Hacks Vitae

Chess Life Lessons, Checked: 19 Claims About the Rules, the History and the Famous Quotes

Hacks Vitae — Wed, 23 Sep 2026 00:00:00 GMT

"Things chess taught me about life" posts are popular, and many of the lessons hold up as life advice. This page does not rule on the advice. It checks the chess inside each one against the rules, the history and the original printings of the famous quotes. The rules quoted are FIDE's Laws of Chess in force from 1 January 2023. English after French, Italian or Latin is our translation.

Part 1: The lessons, checked against the rules

1. "Never reveal your entire strategy"

PARTLY TRUE for chess. Chess is a game of perfect information: the Stanford Encyclopedia of Philosophy gives "a board-game of sequential moves in which both players watch all the action" as its example, and says chess "is an instance of such a game". Nothing on the board is hidden. What is hidden is the plan in each player's head. The chess version of the lesson is: your position is public, your intentions are not.

2. "Every sacrifice has a price; strategy is knowing which prices are worth paying"

PARTLY TRUE for chess. Chess writers split sacrifices in two. Rudolf Spielmann built The Art of Sacrifice in Chess (English edition 1935) around the difference between "temporary and real sacrifices", calling the first kind "sham": in a sham sacrifice the material is given up only for a limited time, so there is no lasting price; "the real sacrifices are much more difficult to treat". The lesson describes the real kind.

3. "Pawns become queens by surviving the entire board"

PARTLY TRUE. A pawn that reaches the far rank "must exchange that pawn as part of the same move for a new queen, rook, bishop or knight of the same colour", and "the player's choice is not restricted to pieces that have been captured previously." So it can become any of four pieces, not only a queen, and a player can have more than one queen. It may capture along the way, so it does not have to cross the board in a straight line. And in the medieval game a promoted pawn became only the weak counsellor piece (see 6), so "the pawn becomes the most powerful piece" is true only of the modern rules.

4. "The knight is dangerous because it refuses straight lines"

PARTLY TRUE. The knight's move is fixed exactly: it "may move to one of the squares nearest to that on which it stands but not on the same rank, file or diagonal." So it is not unpredictable. What sets it apart is that it cannot be blocked. The long-range pieces can: "the bishop, rook or queen may not move over any intervening pieces", and a pawn's two-square first move needs "both squares" empty. The knight's move has no squares in between to block.

5. "The king is the most important piece, not the strongest"

TRUE. The king moves "to an adjoining square" (plus castling), yet the whole game turns on it: the objective is to place the opponent's king under attack "in such a way that the opponent has no legal move." The king is never actually captured; the game ends when it has no escape.

Part 2: The history

6. "The queen was always the most powerful piece"

MYTH. The queen's long move is a late change. In the older game the counsellor piece (Persian firzan, European fers) "also moved one square at a time, moving diagonally", H. J. R. Murray wrote in A History of Chess (1913). He called the old queen "originally far weaker than Rook or Knight, and only little stronger than the Bishop". The new queen moved "to any square in a diagonal, horizontal, or vertical direction, so long as the way is clear". Murray wrote: "I am not disposed to place the invention of the new moves earlier than 1485." Among the earliest records of it is the printed Spanish work of Lucena, from about 1497. Where and when the change began is disputed. Murray found the earliest records in French and Catalan manuscripts and in Lucena's book, and thought "Italy has probably the best claim". The chess historian Ricardo Calvo argued in 1998 for a circle of poets in Valencia, "around 1470-1490", and called their Catalan poem Scachs d'amor the new game's "first written expression", which could put the change earlier than Murray allowed.

7. "Chess comes from India, around the 6th century"

PARTLY TRUE. The documents start at about AD 600. Murray: "Chess was certainly in existence in the 7th century a.d., and it had already at that time penetrated to Persia". The oldest reference is generally taken to be a Middle Persian romance, ascribed "with some hesitation" to the reign of the Persian king Khusraw II, "590-628 a.d.", and early Persian and Arabic tradition "is unanimous in ascribing the game of chess to India". "The 6th century" is an estimate made back from those records, not a dated source. Claims of "five thousand years ago", Murray wrote, are "mere speculation".

8. "Checkmate means 'the king is dead'"

PARTLY TRUE. The word does come from Persian shāh māt, and in English it dates from the "mid-14c.", according to the Online Etymology Dictionary. But Murray reads māt as "a Persian adjective meaning 'at a loss', 'helpless', 'defeated'". The "dead" reading comes through the Arabic verb meaning "to die"; the dictionary, citing Barnhart, calls it "a misinterpretation". It fits the game better too: the king is never taken, only left with no escape.

Part 3: The famous quotes

9. "The pawns are the soul of chess" (Philidor)

TRUE. It is in the preface to Philidor's L'Analyze des échecs, printed in London in 1749: "c'est celle de bien jouer les Pions: Ils sont l'ame des Echècs" (spelling as printed, with the long s modernised). Our translation: "It is how to play the pawns well: they are the soul of chess." He goes on: the pawns alone form the attack and the defence. The English translation of 1750 renders the phrase differently, as "the very Life of this Game".

10. "Never interrupt your enemy when he is making a mistake" (Napoleon)

PARTLY TRUE, and it did not start as a chess saying. The earliest version we found is in Archibald Alison's History of Europe (volume 5, 1836), in an account of the Battle of Austerlitz in 1805. With his marshals pressing to attack at once, Napoleon tells Soult: "let us wait twenty minutes; when the enemy is making a false movement we must take good care not to interrupt him." That is a report written three decades after the battle, and a different wording. The closest nineteenth-century wording Quote Investigator found is from 1888, attributed to Napoleon: "Never interfere with your enemy when he is making a mistake." See also our Napoleon facts.

11. "When you see a good move, look for a better one" (Lasker)

MYTH that Emanuel Lasker coined it. The advice is in Damiano's chess book, printed in Rome in 1512; Murray summarises it as "when you have a good move look for a better". It was still a standard rule in an Italian chess book printed in Modena in 1769, Il giuoco incomparabile degli scacchi, by an anonymous "Modenese author" usually identified as Ponziani. He lists it among "the rules which the authors call prudential": "avendo un buon tratto, cercarne un migliore", having a good move, look for a better one. The next rule in his list: do not risk a small, certain advantage for a larger, uncertain one. We did not find the line in Lasker's Manual of Chess.

12. "Tactics flow from a superior position" (Fischer)

PARTLY TRUE. Bobby Fischer's real words, in a note to his game against Sherwin at the 1957 New Jersey Open in My 60 Memorable Games (1969), are: "As usual, tactics flow from a positionally superior game." It is a comment on one game, not a manifesto.

13. "Chess is life" (Fischer)

UNVERIFIED as to where and when. By December 1972 a periodical wrote that "Robert Fischer was widely quoted as saying" it, and Time printed it as a caption that month. We could not find where or when he said it. The phrase itself was in print by 1936, used by someone else.

14. "Every chess master was once a beginner" (Chernev)

TRUE. It is a chapter epigraph signed by Irving Chernev in Winning Chess (1948), a teaching book he wrote with Fred Reinfeld: "Every chess master was once a beginner".

15. "Help your pieces so they can help you" (Morphy)

UNVERIFIED. The printings we found are modern books, from the twentieth century on, saying only that Morphy "once said" it. We found no source from his lifetime. It is a fair description of how his games are usually summarised, not a record of his words.

16. "The blunders are all there on the board, waiting to be made" (Tartakower)

UNVERIFIED. It is credited to Tartakower in Irving Chernev's anthology The Chess Companion (our copy is dated 1968), with no source given. We did not search his own books.

17. "Chess is a sea in which a gnat may drink and an elephant may bathe" (an Indian proverb)

UNVERIFIED as Indian. Chernev's 1968 anthology labels it an "Indian Proverb"; The Encyclopaedia of Chess (our copy, 1970) calls it a "Russian Proverb". The image is far older than chess books: Gregory the Great described the word of God as a river "shallow and deep, in which a lamb may walk and an elephant swim" (our translation of planus et altus, in quo et agnus ambulet et elephas natet). He meant the Bible, not chess, and a shared image does not prove one borrowed from the other.

18. "Not all artists are chess players, but all chess players are artists" (Duchamp)

PARTLY TRUE. The idea is Marcel Duchamp's, in 1952; the popular wording reorders his: "while all artists are not chess players, all chess players are artists." We have not opened the original transcript, so we give the year and not the occasion.

Part 4: "Chess makes you smarter"

19. "Chess raises IQ"

MYTH as stated. Chess players do score higher: one analysis found "chess players' overall cognitive ability was superior to that of non-chess players by half a standard deviation". But the direction runs the other way, as far as the evidence shows: "smarter individuals are more likely to engage and excel in these fields". For chess lessons in schools, "the effect sizes are inversely related to the quality of the experimental design": among the chess studies reviewed, "only one study with an active control group" existed, "with minimal effects". A two-experiment test against other games found "no statistically significant difference in mathematical problem-solving". A large English school trial found that the groups "did not differ in any of the outcome measures", though Sala, Foley and Gobet doubt that null result because of the trial's design. The fair summary is that the better the study, the smaller the benefit. We found no newer review that changes this, as of September 2026.

Our reading

Most of these lessons use chess as a frame for advice that stands on its own. The chess itself is usually more specific: a pawn chooses among four pieces, the queen's long move is about five centuries old, far younger than the game, and the proverbs attached to famous players mostly reached print in anthologies long after them.

Sources

FIDE, Laws of Chess taking effect from 1 January 2023 (FIDE Handbook E.I.01A).
H. J. R. Murray, A History of Chess (Oxford: Clarendon Press, 1913), including his summary of Damiano's book (Rome, 1512).
Ricardo Calvo, "Valencia Spain: The Cradle of European Chess" (1998).
Online Etymology Dictionary, "checkmate".
A. D. Philidor, L'Analyze des échecs (London, 1749); Chess Analysed (London, 1750).
Archibald Alison, History of Europe, vol. 5 (Edinburgh, 1836); Quote Investigator, "Never Interfere With an Enemy While He's in the Process of Destroying Himself".
Il giuoco incomparabile degli scacchi … Opera d'autore modenese (Modena, 1769); Emanuel Lasker, Lasker's Manual of Chess (1947 edition).
Bobby Fischer, My 60 Memorable Games (1969), game 1.
Irving Chernev and Fred Reinfeld, Winning Chess (1948); Irving Chernev, The Chess Companion (1968 copy); The Encyclopaedia of Chess (1970 copy); Internet Archive full-text search (September 2026).
Gregory the Great, letter to Leander prefacing the Moralia in Iob (Opera, 1594 edition).
Marcel Duchamp, 1952, as printed in A. d'Harnoncourt and K. McShine (eds.), Marcel Duchamp (1973).
Rudolf Spielmann, The Art of Sacrifice in Chess (English edition 1935).
D. Ross, "Game Theory", Stanford Encyclopedia of Philosophy.
G. Sala and F. Gobet, Current Directions in Psychological Science 26 (2017) and Learning & Behavior 45 (2017); G. Sala, J. P. Foley and F. Gobet, Frontiers in Psychology 8 (2017).

Checked September 2026.

How Cybersecurity Is Built: The Layers, the Tools and Who Does What, in Plain Words

Hacks Vitae — Wed, 23 Sep 2026 00:00:00 GMT

Our cybersecurity crash course covers what one person or a small business should do. Our website security crash course covers what a site owner should do. This page explains the bigger picture: how organisations are defended, what the tools are called, and who does what. Each claim is traced to its source, mostly the US National Institute of Standards and Technology (NIST), the US Cybersecurity and Infrastructure Security Agency (CISA), the UK's National Cyber Security Centre (NCSC), MITRE and the Internet Engineering Task Force (IETF). It names no products and gives no legal advice. Versions and figures are as of September 2026.

The idea behind it all: layers

No single barrier is trusted to stop everything. Security people call this defence in depth: "layering heterogeneous security technologies in the common attack vectors to ensure that attacks missed by one technology are caught by another." NIST's other definition makes the point that it is not only technology but a strategy "integrating people, technology, and operations capabilities".

The layers below — network, devices, identity, data, then detection and response, then people — are one common way to picture it, not an official list. The closest official list is CISA's zero trust model, which has five pillars: "Identity, Devices, Networks, Applications and Workloads, and Data."

The frameworks that organise the work

A framework is a checklist of outcomes or controls, not a product. There are four different kinds.

NIST Cybersecurity Framework (CSF) 2.0, dated 26 February 2024, lists outcomes and "does not prescribe how outcomes should be achieved." Its six functions, in NIST's words:
- Govern: "The organization’s cybersecurity risk management strategy, expectations, and policy are established, communicated, and monitored."
- Identify: "The organization’s current cybersecurity risks are understood."
- Protect: "Safeguards to manage the organization’s cybersecurity risks are used."
- Detect: "Possible cybersecurity attacks and compromises are found and analyzed."
- Respond: "Actions regarding a detected cybersecurity incident are taken."
- Recover: "Assets and operations affected by a cybersecurity incident are restored."
CIS Critical Security Controls, version 8.1, published by the non-profit Center for Internet Security: 18 controls and "a total of 153 Safeguards", sorted into three Implementation Groups. The first, IG1, is defined as "essential cyber hygiene," and CIS says "Every enterprise should start with IG1."
ISO/IEC 27001:2022 is a standard for running security as a managed process (an information security management system) that an organisation can be certified against by an auditor. In the 2022 edition "the controls listed in the standard have decreased from 114 to 93", grouped into four areas: "Organizational, People, Physical and Technological."
MITRE ATT&CK is not a checklist of defences but "a knowledge base of cyber adversary behavior". It sorts attacks by tactic, the attacker's goal (MITRE says tactics represent the "why"), and by technique, the method. Its main Enterprise matrix now lists 15 tactics, up from 14: version 19, released in April 2026, made "the split of the Defense Evasion Tactic in Enterprise ATT&CK into the Stealth and Defense Impairment Tactics".
The Cyber Kill Chain, from a Lockheed Martin paper, describes an intrusion as seven ordered steps: "reconnaissance, weaponization, delivery, exploitation, installation, command and control (C2), and actions on objectives." Its defensive point: "just one mitigation breaks the chain". MITRE says the two models are "complementary": "ATT&CK Tactics are unordered and may not all occur in a single intrusion", "whereas the Cyber Kill Chain uses ordered phases".

The network layer

Firewalls. A firewall is "a gateway that limits access between networks in accordance with local security policy." NIST noted long ago that the traditional firewall at the edge of a network has a blind spot: "attacks sent from one internal host to another often do not pass through network firewalls." A next-generation firewall is an industry name rather than a formally defined grade; NIST's glossary gives it no definition, but its network guide describes the feature that sets it apart: "The distinguishing feature of NGFW is application data awareness." In plain words, it can tell which app is talking, not just which address.

Segmentation. Instead of one big flat network, segmentation divides it "into multiple segments, each acting as its own subnetwork", with controlled doors between them. A flat network makes it "easier for threat actors to move laterally" once they are in. Microsegmentation is the same idea at a finer grain, to "prevent attack escalation."

VPNs and zero-trust access. A company VPN puts a remote worker on the company network, like handing them a key to the building. The newer approach grants access to one application at a time after checking who is asking and from what device. CISA and its partners urged "business owners of all sizes to move toward more robust security solutions" in June 2024, noting that CISA had found "over 22 Known Exploited Vulnerabilities (KEVs) related to VPN compromise".

Intrusion detection and prevention. Intrusion detection is "the process of monitoring the events occurring in a computer system or network and analyzing them for signs of possible incidents"; prevention adds "attempting to stop detected possible incidents." Neither is perfect: NIST says they all generate false alarms and misses.

Protective DNS. Every time a device looks up a web address, a protective DNS service checks it against lists of known-bad destinations and "blocks, redirects, or sinkholes the query response" if it matches. CISA runs one for US federal civilian agencies, and offers it to some critical-infrastructure organisations through a limited pilot.

Email: SPF, DKIM and DMARC. Three DNS records let receiving mail servers check whether an email that claims to come from your domain really does. SPF lets you "publish IP addresses which should be trusted for your domain"; DKIM "allows you to cryptographically sign email you send"; DMARC "allows you to set a policy for how receiving email servers should handle email which doesn’t pass either SPF or DKIM checks." The current DMARC standard is RFC 9989, published in May 2026, which replaces RFC 7489. It is frank about its limits: "DMARC can only be used to combat specific forms of exact-domain spoofing directly." It does not stop look-alike domain names or a fake display name.

Denial-of-service protection. A DDoS attack "is a form of DoS attack that originates from more than one source." CISA is candid: "there is no way to completely avoid becoming a target". Protection is mostly upstream — a provider that "detects abnormal traffic flows and redirects traffic away from your network" — plus a plan for when it happens.

The device layer

Antivirus, EDR and XDR. Antivirus is "a program that monitors a computer or network to identify all major types of malware and prevent or contain malware incidents." EDR (endpoint detection and response) has no definition in NIST's glossary, which lists only the abbreviation. In practice it is software on each device that records what the device does so that people can spot and investigate suspicious activity. In January 2025 CISA described its federal rollout as "over 920,000 Endpoint Detection and Response (EDR) agents across 51 agencies, enabling analysts to hunt actively for intrusions". XDR ("extended" detection and response) is a vendor category name for extending the same idea beyond individual devices; NIST's glossary again lists only the abbreviation.

Patching, and the list that tells you what to patch first. CISA's Known Exploited Vulnerabilities catalogue lists flaws already used in real attacks; on 22 September 2026 it held 1,721 entries. CISA says organisations "should use the KEV catalog as an input to their vulnerability management prioritization framework." For US federal agencies the rule changed on 10 June 2026: the old directive, BOD 22-01, "has been revoked. It is superseded by BOD 26-04", which ties deadlines to risk — whether the device is exposed to the internet, whether the flaw is in the catalogue, whether an attacker can automate the exploit, and how much control it gives them. The deadlines in its table run from 3 days to 60 days, and to "fix on system upgrade" for the lowest risk. It binds only US federal civilian agencies. The lesson for everyone else: patch first what is reachable from the internet and already being exploited.

Application allowlisting. Instead of trying to recognise every bad program, only programs "authorized to be present or active on a host according to a well-defined baseline" may run. Australia calls it application control and puts it in its Essential Eight.

Mobile device management. MDM is "the administration of mobile devices such as smartphones, tablets, computers, laptops, and desktop computers" from one place, typically pushing settings and updates and wiping a lost device.

The identity layer

When staff work from anywhere and data lives in cloud services, who is asking matters more than where the request comes from.

Identity and access management is "about establishing and managing the roles and access privileges of individual network users": who works here, what they may touch, and switching access off when they leave.
Single sign-on means "one account and its authenticators are used to access multiple applications". One login becomes very valuable, so it needs the strongest two-step verification.
Two-step verification (MFA) is "authentication using two or more different factors". CISA's warning is that "not all forms of MFA are equally secure", and that "phishing-resistant MFA is the gold standard". The crash course explains the kinds.
Privileged access. Administrator accounts can "perform system control, monitoring, administration functions, or security-relevant functions that ordinary users are not authorized to perform", so they are the prize. Keep them few, separate from everyday accounts and used only when needed. The principle is least privilege: access "to the minimum necessary to accomplish assigned tasks."
Zero trust. NIST's model assumes a network "viewed as compromised" and decides each request on its merits, because "Network location alone does not imply trust." Its seven tenets include "All communication is secured regardless of network location" and "Access to individual enterprise resources is granted on a per-session basis."

The data layer

Encryption is the "cryptographic transformation of data (called “plaintext”) into a form (called “ciphertext”) that conceals the data’s original meaning". "At rest" means stored (disks, databases, backups); "in transit" means moving across a network.
Key management covers "the entire lifecycle of the keys, including their generation, storage, establishment, entry and output, use and destruction." Lose the key and the data is gone; leak it and the encryption is worthless.
Backups. The crash course explains the 3-2-1 rule. The organisational point: "Maintain offline, encrypted backups of critical data", because "many ransomware variants attempt to find and subsequently delete or encrypt accessible backups".
Data loss prevention is a system's ability "to identify, monitor, and protect data in use (e.g. endpoint actions), data in motion (e.g. network actions), and data at rest" — for example, flagging customer records attached to an outgoing email.

Detection and response

Logs and SIEM. Logs are the diary every system keeps. A SIEM (security information and event management) is, in NIST's glossary, "a program that provides centralized logging capabilities for a variety of log types"; NIST's incident-response guide recommends using such tools "to continuously monitor log events for known malicious and suspicious activity". SOAR (security orchestration, automation and response) is an industry category name for tools that automate and coordinate response work; NIST's guide lists it beside SIEM for monitoring and correlating log events.

The security operations centre. A SOC is the team that watches the alerts and starts the response. NIST lists several ways to staff incident handling, including "outsourcing a security operations center [SOC] to a managed security services provider [MSSP]".

Threat intelligence is "threat information that has been aggregated, transformed, analyzed, interpreted, or enriched to provide the necessary context for decision-making processes."

Incident response. NIST's guide was revised in April 2025 (Rev. 3). The familiar four phases — preparation; detection and analysis; containment, eradication and recovery; post-incident activity — are now its previous model. The new one says "All six NIST Cybersecurity Framework (CSF) 2.0 Functions play vital roles in incident response", and that organisations "should use the incident response life cycle framework or model that suits them best." It is blunt about why: "another incident is inevitable."

Practising. A tabletop exercise is a meeting where a team talks through a made-up incident to find the gaps. Free packs exist: CISA offers "over 100" exercise packages, covering physical as well as cyber scenarios, and the NCSC's Exercise in a Box "is used by small and medium sized enterprises".

Testing your own defences, from cheapest to most realistic:

A vulnerability scan is "a technique used to identify hosts/host attributes and associated vulnerabilities." It is broad and cheap, and "can have a high false positive error rate".
A penetration test is testing "in which assessors mimic real-world attacks to identify methods for circumventing the security features" of a system, within agreed limits.
A red team is "a group of people authorized and organized to emulate a potential adversary's attack or exploitation capabilities against an enterprise's security posture."

Letting outsiders report holes. A vulnerability disclosure policy says what finders may do "in order to find and report vulnerabilities in a legally authorized manner." It is "similar to, but distinct from, a “bug bounty.”", in which "organizations pay for valid and impactful findings". Whether a policy protects a finder legally depends on the country; this is not legal advice. For websites, a simple first step is a security.txt file saying who to contact, explained in the website crash course.

Who does what

The US NICE Framework describes cybersecurity work as work roles and says outright that "Work Roles are not synonymous to job titles or occupations." The job titles below are ours, matched to the roles NICE describes:

Common job title	What the work is (NICE's wording)
Head of security (CISO)	"Executive Cybersecurity Leadership": "establishing vision and direction for an organization's cybersecurity operations and resources"
SOC analyst	"Defensive Cybersecurity": "analyzing data collected from various cybersecurity defense tools to mitigate risks."
Incident responder	"Incident Response": "investigating, analyzing, and responding to network cybersecurity incidents."
Governance, risk and compliance	An industry label; the nearest roles include "Security Control Assessment" and "Technology Program Auditing"

Security is not only the IT team's job. NIST's guide gives leadership its own role: it "oversees incident response, allocates funding, and may have decision-making authority on high-impact response actions". The NCSC's version for small organisations: "Every member of the team should realise that cyber security is everyone’s business."

Awareness training is where the evidence is disputed. Government baselines recommend it, while the strongest test we found, a 2025 randomised study, found common training formats "offer limited value". The crash course sets out both sides. The practical answer is the same either way: put first the controls that do not depend on someone noticing, such as phishing-resistant two-step verification and calling back before any payment change.

What a small organisation needs

Almost everything above describes large organisations. None of the small-organisation baselines we checked (the UK NCSC's small organisations guide and the UK and US Cyber Essentials) starts with a security operations centre, a SIEM or a red team. NIST itself says the right incident-response model "depends on many factors", and that "larger and more technology-dependent organizations are likely to benefit more" from one "emphasizing continuous improvement". On who handles incidents, it notes that many organisations may use more than one approach, "such as internally performing basic incident response and engaging third-party resources for assistance with certain incidents." A small firm's security operations centre can be a contract with a provider, or a named person plus the phone numbers of whoever helps in an emergency.

The crash course lists the UK's Cyber Essentials controls and Australia's Essential Eight. Mapped onto the layers above, in our reading:

Layer	First step for a small organisation
Network	Firewall on: the first of the UK Cyber Essentials controls.
Devices	Updates on; patch internet-facing, actively exploited flaws first.
Identity	Two-step verification on every account, phishing-resistant for admin and email; separate admin accounts.
Data	Offline, tested backups.
Detection and response	A written who-to-call plan, one tabletop exercise a year, and rented monitoring if needed.
People	Everyone owns it; controls come before training.

The UK NCSC's guide for small organisations says its steps are ones "many of which can be completed in as little as 5 minutes". And CIS's advice for any organisation starting out: "Every enterprise should start with IG1."

Myths, checked

MYTH: "A firewall is enough." NIST's firewall guide already noted that "attacks sent from one internal host to another often do not pass through network firewalls." Layers exist because any one of them can fail.

MYTH: "Zero trust is a product you can buy." NIST: "ZT is not a single architecture but a set of guiding principles for workflow, system design and operations", and it "cannot simply be accomplished with a wholesale replacement of technology." CISA adds that it "may require a change in an organization’s philosophy and culture around cybersecurity." Products can implement parts of it.

MYTH: "Compliant means secure." The NCSC in so many words: "Compliance and security are not the same thing. They may overlap", and compliance "can coexist with, and mask, very weak security practices." A certificate is evidence of a baseline, not proof of safety. Where a law requires compliance, you still need it; this is not legal advice.

MYTH: "With good detection, you don't need backups." Detection tools miss things: NIST says intrusion detection systems "all generate false positives" and "false negatives (failing to identify malicious activity)." And ransomware goes after backups directly; the NCSC notes that ransomware incidents "have also compromised connected cloud storage locations containing backups."

PARTLY TRUE: "Air-gapped means safe." Cutting a computer off from the network removes one route in, not every route. The US Congressional Research Service described a worm that spread "through the use of thumb drives in computers that were not connected to the Internet", and noted that it "can be spread through an air-gapped network by a removable device, such as a thumb drive".

What changed recently

Topic	Earlier figure	As of September 2026
MITRE ATT&CK Enterprise tactics	14	15, since version 19 (April 2026)
DMARC	RFC 7489	RFC 9989 (May 2026), which replaces it
NIST incident response	The four-phase lifecycle	Rev. 3 (April 2025), organised around the CSF 2.0 functions
CISA patch deadlines (US federal)	Two weeks, or six months for older flaws	BOD 26-04 (June 2026): by risk, from 3 days to 60 days or the next system upgrade
ISO/IEC 27001	114 controls (2013 edition)	93 controls (2022 edition)

Sources

NIST: CSRC glossary (defence in depth, firewall, antivirus software, application allowlisting, mobile device management, identity and access management, single sign-on, multi-factor authentication, privileged user, least privilege, encryption, key management, data loss prevention, SIEM, threat intelligence, vulnerability scanning, red team); CSWP 29, The NIST Cybersecurity Framework 2.0 (February 2024); SP 800-41 Rev. 1 (2009); SP 800-94 (2007); SP 800-115 (2008); SP 800-207, Zero Trust Architecture (2020); SP 800-215 (2022); SP 800-61 Rev. 3 (April 2025); NICE Framework via CISA NICCS.
CISA: Known Exploited Vulnerabilities catalogue (version of 22 September 2026); BOD 22-01 and BOD 26-04 (June 2026); Zero Trust Maturity Model Version 2.0 (April 2023); "Implementing Phishing-Resistant MFA" (October 2022); "Layering Network Security Through Segmentation"; "Modern Approaches to Network Access Security" (June 2024); Protective DNS Resolver; "Understanding Denial-of-Service Attacks"; #StopRansomware Guide; BOD 20-01; Tabletop Exercise Packages; "Securing Federal Networks" (January 2025).
UK NCSC: Small organisations guide to cyber security (2026); email security and anti-spoofing; denial of service guidance; "Offline backups in an online world"; Exercise in a Box; risk management guidance.
IETF: RFC 7208 (SPF), RFC 6376 (DKIM), RFC 9989 (DMARC, May 2026).
MITRE ATT&CK: FAQ, Enterprise tactics, April 2026 update. Lockheed Martin, "Intelligence-Driven Computer Network Defense Informed by Analysis of Adversary Campaigns and Intrusion Kill Chains".
Center for Internet Security: CIS Controls v8.1 and Implementation Groups. BSI, "Transition to the ISO/IEC 27001:2022 standard".
US Congressional Research Service, R41524, "The Stuxnet Computer Worm" (December 2010).
Ho et al., "Understanding the Efficacy of Phishing Training in Practice", IEEE Symposium on Security and Privacy 2025.

Checked September 2026.

Website Security Crash Course: The Terms, the Common Mistakes and a Checklist, in Plain Words

Hacks Vitae — Wed, 23 Sep 2026 00:00:00 GMT

This is the website companion to our cybersecurity crash course, which covers passwords, two-step verification, backups and scams for people in general. This page adds the layer that belongs to whoever runs a website: a blog, a small shop, a portfolio, a WordPress site or a static one. It explains what each risk is and how to prevent it, in plain words, with each claim traced to its source: OWASP (the Open Worldwide Application Security Project), MDN (the web's reference documentation), NIST, CISA, the certificate industry's own rules, and the platforms' own documentation. It recommends nothing you have to buy and gives no legal advice. Figures and rules are as of September 2026.

Do these first

None of the official guides we checked (the UK NCSC's small organisations guide, CISA's Secure Our World and OWASP's verification standard) puts these steps in order for a small website, so the order below is ours. It follows the evidence: in the breaches Verizon studied for its 2026 report, exploiting unpatched software was the most common way in, ahead of phishing and of stolen or misused logins.

Put two-step verification on four accounts: your domain registrar, your host, your site's admin login and your code repository or deploy account. The registrar and the host come first, because they control everything else.
Turn updates on, and delete plugins and themes you do not use — delete, not just deactivate.
Back up both the files and the database, keep at least one copy away from the host, keep several dates, and test a restore.
Serve everything over HTTPS, renewing automatically, with an expiry alert from something other than the company that issues the certificate.
Give each person their own account at the lowest role that works, and remove old accounts.
Keep the domain safe: auto-renew with a working card, transfer lock on, a contact email you read, and no DNS records left pointing at services you stopped using.
Add the security headers described below.
Know when something changes: an uptime alert, a file-change alert, and a look at the admin-user list now and then. Verify the site in Google Search Console so Google can tell you if it finds a problem.
Keep nothing secret in the public folder or a public repository, and replace any key that ever was.
Give people a way to report a problem, with a security.txt file (see below).

The words you will meet

HTTPS, TLS and certificates

HTTPS "is an encrypted version of the HTTP protocol". "It uses TLS to encrypt all communication between a client and a server." TLS is the encryption layer, "formerly known as Secure Sockets Layer (SSL)". A certificate is the file that proves to the browser that the server really is the domain it claims to be; TLS requires "the server to provide a valid digital certificate confirming its identity".

Certificates are getting shorter-lived, which makes automatic renewal essential. The certificate industry's rules (the CA/Browser Forum) set the maximum lifetime at 200 days from 15 March 2026, 100 days from 15 March 2027 and 47 days from 15 March 2029. Let's Encrypt, the free certificate authority, already issues shorter certificates. In December 2025 it said: "We currently issue certificates valid for 90 days, which will be cut in half to 45 days by 2028." The 47 and the 45 are not a conflict: one is the industry ceiling, the other is one authority's own choice. Let's Encrypt also stopped emailing expiry warnings on 4 June 2025, and now recommends monitoring "to alert appropriately if certificates aren’t renewed when expected." Many hosts renew for you; check that yours does.

DNS, the registrar and registrar lock

DNS is the internet's address book: its main job "is the translation of human-friendly domain names (such as mozilla.org) to a numeric IP address". The registrar is the company you rent the domain from. Whoever gets into that account can move the domain away from you. A transfer lock (clientTransferProhibited) "tells your domain's registry to reject requests to transfer the domain from your current registrar to another". Some registries also offer a stronger registry lock; ask your registrar.

CMS, plugins, themes and the admin panel

A CMS (content management system — WordPress, Joomla, Drupal and the hosted builders) is the software that lets you edit a site without writing code. Plugins add features and themes change the look; both are other people's code running with your site's permissions. The admin panel is where you log in to change everything, which is why attackers go after it. Your host usually is not responsible for the software you install on top. WordPress's own handbook: "Web hosts are often responsible for the infrastructure on which your website sits, they are not responsible for the application you choose to install."

Dependencies and the supply chain

Your site is built on other people's code: the CMS, its plugins, JavaScript libraries, build tools, the script a chat widget loads. A supply-chain attack is "an incident where an adversary exploits vulnerabilities in the product or service supply network of the intended target", so the harm arrives through a trusted update.

Secrets and API keys

A secret is anything that lets a program act as you: a database password, an API key, a payment-provider key. It belongs in the server's settings, never in code sent to the browser and never in a public repository. Deleting it later does not undo a leak: "Secrets persist in Git history even after removal from current code." The fix is to revoke it and issue a new one; GitHub's advice is to "rotate the affected credential immediately".

Backups, site by site

For backups in general, see the crash course. The website point: "There are two parts to backing up your WordPress site: Database and Files." "You need both to be able to fully restore a typical WordPress site." Keep several dates, because a hack is often noticed late — the handbook's example is a site "compromised on May 1st but the compromise is not detected until May 12th".

Firewalls, CDNs, floods and rate limiting

A web application firewall (WAF) sits in front of a site and "applies a set of rules to an HTTP conversation" to block common attacks. It is a filter, not a fix (see the myths).
A CDN "is a group of servers spread out over many locations", so visitors load your files from nearby. Many also cope better with heavy traffic.
A DDoS attack is a flood of traffic from many machines meant to knock a site offline. CISA and its partners describe three kinds: "Volumetric, attacks aiming to consume available bandwidth", protocol attacks and application attacks.
Rate limiting is the server telling any one visitor to slow down. MDN calls asking a visitor to slow down "rate limiting", and its error code is 429.

Security headers

Security headers are short instructions your server sends with every page, telling the browser what it may do with it. On most hosts they cost nothing to add.

Header	What it does
`Strict-Transport-Security` (HSTS)	Tells browsers the site "should only be accessed using HTTPS", and stops visitors clicking past certificate errors.
`Content-Security-Policy` (CSP)	Lists where scripts may load from; "mainly used as a defense against cross-site scripting (XSS) attacks".
`frame-ancestors` (in CSP) or `X-Frame-Options`	Stops other sites showing your page inside a frame, to "avoid clickjacking attacks".
`X-Content-Type-Options: nosniff`	Tells the browser not to guess a file's type.
`Referrer-Policy`	"controls how much referrer information" is passed on when a visitor follows a link.

One header you should not add is X-XSS-Protection, still found in old checklists. OWASP's guidance now is: "Do not set this header or explicitly turn it off."

HSTS preload is a genuine disagreement. OWASP's recommended header includes preload, which signals that you consent to your domain being built into browsers as HTTPS-only; the domain is added only once it is submitted to the list. The people who run that list ask anyone who gives HTTPS configuration advice: "do not include the preload directive by default", because "inclusion in the preload list cannot easily be undone" and removal "takes months". Both agree HSTS itself is good. Our reading: turn HSTS on, raise its duration in stages, and treat preloading as a separate decision for later, once every subdomain works over HTTPS.

Cookie flags

A cookie is how a site remembers that someone is logged in. Secure means it is only sent over HTTPS; HttpOnly "forbids JavaScript from accessing the cookie"; SameSite limits cross-site request forgery. MDN's warning: "Do not assume that Secure prevents all access to sensitive information in cookies". On a CMS you rarely set these by hand; they matter when you choose software or build your own login.

Two-step logins and least privilege

For the kinds of two-step verification and why some resist phishing better, see the crash course. Every account that can publish, install plugins or change settings is a door into the site and needs its own login with two-step verification. Least privilege means restricting access "to the minimum necessary to accomplish assigned tasks": an author does not need to be an administrator.

Logs and alerts

A log is the site's diary of who asked for what and when. "Without logging and monitoring, attacks and breaches cannot be detected", and OWASP adds that "great logging with no alerting is of minimal value". Something has to tell you.

security.txt

If a stranger finds a hole in your site, how do they tell you? security.txt is a small file at /.well-known/security.txt saying who to contact. Its format is described in RFC 9116 (April 2022), which is informational rather than a formal standard. Two fields are required: Contact and Expires. The RFC's own warning is that having no file at all "may be preferable to having stale information in this file."

The OWASP Top 10, in plain words

The OWASP Top 10 is "a standard awareness document for developers and web application security", not a rulebook you comply with. The current edition is the 2025 list; OWASP says "The most current released version is the OWASP Top 10 2025". It is written for developers. In our reading, a blog or small-shop owner mostly meets numbers 2, 3, 7 and 9 — settings, plugins and dependencies, logins, and noticing trouble.

#	Name	In plain words
A01	Broken Access Control	People can see or do what they should not: another customer's order, an admin page.
A02	Security Misconfiguration	The software is fine, the settings are not: defaults left on, debug pages exposed.
A03	Software Supply Chain Failures	Something you depend on is out of date or has been tampered with.
A04	Cryptographic Failures	Data that should be encrypted is not, or the keys leak.
A05	Injection	Text typed into a form is treated as an instruction instead of data.
A06	Insecure Design	The plan was flawed before any code was written.
A07	Authentication Failures	The login can be fooled or worn down by guessing.
A08	Software or Data Integrity Failures	The site trusts code or data it has not checked.
A09	Security Logging and Alerting Failures	Nobody would notice an attack.
A10	Mishandling of Exceptional Conditions	The site behaves badly when something unexpected happens.

What changed from 2021: "There are two new categories and one consolidation in the Top Ten for 2025." The two new ones are supply chain failures, which OWASP calls "an expansion of A06:2021-Vulnerable and Outdated Components" (a new category for an old risk, widened), and mishandling of exceptional conditions. Server-side request forgery "has been rolled into" broken access control. And security misconfiguration "moved up from #5 in 2021 to #2 in 2025".

The mistakes that let attackers in

Outdated and unused plugins and themes

On a CMS site, this is the main door. The figures come from one security company, Patchstack, which sells protection against exactly this, so read them as one company's data rather than a census. It counted 11,334 new vulnerabilities in the WordPress ecosystem in 2025, "a 42% increase compared to 2024"; "91% of new vulnerabilities were found in plugins, and 9% were found in themes"; and only 6 were found in WordPress itself, all low priority. For the most-exploited ones, the weighted median time from disclosure to first attack was 5 hours, and "46% of vulnerabilities did not receive a fix from the developer in time for public disclosure." Google's own guidance agrees on the principle: "outdated or unpatched themes and plugins are a major source of vulnerabilities." It also warns against pirated copies: "It's a common tactic for attackers to add malicious code to free versions of paid plugins or themes."

Updates are necessary but not always enough on their own, because a fix may not exist yet. That is why the second defence is having fewer plugins: "if you are not using a specific plugin, delete it from the system."

Default or weak admin logins

"Default accounts and their passwords are still enabled and unchanged" is one of OWASP's own examples of misconfiguration. Guessing is automated: "A brute force attack is the simplest way to break in: an attacker repeatedly tries username/password combinations until one works." Two-step verification on every admin account is the answer; for passwords themselves, see the crash course.

Files left where anyone can download them

Anything inside the public folder of a site can be downloaded by anyone who guesses its name, and the guessing is done by scripts. Four common leaks: a .env or settings file holding passwords; a .git folder, which holds the project's full history ("If these subdirectories are stored on a web server or added to an archive, then these could be used by an attacker"); a backup archive left next to the live files ("A backup file is stored in a directory or archive that is made accessible to unauthorized actors"); and directory listing switched on, which "exposes a directory listing with an index of all the resources located inside of the directory." Keep secrets and backups outside the public folder, and switch directory listing off.

Cloud storage set to public

Amazon's storage service, for one, now makes new storage private by default: "By default, new buckets, access points, and objects don't allow public access." The leak today is usually a setting someone changed to make something work: "users can modify bucket policies, access point policies, or object permissions to allow public access."

Mixed content and expired certificates

Mixed content "refers to securely loaded web pages that use resources to be fetched via HTTP", usually old http:// links to images or scripts left in posts after a move to HTTPS. Browsers now auto-upgrade images, video and audio and "block insecure requests for all other resource types", so the symptom is usually something broken rather than a warning. Expired certificates are becoming more likely as lifetimes shrink; the visitor sees a full-page browser warning (below).

Roles that are too broad

WordPress's roles run from Subscriber to Administrator, "somebody who has access to all the administration features". Most people who write for a site need Author or Editor. Google lists "granting administrative access to users who don't require it" among the ways sites get hacked.

Unpatched server software

If you run your own server rather than managed or static hosting, the web server, language and database need updates too. CISA keeps a catalogue of flaws already being used in real attacks, and "strongly recommends all organizations review and monitor the KEV catalog". On 22 September 2026 it listed 1,721 of them. On managed or static hosting, the host patches the server; your part is the CMS, the plugins and the accounts.

Losing the domain

Three separate mistakes. First, a registrar account with no two-step verification and no transfer lock. Second, letting the domain expire: for most generic endings, "ICANN policy requires registrars to send you two renewal reminders approximately one month and one week before expiration", which only help if the contact email works; after the grace periods the name is "released and made available for registration on a first-come-first-served basis". Third, a dangling DNS record: a subdomain still pointing at a service you stopped using, which someone else can then claim so that your subdomain shows their content. OWASP's summary: an attacker can "claim and take control of the victim's subdomain." Delete DNS records when you stop using what they point at.

What the warnings and error codes mean

"Your connection is not private"

The browser checked the certificate and something did not add up. Chrome's own description: "there's a problem with the site, the network, or your device." Three common codes, with the meaning Chrome's source code gives:

NET::ERR_CERT_DATE_INVALID — the certificate has expired or is not yet valid. For an owner, the likely cause is a failed renewal, though a visitor's wrong clock causes it too: "3. Our clock is wrong."
NET::ERR_CERT_COMMON_NAME_INVALID — the certificate is for a different name, for example it covers www. but not the bare domain: "The server is misconfigured and responding with the wrong cert."
NET::ERR_CERT_AUTHORITY_INVALID — signed by an authority the browser does not trust, or self-signed.

Do not tell visitors to click through. With HSTS on, the browser will not let them, by design.

"Not secure" and the red warning page

"Not secure" beside the address means the page loaded over plain HTTP: "Someone may be able to view and change the information you send and get through this site." It is not an accusation; "the site owner must secure the site and your data with HTTPS." A full-page red warning is different: "If you get a full-page red warning screen, the site is flagged as unsafe by Google Safe Browsing". For an honest owner, a hack is one of the likely causes; Search Console's Security issues report says which problem Google found.

HTTP error codes

Code	Name	In plain words
401	Unauthorized	You need to log in first.
403	Forbidden	The server "understood the request but refused to process it" — often a security rule or a file permission.
404	Not Found	Nothing at that address.
429	Too Many Requests	Slow down: rate limiting.
500	Internal Server Error	A catch-all for problems on the server; only the owner can fix it.
502	Bad Gateway	A middle server, such as a CDN, got a bad answer from the real one behind it.
503	Service Unavailable	"down for maintenance or overloaded" — meant to be temporary.

A detailed error page that prints file paths or software versions is itself a leak; OWASP lists "a lack of central configuration for intercepting excessive error messages" as a misconfiguration.

Google Search Console's "Security issues"

If Google decides a site is hacked or harmful, its pages "can appear with a warning label in search results or an interstitial warning page in the browser". The details are in Search Console under Security issues. Once every affected page is fixed, you ask for a review. "Fixing the issue on just some pages will not earn you a partial return to search results", and "a review can take from a few days to a few weeks to complete."

Myths, checked

MYTH: "My site is too small to be attacked." Many website attacks are automated: "Many WordPress attacks are carried out autonomously by malicious software bots." Nobody has to choose your site; being reachable is enough. The UK NCSC puts it plainly: "if you think your business is too small to be a target, think again."

MYTH: "HTTPS means my site is safe." HTTPS protects data on its way between the visitor and the server. It says nothing about whether the server, the CMS or the plugins are secure, and "hacks are often invisible to users, yet remain harmful to anyone viewing the page". This agrees with the padlock myth in the crash course, seen from the owner's side.

MYTH: "A security plugin is enough." It is a useful layer. It does not update your other plugins, secure your registrar or hosting login, or make a backup you have tested, and it is itself a plugin. WordPress's own handbook: "What security is though is risk reduction, not risk elimination."

PARTLY TRUE: "Static sites can't be hacked." A site made of plain files has no login page, no database and no plugin code on the live server, so the most common way CMS sites are hacked is simply absent. This site is one. But the risk moves rather than vanishes. The browser runs code too. A page's own scripts can still be open to injection, which OWASP defines as untrusted input "sent to an interpreter (e.g. a browser, database, the command line)", and a page that loads third-party scripts inherits their risk: OWASP calls "a compromise of the third party JavaScript server" "the single greatest risk" of such scripts. The hosting account, the code repository and the registrar all have logins, and whoever takes one over can replace the whole site. And the build process pulls in dependencies. No standards body states this verdict in one sentence; it is our reading of those parts.

MYTH: "Hide the admin URL and you're safe." Moving the login page is obscurity, not a lock. NIST's principle: "System security should not depend on the secrecy of the implementation or its components." WordPress's handbook agrees that "Security through obscurity is generally an unsound primary strategy", while allowing it a small role, such as not calling your admin account "admin". It is no substitute for two-step verification and updates.

MYTH: "A firewall replaces patching." A WAF rule can block a known attack while you wait for a fix, and OWASP values that time: "50% reduction in 10 minutes is better than 100% reduction in 48 hrs." But it is explicit that "the number one remediation strategy" is fixing the software itself, and that "Code level fixes and Virtual Patching are NOT mutually exclusive".

If your site has been hacked

Google's step-by-step guidance for site owners is the most complete public guide we found; the steps below are drawn from it.

Confirm it. Check Search Console's Security issues report and the example pages it lists. Hacked content is often hidden from the owner: "Cloaking makes cleaning a site more difficult by showing different content to different types of users."
Tell your host. "Your hoster can make sure their other customers weren't affected, and they can potentially help recover your site."
Take the site offline properly, with the offline notice served "from outside your compromised server or site". Blocking search engines is not enough: "Using a robots.txt disallow is also insufficient because it only blocks search engine crawlers."
Lock the accounts. Remove any accounts the attacker created, and "change the passwords for all site users and accounts" — hosting, file transfer, database and CMS — and revoke and reissue any API keys and tokens the site uses, as with any leaked secret.
Find how they got in, and keep looking after the first hole. Check the admin's own computer too: "On an administrator's virus-infected computer, the hacker might have installed spyware".
Think before deleting anything if visitors' personal data may have been taken: "consider any business, regulatory, or legal responsibilities before you begin cleaning your site or deleting any files." Rules vary by country; this is not legal advice.
Clean up. "First, check that your backup was created before your site was hacked." Otherwise do a clean installation — "Upgrades can leave files from a previous version" — and bring back only known-clean content. Update everything and remove what you do not use.
Ask Google for a review once every page is clean, and describe what you fixed. Asking too early "will only prolong the period of time that your site is flagged".
Report it and tell the people who need to know. Where to report depends on where you are; the crash course has the list.

Sources

OWASP: Top 10:2025 (introduction and the ten category pages) and the project page; Top 10:2021; cheat sheets on HTTP security headers, virtual patching and third-party JavaScript; Web Application Firewall community page; Web Security Testing Guide (subdomain takeover).
MDN Web Docs: HTTPS, TLS, DNS, CDN, mixed content, Set-Cookie, Strict-Transport-Security, Content-Security-Policy, X-Frame-Options, Referrer-Policy, and HTTP status codes 401 to 503.
CA/Browser Forum, TLS Baseline Requirements and Ballot SC081v3 (April 2025); Let's Encrypt, "Expiration Notification Service Has Ended" (June 2025) and "Decreasing Certificate Lifetimes to 45 Days" (December 2025); hstspreload.org.
NIST CSRC glossary (supply chain attack, least privilege); NIST SP 800-123, Guide to General Server Security (2008).
CISA: Known Exploited Vulnerabilities catalogue (version of 22 September 2026); "Understanding and Responding to Distributed Denial-of-Service Attacks" (March 2024).
MITRE CWE-527, CWE-530 and CWE-548; ICANN, EPP status codes and domain renewal FAQs; IETF RFC 9116 (April 2022).
UK NCSC, Small organisations guide to cyber security (2026).
WordPress Advanced Administration Handbook (hardening, backups, brute force) and Roles and Capabilities; GitHub Docs (secret leakage and secret scanning); Amazon S3 documentation (block public access); Google Chrome Help; Chromium source (net_error_list.h); Google Search Console Help (security issues report); Google web.dev hacked-site guides.
Verizon, 2026 Data Breach Investigations Report, executive summary.
Patchstack, "State of WordPress Security in 2026" (February 2026) — vendor research.

Checked September 2026.

AI Design Tools and Component Libraries, Checked: 7 Tools, Their Licences and Their Traps

Hacks Vitae — Tue, 22 Sep 2026 00:00:00 GMT

We checked each of these seven tools against its own website, licence file and code repository in September 2026. Prices change often, so treat them as a snapshot. For the wider picture, see our web design resources guide.

1. Cult UI

What it is. A free library of components you copy into your own project, in the shadcn/ui style: "Accessible and customizable components that you can copy and paste into your apps. Free. Open Source." It is published by Nolly Studio; its licence names Jordan Gilliam.

Licence and price. The free code is under the MIT licence ("MIT License Copyright (c) 2023 Jordan-Gilliam"). A separate product, Cult UI Pro, sells premium blocks for a one-time "lifetime" payment: $129 on a sale, against $179.

The catch. Pro is not MIT. Its terms give "a non‑exclusive, non‑transferable license", and "the license is non‑transferable and tied to the purchaser." They allow Pro blocks in "personal and commercial projects", but say "You may not publish our code publicly (e.g. open‑source repositories)", even though the free Cult UI code next to it is MIT.

2. Componentry

What it is. "Animated React components with editable source code. Install via shadcn CLI or your AI editor with MCP." Its home page lists 52 components, built by one developer, Harsh Jadhav.

The catch. It is young (the repository was created in December 2025) and has one maintainer. Its terms say the code is "generally provided under the MIT License unless otherwise specified", so check each component you use.

3. Refero and Refero Styles

What it is. Refero is a paid library of screenshots from real apps and websites. Refero Styles, marked Beta, lets you "browse 2,000+ AI-readable design systems from leading product websites": colour, type and spacing notes, as DESIGN.md files, for pasting into an AI coding assistant.

Licence and price. You can "browse and copy any DESIGN.md for free". Refero Pro costs $17 a month, $120 a year, or $250 as a "one-time payment"; the free plan gives "access to only ~3% of the full collection".

The catch. Each style is named after a real company. Refero's own guide says: "You should not ask the agent to recreate ElevenLabs, Linear, Stripe, or any other source exactly." Refero's pages we read say nothing about licensing the fonts or the brand a style describes.

4. Taste Skill

What it is. "Open-source skill files that stop Cursor, Claude Code, Codex, Gemini CLI, v0, Lovable, and more from generating generic frontends." "Taste" in videos most likely means this project (tasteskill.dev), the one by far the most starred on GitHub under that name, but the name alone does not prove it.

The catch. Its README and its website both carry sponsor offers with tracking links. The README also warns that "Taste Skill has no official token, coin, or crypto project", so anything selling one is not the project.

5. Impeccable

What it is. "Design guidance for AI coding agents. 1 skill, 24 commands, live browser iteration, and 61 deterministic detector rules for AI-generated frontend design." It is by Paul Bakaus. We use it on this site.

The catch. On Claude Code, Codex, Cursor, GitHub Copilot and Grok Build its installer also offers to add hooks, with yes as the default, and its engine "either sits next to the launcher or is downloaded once on first run into ~/.impeccable/bin/". For Claude Code, the README warns that "the first edit or Stop event can therefore download and cache the engine even if the session denies the model's launcher command", and advises: "Review installed hooks before unattended runs".

6. Playwright CLI

What it is. Microsoft's browser controller for coding agents: "Playwright comes with playwright-cli, a command-line interface for browser automation designed for coding agents." Microsoft's reason for it: "CLI commands avoid loading large tool schemas and verbose accessibility trees into the model context."

Licence and price. Free, Apache 2.0 ("Copyright (c) Microsoft Corporation."). It is installed as its own package, @playwright/cli, and was at version 0.1.21 in September 2026, so it is still pre-1.0.

The catch. It is not the same thing as npx playwright test, the command that runs a Playwright test suite. Microsoft's own pages also disagree on requirements: the README says "Node.js 18 or newer", the docs page "Node.js 20 or newer". Use 20 to be safe.

7. Awesome DESIGN.md

What it is. A collection of DESIGN.md files taken from real websites: "Copy a DESIGN.md into your project, tell your AI agent 'build me a page that looks like this'". Its README says DESIGN.md "is a new concept introduced by Google Stitch". "Awesome Design" in videos most likely means this project; an older, unrelated list called Awesome Design Tools also exists.

Licence and price. The repository is free, MIT ("MIT License Copyright (c) 2026 VoltAgent"). Its companion site, getdesign.md, sells a "Private DESIGN.md" for $34 (shown reduced from $54) and a "Website Starter Kit + DESIGN.md" for $199 (reduced from $299), both one-time.

The catch. The MIT licence covers the files, not the brands. The README says: "We do not claim ownership of any site's visual identity."

Left out

A tool named in one video as "Image 23.js" could not be identified. We searched npm and GitHub for the name and close spellings and found no image library by that name, so it may have been misheard.

Our reading

The pattern across most of the seven: a free core, and a catch that sits right next to it. The licence that counts is the one on the exact block or file being copied, not the one on the project's front page.

Sources

Cult UI: GitHub repository nolly-studio/cult-ui (README, LICENSE); pro.cult-ui.com pricing and terms.
Componentry: componentry.dev (home, blocks, terms of service); GitHub harshjdhv/componentry (LICENSE).
Refero: styles.refero.design (home, DESIGN.md guide); refero.design/pricing.
Taste Skill: tasteskill.dev; GitHub Leonxlnx/taste-skill (README, LICENSE).
Impeccable: impeccable.style; GitHub pbakaus/impeccable (README, LICENSE).
Playwright CLI: playwright.dev (coding agents and test CLI pages); GitHub microsoft/playwright-cli (README, LICENSE); npm registry, @playwright/cli.
Awesome DESIGN.md: GitHub VoltAgent/awesome-design-md (README, LICENSE); getdesign.md.

Checked September 2026.

Buddha Quotes, Checked: What the Early Texts Say and What They Don't

Hacks Vitae — Tue, 22 Sep 2026 00:00:00 GMT

Buddhism is a living religion with many traditions, and this page is not about any of them. It checks popular quotations against one of the oldest collections of the Buddha's teachings, the discourses preserved in Pali. (Parallel early collections survive in Chinese translation.) Even those, Mark Siderits notes in the Stanford Encyclopedia of Philosophy, are "texts that were not written down until several centuries after his death", though they "were probably rehearsed shortly after his death and preserved through oral transmission", and his dates are uncertain: "Up until recently his dates were thought to be approximately 560–480 BCE, but many scholars now hold that he must have died around 405 BCE." So "the Buddha said" here means "the early texts have the Buddha say".

We quote Bhikkhu Sujato's English translations on SuttaCentral, which are released into the public domain, F. Max Müller's 1881 translation of the Dhammapada, and, briefly, Thanissaro Bhikkhu's translations on Access to Insight. To check whether a line is in the early texts, we searched the whole of Sujato's English translation of the discourses, from the four main collections to the Dhammapada and Sutta Nipāta. That tells us whether the words are there, not whether the idea is, so each "not found" below comes with the nearest real passage.

1. "Buddhism teaches that life is suffering"

PARTLY TRUE. Traditionally the first teaching, the Dhammacakkappavattana Sutta (Saṃyutta Nikāya 56.11), does say, in Sujato's translation, "Now this is the noble truth of suffering", and "In brief, the five grasping aggregates are suffering." But it is the first of four truths: suffering, its origin in craving, its cessation, and "the practice that leads to the cessation of suffering", the eightfold path of "right view, right purpose, right speech, right action, right livelihood, right effort, right mindfulness, and right immersion."

The same discourse opens with the "middle way" between two extremes: "Indulgence in sensual pleasures, which is low, crude, ordinary, ignoble, and pointless. And indulgence in self-mortification, which is painful, ignoble, and pointless."

The key word is dukkha, and translators do not agree on it. Sujato writes "suffering", Thanissaro "stress", and Müller, in the Dhammapada, "pain". Siderits explains that "'suffering' is understood to mean not mere pain but existential suffering", and that the first truth is "not the implausible claim that all of life's states and events are necessarily experienced as unsatisfactory".

2. "The Buddha taught that there is no self"

DISPUTED. The text is clear; its meaning is argued. In the discourse on "not-self" (Saṃyutta Nikāya 22.59), each part of a person is examined in turn: "Mendicants, form is not-self", and the listener is taught to see each one as "This is not mine, I am not this, this is not my self." Whether that adds up to "there is no self" is what scholars disagree about. Siderits sets out both sides. Some scholars "(correctly) point out that the Buddha never categorically denies the existence of a self that transcends what is empirically given", though he does deny that any part of a person is a self; others read the teaching as "not a bit of metaphysics, just some practical advice". Against the first reading he records the objection that "all of classical Indian philosophy—Buddhist and orthodox alike—understood the Buddha to have denied the self tout court", and he points out that both readings assume we can know what the Buddha thought "without relying on the understanding of his teachings developed in the subsequent Buddhist philosophical tradition".

3. "What we think, we become"

PARTLY TRUE. It descends from a real verse, Dhammapada 1, but not in these words. In Müller's 1881 translation: "All that we are is the result of what we have thought: it is founded on our thoughts, it is made up of our thoughts." The verse goes on: "If a man speaks or acts with an evil thought, pain follows him, as the wheel follows the foot of the ox that draws the carriage." It is about intention shaping what you say and do, and what follows from it.

Müller's reading was contested in his own lifetime. In his notes he records that the Pali scholar Childers came round to a different reading, and adds: "I therefore adhere to my original translation." Modern translators begin the verse differently again: Sujato, "Intention is the leader of things"; Thanissaro, "Phenomena are preceded by the heart".

4. "The mind is everything. What you think you become."

MYTH as a Buddha quote. Neither sentence is in Sujato's translation of the early discourses, nor in Müller's Dhammapada. The nearest real teaching is about habit, in Majjhima Nikāya 19: "Whatever a mendicant frequently thinks about and considers becomes their heart's inclination."

5. "Peace comes from within. Do not seek it without."

PARTLY TRUE. The wording is not in the early texts, but a verse in the Sutta Nipāta (4.14) says something close. In Sujato's translation: "A mendicant would be settled internally, and not seek peace from another. For one at peace inside themselves, there's no picking up, whence putting down?" It describes a mendicant, in a poem answering a question about "seclusion and the state of peace".

6. "Holding on to anger is like grasping a hot coal; you are the one who gets burned"

PARTLY TRUE. The idea is early; the hot coal is not. No discourse in Sujato's translation of the early texts pairs coals with anger, but the idea that anger harms the angry person first is there. In Aṅguttara Nikāya 7.64, in Sujato's translation: "Anger creates harm; anger upsets the mind", and "when the anger is spent, they're tormented as if burnt by fire". The coal image itself comes from a much later Buddhist work, Buddhaghosa's Visuddhimagga, which the Stanford Encyclopedia dates to the "fifth century CE". In Pe Maung Tin's English translation, a person reflecting on their own anger is told they are "like a man who seizes with both hands glowing live coals or dung in order to strike another man therewith", but who is burned and soiled first. Buddhaghosa then applies a second image to the other person's anger and cites a verse of the Buddha for it, a verse also found in the Dhammapada (125), where the evil "backfires on the fool, like fine dust thrown upwind".

For what the early texts do say about grudges, see Dhammapada 3 and 4, in Sujato's translation: "For those who bear such a grudge, hatred is never laid to rest."

7. "Three things cannot be long hidden: the sun, the moon, and the truth"

PARTLY TRUE. A real discourse lies behind it, Aṅguttara Nikāya 3.131, but the popular line changes its words and its point. The discourse lists three things "conveyed under cover" and three that "dazzle in the open": "The moon dazzles in the open, not under cover. The sun dazzles in the open, not under cover." The third is not "the truth" but "The teaching and training proclaimed by a Realized One". Nothing is said about how long anything stays hidden.

8. "You, yourself, as much as anybody in the entire universe, deserve your love and affection"

MYTH as a Buddha quote. It is not in the early discourses as translated. There is a real discourse about holding oneself dear (Saṃyutta Nikāya 3.8), and it draws a different conclusion. In Sujato's translation: "Having explored every quarter with the mind, one finds no one dearer than oneself. Likewise for others, each holds themselves dear; so one who desires self-knowledge would harm no other."

9. "Believe nothing, no matter where you read it, unless it agrees with your own reason"

PARTLY TRUE. It comes from a real discourse, Aṅguttara Nikāya 3.65, often called the Kālāma Sutta, where the Kālāmas, confused by rival teachers, ask whom to believe. The Buddha tells them not to go by tradition, scripture or a teacher's standing: "don't go by testament, don't go by canonical authority, don't rely on logic, don't rely on inference". So logic and inference are on the list of things not to go by, the opposite of "your own reason". The text does then say "when you know for yourselves", but what you are to know is whether things are "unskillful, blameworthy, criticized by sensible people", and lead "to harm and suffering". "Even if I have said it" has no clear counterpart in the text.

10. "Do not dwell in the past, do not dream of the future"

PARTLY TRUE. This is a fair gist of a real verse, the opening of Majjhima Nikāya 131. In Sujato's translation: "Don't run back to the past, don't anticipate the future. What's past is left behind, the future has not arrived". The verse does ask for clear seeing of "any present phenomenon", but it goes on with an urgency the popular version drops: "Today's the day to keenly work— who knows, tomorrow may bring death!"

11. "Hatred is never ended by hatred"

TRUE. It is Dhammapada 5. In Müller's translation: "For hatred does not cease by hatred at any time: hatred ceases by love, this is an old rule." Sujato: "For never is hatred laid to rest by hate, it's laid to rest by love: this is an ancient teaching." The Pali word translated "love" here, avera, is literally non-hatred (our gloss); Thanissaro renders the line "Hostilities are stilled through non-hostility".

Our reading

Most of the misattributed quotes here have something in common: they are about you. They promise that your thoughts shape reality, that you deserve your own love, that your own reason is the final test. The early texts are usually pointing the other way: at intention and action, at not harming others, at seeing that the things you call "mine" are not you. The real lines are less comforting and more interesting. When a Buddha quote comes without a text and a number, it is worth asking where it is from.

Sources

Bhikkhu Sujato, translations of Saṃyutta Nikāya 3.8, 22.59 and 56.11; Majjhima Nikāya 19 and 131; Aṅguttara Nikāya 3.65, 3.131 and 7.64; Dhammapada 1–20 and 125; Sutta Nipāta 4.14, SuttaCentral (public domain), with the Pali texts.
F. Max Müller, trans., The Dhammapada, Sacred Books of the East vol. 10 (Oxford, 1881), Project Gutenberg and Internet Archive.
Thanissaro Bhikkhu, translations of Saṃyutta Nikāya 56.11 and Dhammapada 1, Access to Insight.
Buddhaghosa, The Path of Purity (Visuddhimagga), Part II, trans. Pe Maung Tin, Pali Text Society, Internet Archive.
Mark Siderits, "Buddha", and Noa Ronkin, "Abhidharma", Stanford Encyclopedia of Philosophy.

Checked September 2026.

Confucius Quotes, Checked: What the Analects Say and What They Don't

Hacks Vitae — Tue, 22 Sep 2026 00:00:00 GMT

A great many sayings are credited to Confucius. His name, the Stanford Encyclopedia of Philosophy notes, is "a Latinized combination of the surname Kong 孔 with an honorific suffix 'Master'", and his dates are traditional: "Confucius (trad. 551–479 BCE)". Here are ten claims, checked against the Analects (Lunyu 論語) in Chinese on Chinese Wikisource and in James Legge's public-domain English translation, second edition of 1893. Legge numbers some chapters one lower than modern editions such as the one the Stanford Encyclopedia uses; we give the modern number and note Legge's where they differ. English after a Chinese line is our translation unless it is marked as Legge's.

1. "Confucius wrote the Analects"

MYTH. Not even the traditional account says he wrote it. The Analects is a collection of his sayings and conversations. Tradition does credit him with one book of his own, the chronicle called the Spring and Autumn Annals; James Legge thought it the only classic that could, "with an approximation to correctness, be described as of his own 'making'". The traditional account, as the Stanford Encyclopedia reports it, is that the Analects was "compiled by first and second generation disciples of Confucius and then transmitted privately for centuries". Some scholars question that: one argument it reports is that the book's "traditional status as the oldest stratum of the teachings of Confucius was undeserved", and another that writers "started to demonstrate an acute interest in the Analects only in the late second and first centuries BCE". The encyclopedia's summary is that "the philosophy of Confucius is historically underdetermined".

Even inside the Analects, not every famous line is his. The saying that filial piety is the root of goodness, in Legge's translation "Filial piety and fraternal submission!-- are they not the root of all benevolent actions?", is spoken by a disciple, "The philosopher Yu", in the book's second passage.

2. "Confucius taught the Golden Rule"

PARTLY TRUE. He taught it mostly in a negative form, and that form appears three times. When his disciple Zigong asks for one word to live by: "其恕乎！己所不欲，勿施於人", "Is it not reciprocity? What you do not want for yourself, do not impose on others" (15.24; Legge and Wikisource number it 15.23; our translation). Legge: "What you do not want done to yourself, do not do to others." The same rule appears at 12.2, and a variant at 5.12 (Legge 5.11), where Zigong says that what he does not want others to do to him, he wishes not to do to them, and Confucius answers "賜也，非爾所及也", "Ci, that is not something you have reached" (our translation; Ci is Zigong's personal name).

A positive version appears at 6.30 (6.28 in Legge and on Wikisource): "己欲立而立人，己欲達而達人", "wanting to be established himself, he establishes others; wanting to succeed himself, he helps others succeed" (our translation). Whether it is the same principle as the Christian Golden Rule is a matter of interpretation. The Stanford Encyclopedia notes that "their interpretation as abstract principles has been influenced by their perceived similarity to the Biblical examples", and that "some scholars, including E. Bruce Brooks, believe these passages to be interpolations."

3. "Real knowledge is to know the extent of one's ignorance"

PARTLY TRUE. A real passage lies behind it (2.17), but it says something slightly different: honesty in both directions. Confucius tells his disciple Zilu: "知之爲知之，不知爲不知，是知也", "To take what you know as known and what you do not know as not known — that is knowing" (our translation). Legge: "When you know a thing, to hold that you know it; and when you do not know a thing, to allow that you do not know it;-- this is knowledge."

4. "It does not matter how slowly you go as long as you do not stop"

MYTH as a Confucius quote. It is not in the Analects: we searched Legge's complete translation and found no "slowly", and nothing that says it in the Chinese. The nearest warning is about haste, not persistence: "Do not be desirous to have things done quickly" (13.17, Legge). Two other passages are closer in spirit. In 9.19 (Legge 9.18), learning is like building a mound: "If there want but one basket of earth to complete the work, and I stop," the stopping is one's own. In 9.21 (Legge 9.20), Confucius says of his student Yan Yuan: "Alas! I saw his constant advance. I never saw him stop in his progress."

5. "The man who moves a mountain begins by carrying away small stones"

MYTH as a Confucius quote. It is not in the Analects. A famous Chinese story about moving mountains does exist, but it is in a different book, the Liezi: the Foolish Old Man of the North Mountain, "年且九十", "nearly ninety" (our translation), sets out with his sons and grandsons to break rocks and carry the earth away in baskets.

6. "Our greatest glory is not in never falling, but in rising every time we fall"

MYTH as a Confucius quote. It is not in the Analects. The earliest source we found is Oliver Goldsmith. His letters from an invented Chinese philosopher, most of which first ran in a London newspaper, the Public Ledger, from 1760, were collected in 1762 as The Citizen of the World, where the line appears: "Our greatest glory is, not in never falling, but in rising every time we fall." It is spoken by Goldsmith's invented Chinese philosopher, Lien Chi Altangi, in Letter VII. The book's own note to that letter says it seems to be "little more than a rhapsody of sentences borrowed from Confucius", which may explain how the line came to be credited to him. We did not find it in the Analects, in Chinese or in Legge's English.

7. "Choose a job you love, and you will never have to work a day in your life"

MYTH as a Confucius quote. Nothing like it is in Legge's complete translation, and we could not find where it comes from.

8. "Before you embark on a journey of revenge, dig two graves"

MYTH as a Confucius quote. We found no word for revenge, and none for graves, in the Chinese text of the Analects. What it does have is Confucius's own answer on repaying harm (14.36). Asked whether resentment should be repaid with kindness, he replies: "何以報德？以直報怨，以德報德", "Then with what will you repay kindness? Repay resentment with uprightness, and kindness with kindness" (our translation). Legge renders it: "Recompense injury with justice, and recompense kindness with kindness."

9. "Everything has beauty, but not everyone sees it"

MYTH as a Confucius quote. We read every passage in the Chinese Analects that uses 美, the word for beautiful or fine, and none says anything like it.

10. "I hear and I forget. I see and I remember. I do and I understand."

MYTH as a Confucius quote, but it has a real Confucian ancestor. It is not in the Analects. The philosopher Xunzi, whom the Stanford Encyclopedia dates to the "third century BCE", wrote: "不聞不若聞之，聞之不若見之，見之不若知之，知之不若行之", "not hearing is not as good as hearing; hearing is not as good as seeing; seeing is not as good as knowing; knowing is not as good as doing" (our translation). The famous three-line English, with "forget", "remember" and "understand", is not in that text either.

Our reading

The Confucius of the popular quotes tends to be short, soothing and about you. The one in the Analects is a teacher talking to students, often telling them they have not got there yet. When a line is credited to Confucius without a book and passage number, it is worth checking; the Analects is short enough to search in an afternoon.

Sources

Lunyu (Analects), Xunzi ch. 8 and Liezi "Tang wen", Chinese texts on Chinese Wikisource.
James Legge, trans., The Chinese Classics, vol. 1, Confucian Analects, 2nd ed. (Oxford: Clarendon Press, 1893; first edition 1861), Project Gutenberg and Internet Archive.
Mark Csikszentmihalyi, "Confucius", and Paul R. Goldin, "Xunzi", Stanford Encyclopedia of Philosophy.
Oliver Goldsmith, The Citizen of the World, Letter VII (1762; 1840 and 1891 reprints), Internet Archive.

Checked September 2026.

Deciding Under Uncertainty: 7 Famous Decision Tools and Where They Come From

Hacks Vitae — Tue, 22 Sep 2026 00:00:00 GMT

When you cannot know how things will turn out, a handful of tools get recommended again and again. Here are seven, traced to where they started and what their authors claimed. This page explains the ideas; it gives no money, betting or investment advice, and the examples are about ordinary choices.

1. "Decision theory says: always choose the option with the highest expected value"

PARTLY TRUE. The formal version is expected utility theory. The Stanford Encyclopedia of Philosophy sums it up: "Its basic slogan is: choose the act with the highest expected utility." Two things are usually lost. First, utility is not money: "Utiles are typically not taken to be units of currency", and "Bernoulli (1738) argued that money and other goods have diminishing marginal utility". Second, the theory has well-known puzzles; "one problematic example is the St. Petersburg game, originally published by Bernoulli". The entry treats it as "a normative theory — that is, a theory of how people should make decisions", and whether you should always follow it is argued at length there. For choosing between two job offers, the useful part is the question it forces: how much does each outcome matter to you, and how likely is it?

2. "The Kelly criterion is an investing formula"

PARTLY TRUE. It is J. L. Kelly Jr.'s paper "A New Interpretation of Information Rate", in the Bell System Technical Journal for July 1956, and it is about information. Kelly imagines a channel "used to transmit the results of a chance situation before those results become common knowledge", and shows that "the maximum exponential rate of growth of the gambler's capital is equal to the rate of transmission of information over the channel." Its later use in betting and investing is another matter, and not one this page covers.

3. The pre-mortem

TRUE as Gary Klein's, described in Harvard Business Review in September 2007. In a premortem, "team members assume that the project they are planning has just failed", and then explain why. Its purpose, in HBR's summary, is "making it safe for dissenters who are knowledgeable about the undertaking and worried about its weaknesses to speak up".

UNVERIFIED is the figure often attached to it, that imagining an event has already happened improves people's ability to find reasons for it by 30%. The study usually cited is Mitchell, Russo and Pennington (1989), on "prospective hindsight", which "involves generating an explanation for a future event as if it had already happened". Its abstract does not report that figure. It says that "in the first experiment, temporal perspective showed little influence while outcome uncertainty strongly affected the nature of explanations for events", and that "explanations for sure events tended to be longer". A pre-mortem treats failure as certain, so that is the factor that matters here. Longer explanations are not the same as better ones, and we could not read the full paper, so we do not repeat the number.

4. One-way doors and two-way doors

TRUE as Jeff Bezos's, in his letter to Amazon shareholders published with the 2015 annual report: "Some decisions are consequential and irreversible or nearly irreversible – one-way doors – and these decisions must be made methodically, carefully, slowly". "We can call these Type 1 decisions." "But most decisions aren't like that – they are changeable, reversible – they're two-way doors." Those, he wrote, "can and should be made quickly by high judgment individuals or small groups." In a footnote on the opposite mistake, which he calls "less interesting", he adds: "there is undoubtedly some survivorship bias. Any companies that habitually use the light-weight Type 2 decision-making process to make Type 1 decisions go extinct before they get large."

The rule of thumb often quoted with it is from the next year's letter, which points back ("I wrote about this in more detail in last year's letter") and adds that "most decisions should probably be made with somewhere around 70% of the information you wish you had. If you wait for 90%, in most cases, you're probably being slow." Note the hedges, and that he was writing about decisions inside a company.

5. Regret minimisation

TRUE as Bezos's own account of one decision, in an interview with the Academy of Achievement in San Antonio on 4 May 2001. Deciding whether to leave a job for what he calls "this crazy thing", after being told the idea would suit "somebody who didn't already have a good job", he used what he called "a 'regret minimization framework.'" "I wanted to project myself forward to age 80". "I knew that if I failed I wouldn't regret that, but I knew the one thing I might regret is not ever having tried." It is one person's account of one choice, though he went on to suggest it to others: "If you can project yourself out to age 80 and sort of think, 'What will I think at that time?' it gets you away from some of the daily pieces of confusion."

6. The inside view

TRUE as Daniel Kahneman and Dan Lovallo's, in "Timid Choices and Bold Forecasts", Management Science (January 1993). They wrote that "Decision makers have a strong tendency to consider problems as unique", and that "overly optimistic forecasts result from the adoption of an inside view of the problem, which anchors predictions on plans and scenarios." The paper pairs that with a second problem that gets less attention: "Overly cautious attitudes to risk result from a failure to appreciate the effects of statistical aggregation". Bold forecasts and timid choices, from the same habit of treating each case alone. "Reference-class forecasting" is a later name for the method, and we did not check who coined it.

7. Superforecasting

TRUE that the research behind it was a real tournament with published results. In Mellers and colleagues' Psychological Science paper (2014), "five university-based research groups competed to recruit forecasters, elicit their predictions, and aggregate those predictions". The authors' group "found support for three psychological drivers of accuracy: training, teaming, and tracking". The training "corrected cognitive biases, encouraged forecasters to use reference classes", the same idea as in 6, and tracking "placed the highest performers (top 2% from Year 1) in elite teams that worked together." A 2015 paper by Mellers and colleagues in Perspectives on Psychological Science describes this as "the winning strategy: culling off top performers each year and assigning them into elite teams of superforecasters". We did not use the claims on the commercial company's own website.

Our reading

Several of these tools came with hedges from the people who proposed them: a 70% rule full of "probably", a paper about information rather than advice, a regret framework described as "a decision that I had to make for myself". The versions that circulate are more certain. The authors' own hedges are worth keeping.

Sources

R. A. Briggs, "Normative Theories of Rational Choice: Expected Utility", Stanford Encyclopedia of Philosophy.
J. L. Kelly Jr., "A New Interpretation of Information Rate", Bell System Technical Journal (July 1956).
Gary Klein, "Performing a Project Premortem", Harvard Business Review (September 2007), summary page.
D. J. Mitchell, J. E. Russo and N. Pennington, "Back to the future: Temporal perspective in the explanation of events", Journal of Behavioral Decision Making 2 (1989).
Jeffrey P. Bezos, letters to Amazon.com shareholders with the 2015 and 2016 annual reports, SEC Exhibit 99.1.
Academy of Achievement, interview with Jeffrey P. Bezos, San Antonio, 4 May 2001.
D. Kahneman and D. Lovallo, "Timid Choices and Bold Forecasts", Management Science 39 (1993).
B. Mellers et al., "Psychological strategies for winning a geopolitical forecasting tournament", Psychological Science 25 (2014); B. Mellers et al., "Identifying and Cultivating Superforecasters as a Method of Improving Probabilistic Predictions", Perspectives on Psychological Science 10 (2015).

Checked September 2026.

Cybersecurity Crash Course: Everything You Need to Know, in Plain Words

Hacks Vitae — Tue, 22 Sep 2026 00:00:00 GMT

This is a crash course for ordinary people and small businesses, not security professionals. It explains how the common attacks work and how to protect yourself, in plain words, drawing on official sources: NIST (the US standards body), the UK's National Cyber Security Centre (NCSC), the US Cybersecurity and Infrastructure Security Agency (CISA), the FBI, Australia's Signals Directorate and others. Figures are as of September 2026. This page gives no legal or financial advice.

Do these first

Our ranking, from what the sources below say matters most for one person's accounts:

Turn on two-step verification on your email first, then everywhere else. Passkeys or a security key are the strongest kind; codes by text message are the weakest, but still better than nothing.
Use a different password for every account, kept in a password manager. The one built into your browser or phone counts.
Switch on automatic updates for your phone, computer, browser and router.
Keep a backup that is not plugged in, and check once that you can restore from it.
Never act on a message's own contact details. If anyone asks for money, a code or new bank details, contact them yourself using details you already had.

The basics in four ideas

Security has three parts, not one. The standard definitions are about keeping information secret (confidentiality: "preserving authorized restrictions on information access and disclosure"), unaltered (integrity: "guarding against improper information modification or destruction") and reachable (availability: "ensuring timely and reliable access to and use of information"). Ransomware attacks the third.

Threat, vulnerability and risk are different words. A threat is who or what could hurt you. A vulnerability is the weak spot, a "weakness in an information system … that could be exploited or triggered by a threat source". Risk depends on how bad it would be and how likely it is. You cannot remove criminals; you can remove weak spots.

Your attack surface is every way in: each account, app, device and person, anywhere "an attacker can try to enter, cause an effect on, or extract data from". Fewer old accounts and apps means fewer doors.

A lot of "hacking" is persuading a person. Social engineering is "an attempt to trick someone into revealing information (e.g., a password) that can be used to attack systems or networks". Verizon's 2026 breach report found a "human element was present in 62% of breaches", and social engineering on its own was "16% of all breaches".

What the numbers measure

Two big annual reports are widely quoted, and they measure different things.

The FBI's IC3 report counts crimes that victims report, mostly in the US. For 2025 it recorded "1,008,597 complaints; $20.877 billion in losses; 26% increase in losses from 2024". It "received complaints from more than 200 countries".
Verizon's 2026 Data Breach Investigations Report (DBIR) studies breaches of organisations: "more than 22,000 were confirmed data breaches involving organizations in 145 countries", with a dataset that "covers Oct 2024 through Nov 2025". Its headline finding: "Exploitation of vulnerabilities is now the most common initial access vector for breaches", at 31%, ahead of stolen or misused logins ("credential abuse") at 13%. In plain words, unpatched software is now the most common way into the organisations Verizon studied.

The threats that matter now

Phishing: fake messages by email, text, call or QR code

A phishing message pretends to be someone you trust. It comes by email, by text (smishing), by phone (vishing) or by QR code (quishing). The UK NCSC says: "It used to be easier to spot scams." The FBI warns that AI tools "can correct for human errors that might otherwise serve as warning signs of fraud", so bad spelling is no longer a reliable tell.

What still gives scams away is the pressure. The NCSC lists five signs: authority, urgency, emotion, scarcity and current events. Its advice: "If you have any doubts about a message, contact the organisation directly", and "use the details from their official website." Phishing and spoofing was the most-reported crime type to the FBI in 2025.

QR codes. The NCSC says "the QR codes used in pubs or restaurants are probably safe for you to scan", but "scanning QR codes in open spaces (like stations and car parks) might be riskier". The FBI adds: check that the code "has not been tampered with, such as with a sticker placed on top of the original code", and "Do not download a QR code scanner app"; use the camera your phone already has.

Business email compromise

A criminal gets into, or imitates, a real email account (a supplier, a boss, a lawyer) and asks for a payment to "new bank details". It was the second-largest loss category in the FBI's 2025 report, at over $3 billion. The FBI's advice is to use "secondary channels and/or two-factor authentication to verify requests for changes in account information". In practice: any change of bank details is confirmed by phone, on a number you already had.

Ransomware

Ransomware is "a form of malware designed to encrypt files on a device, rendering them and the systems that rely on them unusable". Criminals now often steal a copy first and threaten to publish it; CISA calls this "double extortion", which is why a backup alone no longer ends the story. Ransomware was in 48% of the breaches in Verizon's 2026 report, and "small organizations are disproportionally impacted".

Should you pay? The UK's NCSC says it and UK police "do not encourage, endorse nor condone the payment of ransom demands": "there is no guarantee that you will get access to your data", and "you're more likely to be targeted in future". Its protection is simple: "always have a recent offline backup of your most important files and data."

Stolen and reused passwords

When one website leaks passwords, criminals feed them automatically into other sites' login forms. This is called credential stuffing, and it works "since many users will re-use the same password". A different password for every site defeats it, and the OWASP security project calls multi-factor authentication "by far the best defense against the majority of password-related attacks".

Infostealers

An infostealer is malware that quietly copies saved passwords, cards and crypto wallets, and also "application tokens and session cookies", the small files that keep you logged in. With those, a criminal can sometimes get in without your password or your code. Europol says "phishing techniques are the main vector for the distribution of infostealers". In one 2025 operation, Microsoft found "over 394,000 Windows computers globally infected" by a single infostealer in two months.

SIM swapping

A criminal persuades your mobile provider to move your number to their SIM card. "Once the SIM is swapped, the victim's calls, texts, and other data are diverted to the criminal's device", including text-message login codes. This is one reason text-message codes rank below other kinds of two-step verification. Our advice: ask your mobile provider what extra protection it offers for your number.

Tech-support and "government agent" scams

"Tech support scams often start with a bogus warning about a problem with your computer." The fake technician wants remote access and payment, and "they often insist that you pay with gift cards, a wire transfer, a bank transfer, cryptocurrency, or a payment app". A variant hands you to a fake official who tells you to move your money somewhere "safe". The US FTC: "Someone who works for the government will never tell you to put your money in a federal safety locker." Tech and customer-support scams were the third-largest loss category in the FBI's 2025 report.

Relationship investment scams

A stranger met through a text, social media or a dating app builds a friendship, then introduces an "investment" platform showing fake profits. "Cryptocurrency investment fraud was the highest source of financial losses to Americans in 2025 with $7.2 billion reported in losses." When victims try to withdraw, "they are told they need to pay a fee or taxes." Afterwards, fake "recovery" services often target the same people; the FBI says: "Do not pay for services that claim to be able to recover lost funds."

Deepfake voices and video

"Criminals generate short audio clips containing a loved one's voice to impersonate a close relative in a crisis situation", and fake live video calls with "alleged company executives, law enforcement, or other authority figures". In early 2024, Hong Kong police said an employee of a multinational firm was tricked into a video call with fakes of the finance chief and colleagues and paid out 200 million Hong Kong dollars. The FBI says such content "is often difficult to identify", so do not rely on spotting glitches. The defences are procedural: "Create a secret word or phrase with your family to verify their identity", and hang up and call back on a number you already have.

Data breaches

When a company you use is breached, your email, phone number or password may circulate. You cannot prevent it, but you can make it harmless: unique passwords and two-step verification. The free site Have I Been Pwned lets you check whether your email address appears in known breaches; in September 2026 it listed more than 17.8 billion breached account records.

The protections that work

Passwords: length beats complexity

NIST SP 800-63B, the US government's rules for logins to its own services (revised in July 2025), now says:

a password used on its own must be "a minimum of 15 characters in length"; one used with a second step may be shorter, but must be "a minimum of eight characters in length";
services "SHALL NOT impose other composition rules (e.g., requiring mixtures of different character types)";
they "SHALL NOT require subscribers to change passwords periodically", but "SHALL force a change if there is evidence that the authenticator has been compromised";
they must check new passwords "against a blocklist that contains known commonly used, expected, or compromised passwords", and "SHALL allow the use of password managers and autofill functionality".

These rules bind US government services, not you or other websites, but they tell you what works. The UK NCSC's advice is to "combine three random words to create a password". It says "longstanding advice around making your passwords very complex … is not helpful", because complex passwords are hard to remember, and swapping letters for symbols is a trick "cyber criminals know … as well". NCSC's own three-word example is 13 characters, under NIST's 15 for a password used on its own, so use three or four random words, and add two-step verification. Writing a password down is "also OK, provided you keep it somewhere safe."

Password managers

"A password manager stores passwords safely for you, meaning that you can have unique passwords for each service." It also "helps protect you from phishing attacks as the password will only autofill on the correct website." The one built into your browser or phone is fine on your own devices, but on shared computers "you should never save your password in the browser." Protect the manager itself with two-step verification.

Two-step verification (MFA), and why some kinds are stronger

Multi-factor authentication means proving it is you with a second thing beyond the password. CISA says "any form of MFA is better than no MFA", but "not all forms of MFA are equally secure", and "phishing-resistant MFA is the gold standard". From strongest to weakest:

Passkeys and security keys (the FIDO standard), which a fake website cannot capture.
Authenticator apps, or login approvals where you type a matching number.
Plain "approve this login?" pop-ups, which criminals abuse by flooding you with requests until you tap yes ("push bombing").
Codes by text or voice call, which a SIM swap can redirect.

A 2023 preprint by Microsoft researchers, on business accounts that showed suspicious activity, found MFA "reduces the risk of compromise by 99.22%". Microsoft's widely quoted "99.9 percent" comes from a 2019 company blog post.

Passkeys

"A passkey is an authentication credential based on FIDO standards, that can be stored on your phone or computer, or in a hardware security key", and you unlock it with "the same process that they use to unlock their device". It resists phishing because it only works on the website that created it; NIST says this "prevents a falsified web page from being able to capture and reuse an authenticator output." Use passkeys wherever a site offers them.

Updates

The NCSC calls updating "one of the most important (and quickest) things you can do to keep yourself safe online", and says to "turn on 'automatic updates' in your device's settings". CISA warns that "attackers may target vulnerabilities for months or even years after updates are available". A device that no longer gets updates "won't receive the security updates from the manufacturer", so plan to replace it.

Backups: the 3-2-1 rule

"Keep 3 copies of any important file: 1 primary and 2 backups", "keep the files on 2 different media types", and "store 1 copy offsite". CISA's own paper credits the rule to a photographers' book, Peter Krogh's The DAM Book. The ransomware lesson on top: "When the removable media isn't in use, it's important that you disconnect it", because malware "can move to attached media automatically".

Device encryption

Encryption protects your data if a device is lost or stolen. Windows turns on "BitLocker encryption automatically" on many machines, though "if you're using a local account, Device Encryption isn't turned on automatically". On Macs with Apple silicon, "your data is encrypted automatically", but Apple says turning on FileVault is what keeps "someone from decrypting or getting access to your data without entering your login password", so turn it on. Keep the recovery key safe: Apple warns that if you forget both your password and the key, "your files and settings will be lost forever."

Antivirus

You do not need to buy it for most devices. The NCSC's device guidance says "you should not need to use AV products on platforms like Chrome OS, Android and iOS in their default configuration." Windows includes "Microsoft Defender Antivirus, Windows Firewall, and Smart App Control". Keep the built-in protection on, and install software only from official stores and websites.

VPNs

A VPN moves who can see your traffic from your Wi-Fi network and internet provider to the VPN company. The digital- rights group EFF says "a VPN is not a tool for anonymity", and "while a VPN hides your browsing data from the ISP, it's all visible to the VPN provider." It can help on an untrusted network, "and you have a VPN you trust". A VPN from your employer, for reaching work systems, is a different thing.

Your browser and home network

Leave your browser's built-in dangerous-site protection on; in Chrome, "Standard protection offers security from known dangers." For your router, CISA's advice: "Change your router's administrator password", "use the strongest encryption protocol available", and "disable Wi-Fi Protected Setup (WPS)".

What you share

Public details feed scams: birthdays and pet names answer security questions, and voice clips feed voice clones. The NCSC says "you can reduce the likelihood of being phished by thinking about what personal information you (and others) post about you", and the FBI suggests you "limit online content of your image or voice".

Myths, checked

MYTH: "I'm not a target." CISA lists "their home network is too small to be at risk of a cyberattack" as a misconception, because "most attacks are not personal in nature". The NCSC compares most attacks to "a thief trying your front door to see if it's unlocked."

MYTH: "Macs don't get viruses." Apple itself says "macOS includes built-in antivirus technology called XProtect". Microsoft's researchers report that "since late 2025", they have seen "macOS targeted infostealer campaigns using social engineering techniques". Apple's built-in protection helps; it does not make Macs immune.

MYTH: "Incognito makes you anonymous." Google's own help page: "While Incognito can help keep your browsing private on your device, it doesn't make you invisible." Your school, employer or internet provider "may be able to observe your activity in Incognito".

MYTH: "A padlock means the site is safe." The padlock means the connection is encrypted, not that the site is honest. Google's Chrome team wrote that "the lock icon does not indicate website trustworthiness", that "only 11% of study participants correctly understood the precise meaning of the lock icon", and that "nearly all phishing sites use HTTPS, and therefore also display the lock icon."

PARTLY TRUE: "Public Wi-Fi will get you hacked." The agencies differ in emphasis. The US FTC says "because of the widespread use of encryption, connecting through a public Wi-Fi network is usually safe." Australia's ACSC still calls hotspots "an attractive target for cybercriminals", and advises turning off "auto-join" and: "If your browser displays a warning message when you try to visit a website, do not continue." The ACSC also says: "Reconsider your need to access sensitive information, such as your online banking." The risk is small, not zero.

MYTH: "Change your password every 90 days." The NCSC advised against forced changes in 2015, because "the new password will be similar to the old one", and NIST now forbids US government services from requiring it. Change a password when there is a reason: a breach, a phishing slip, a shared login. Not every official source agrees: the FBI's 2024 advice on business email compromise still says to change passwords "periodically". We follow NIST and the NCSC, because forced changes push people toward predictable passwords.

MYTH: "A VPN makes you anonymous and secure." See above: the VPN company sees what your internet provider used to see. It does not stop phishing, malware or you logging in to your own accounts.

MYTH: "Caller ID proves who is calling." CISA: "VoIP easily allows caller identity (ID) to be spoofed".

If you have been hacked or scammed

Order matters. Money first, then your email (it resets everything else), then everything else.

Lost money? Call your bank or card provider now, on the number on your card, and ask them to stop or recall the payment. The FBI: "If you discover a fraudulent transfer, time is of the essence."
Recover the account through the provider's own help pages, not a search ad or a "recovery service".
Check your email for forwarding rules. "A common tactic used by cyber criminals is to set up a forwarding rule" that copies your mail to them.
Change the password on the hacked account and "for any accounts that are using the same password".
"Log all devices and apps out of your account". This also shuts out stolen login cookies.
Turn on two-step verification, and update the device.
If a device is infected, disconnect it from the network and restore it from a clean backup. Our advice: change passwords from a different, clean device.
US readers whose identity details leaked can freeze their credit: "When a credit freeze is in place, nobody can open a new credit account in your name", and "there's no cost to place or lift a credit freeze". Other countries have different systems.
Keep evidence (screenshots, numbers, transaction references) and report it, even if you lost nothing.

Where to report it

Where you are	Report to	Phone
Anywhere	Your local police; the FBI's IC3 also accepts complaints from outside the US	—
United States	ic3.gov (FBI); ReportFraud.ftc.gov (FTC) for scams	—
UK (England, Wales, Northern Ireland)	Report Fraud, reportfraud.police.uk	0300 123 2040
UK (Scotland)	Police Scotland	101
UK, scam emails and texts	Forward emails to report@phishing.gov.uk; forward texts to 7726	7726
Australia	ReportCyber at cyber.gov.au; Scamwatch for scams	1300 292 371 (emergencies: 000)
Canada	Canadian Anti-Fraud Centre, reportcyberandfraud.canada.ca	1-888-495-8501
India	National Cyber Crime Reporting Portal, cybercrime.gov.in	1930

Elsewhere, contact your national police or cybercrime unit. Criminals impersonate these agencies too: the FBI warns about anyone "impersonating or claiming to work with IC3".

For small businesses

In our reading, government baselines agree more than they differ. The UK's Cyber Essentials is "the minimum standard of cyber security recommended by the Government for organisations of all sizes", with five controls: firewalls, secure configuration, security update management, user access control and malware protection. Australia's Essential Eight is: "patch applications patch operating systems multi-factor authentication restrict administrative privileges application control restrict Microsoft Office macros user application hardening regular backups." The US NIST Cybersecurity Framework 2.0 organises everything under six headings: "GOVERN, IDENTIFY, PROTECT, DETECT, RESPOND, and RECOVER". (The US CISA also has a guide called Cyber Essentials, for small-business leaders; it is a different thing from the UK scheme.)

The short version for a small team:

"Too small to be a target" does not hold up. The NCSC: "if you think your business is too small to be a target, think again."
Least privilege: give each person only the access their job needs, "the minimum necessary to accomplish assigned tasks", and, our advice, remove access the day someone leaves.
MFA for everyone. Where phishing-resistant MFA is not yet possible, CISA says app-based codes or "mobile push with number matching are the best options for small- and medium-size business".
A payment rule: no change of supplier bank details without a phone call to a known number.
Backups: offline, encrypted and tested.
Staff training is recommended by several government baselines, but its effect is disputed. Earlier studies found people scored better on security quizzes straight after training, but within four to five months they were back where they started. The strongest test we found, a randomised study of "over 19,500 employees at a large healthcare organization" (2025), found that common training formats "offer limited value", and an earlier study of about 14,000 employees found no benefit either. The 2025 study's authors suggest technical controls "may offer a better return on investment". That does not make training useless; it means controls that do not depend on people noticing, like MFA and payment call-backs, should come first.

Glossary

Term	In plain words
Attack surface	Every way in: accounts, apps, devices, people.
Authenticator app	An app that shows login codes or approves sign-ins.
Backup	A spare copy you can restore from.
Breach	An incident where data is confirmed to have reached someone it should not have.
Business email compromise	A real or faked business email account used to redirect a payment.
Credential stuffing	Trying leaked username-and-password pairs on other sites, automatically.
Deepfake	Fake audio, image or video of a real person, made with AI.
Double extortion	Ransomware that also steals the data and threatens to publish it.
Encryption	Scrambling data so only someone with the key can read it.
Firewall	A gatekeeper that blocks unwanted network traffic.
HTTPS / padlock	An encrypted connection to a website; not proof the site is honest.
Infostealer	Malware that copies saved passwords, cards, wallets and login cookies.
Least privilege	Each person gets only the access their job needs.
Malware	Any software made to do harm.
MFA / 2FA / 2SV	Proving it is you with two different kinds of thing: something you know, have or are.
Passkey	A login that replaces the password with a key on your device, unlocked by face, fingerprint or PIN.
Password manager	An app or browser feature that creates and remembers a different password for every site.
Patch / update	A fix for a flaw in software.
Phishing	A fake message pretending to be someone you trust, to get your details or money.
Phishing-resistant MFA	Two-step verification a fake website cannot capture: passkeys and security keys.
Push bombing	Flooding someone with login approvals until they tap "accept".
Quishing	Phishing through a QR code.
Ransomware	Malware that locks your files and demands payment.
Session cookie	A small file that keeps you logged in; stealing it can skip the login.
SIM swap	Getting your phone number moved to a criminal's SIM card.
Smishing / vishing	Phishing by text message / by phone call.
Social engineering	Tricking a person rather than breaking a machine.
VPN	An encrypted tunnel to a provider, who then sees your traffic instead of your internet provider.
Zero-day	An attack on a flaw the maker did not yet know about, so no fix exists.

Sources

NIST, SP 800-63B (revision 4, July 2025); NIST Computer Security Resource Center glossary; NIST Cybersecurity Framework 2.0 (February 2024).
UK NCSC: How to spot a scam; Report a scam email; Report a scam text; Three random words; Managing your passwords; Install the latest updates; Backing up your data; Recovering a hacked account; Ransomware; QR codes: what's the real risk?; The problems with forcing regular password expiry; Antivirus and other security software; Social media: how to use it safely; Cyber Essentials; Small organisations guide.
CISA: Avoiding social engineering and phishing attacks; Implementing phishing-resistant MFA (October 2022); #StopRansomware Guide; Understanding patches and software updates; Home network security; Data backup options; Cyber Essentials.
FBI Internet Crime Complaint Center: 2025 IC3 Annual Report; public service announcements on generative AI fraud (December 2024), business email compromise (September 2024), SIM swapping (February 2022), QR codes (January 2022), cryptocurrency investment fraud (March 2023) and impersonation messaging (May 2025).
US FTC: tech support scams; public Wi-Fi; AI family-emergency scams; QR codes; credit freezes.
Verizon, 2026 Data Breach Investigations Report, executive summary.
Europol, IOCTA 2025; Microsoft, Lumma Stealer disruption (May 2025), macOS infostealers (February 2026) and a 2023 MFA study (arXiv 2305.00945); OWASP; FIDO Alliance; EFF Surveillance Self-Defense; Google Chrome Help; Chromium Blog (May 2023); Apple and Microsoft support pages; Have I Been Pwned.
Australia: ASD's ACSC (report, public Wi-Fi, Essential Eight) and Scamwatch. Canada: Canadian Anti-Fraud Centre. India: National Cyber Crime Reporting Portal and PIB (February 2024). UK: Report Fraud.
Ho et al., "Understanding the Efficacy of Phishing Training in Practice", IEEE Symposium on Security and Privacy 2025.
CNN (May 2024), citing Hong Kong police.

Checked September 2026.

Epicurus: What He Actually Taught About Pleasure, Death and the Gods

Hacks Vitae — Tue, 22 Sep 2026 00:00:00 GMT

Epicurus ran a philosophical school in Athens called the Garden, and his name now means someone who loves fine food. That is close to the opposite of what he taught. Diogenes Laertius counted about three hundred rolls of his writing. Three of his letters and a list of his key sayings, the Principal Doctrines, survive in full because that later biographer copied them into his Lives of Eminent Philosophers. We quote them from R. D. Hicks's 1925 translation. Here are eight common claims, checked.

1. "Epicurus taught people to live for pleasure"

PARTLY TRUE. He did make pleasure the goal of life, so in the technical sense he was a hedonist. But he meant something unusual by it. In his Letter to Menoeceus he writes: "When we say, then, that pleasure is the end and aim, we do not mean the pleasures of the prodigal or the pleasures of sensuality". Instead: "By pleasure we mean the absence of pain in the body and of trouble in the soul." It is "not an unbroken succession of drinking-bouts and of revelry", but "sober reasoning, searching out the grounds of every choice and avoidance".

One of his Principal Doctrines sets the ceiling: "The magnitude of pleasure reaches its limit in the removal of all pain." On his view, once nothing hurts and nothing troubles a person, more luxury adds nothing. Epicurus's own word for the untroubled mind is ataraxia, usually translated "tranquillity". The philosopher Tim O'Keefe, writing in the Internet Encyclopedia of Philosophy, suggests "it might be less misleading to call him a 'tranquillist' instead of a 'hedonist.'"

He was not against bodily pleasure, though. Diogenes quotes, in a passage listing what his critics said about him, a line from his book On the Ethical End: "I know not how to conceive the good, apart from the pleasures of taste, sexual pleasures, the pleasures of sound and the pleasures of beautiful form." How that fits with the rest of his ethics is still argued; the Stanford Encyclopedia of Philosophy distinguishes what Epicurus called "static" from "kinetic" pleasure and notes that the precise nature of this distinction "is debated".

2. "An epicure loves fine food, and so did Epicurus"

MYTH. By his own account he lived very plainly. Diogenes reports that "In his correspondence he himself mentions that he was content with plain bread and water", and quotes him asking a friend: "Send me a little pot of cheese, that, when I like, I may fare sumptuously." In the Letter to Menoeceus he writes that "bread and water confer the highest possible pleasure when they are brought to hungry lips."

The Roman philosopher Seneca describes a welcome at the Garden of "barley-meal" and "water also in abundance". The food-lover meaning of epicure came much later. According to the Online Etymology Dictionary, the Stoics "first gave his name a reproachful sense"; English dictionaries date the insulting sense of epicure to the mid-1500s, and the sense of a person with refined taste in food and drink came after it.

3. "Epicurus said death is nothing to us"

TRUE. He said it, and argued for it: "Accustom thyself to believe that death is nothing to us, for good and evil imply sentience, and death is the privation of all sentience". The famous short version, often quoted as "Where death is, I am not", is a paraphrase of the next lines: "when we are, death is not come, and, when death is come, we are not." He adds that people fear death "not because it will pain when it comes, but because it pains in the prospect."

The Roman poet Lucretius, who set out Epicurus's system in the Latin poem On the Nature of Things, restates it: in William Ellery Leonard's verse translation, "Therefore death to us Is nothing, nor concerns us in the least". Lucretius also adds the mirror argument: we felt nothing in all the ages before we were born.

4. "Epicurus was an atheist"

MYTH. It was an ancient charge, and he denied it. "For verily there are gods, and the knowledge of them is manifest; but they are not such as the multitude believe". What he rejected was the idea that gods punish, reward or run the world. The first of his Principal Doctrines says: "A blessed and eternal being has no trouble himself and brings no trouble upon any other being". O'Keefe notes that Epicureanism was "charged in antiquity with being a godless philosophy, although Epicurus and his followers denied the charge."

What he meant by "gods" is a real dispute. O'Keefe describes the gods as "unaware of our existence", living eternally "in the intermundia, the space between the cosmoi", but he also reports that some scholars "conjecture that Epicurus' 'gods' are thought-constructs, and exist only in human minds as idealizations". The Internet Encyclopedia also notes that ancient critics thought the gods "a thin smoke-screen to hide Epicurus' atheism". The texts do not settle it.

5. "The Epicurean paradox about evil is Epicurus's"

DISPUTED. The argument goes: if God wants to stop evil but cannot, he is not all-powerful; if he can but will not, he is not good; so where does evil come from? It survives in the Christian writer Lactantius's On the Anger of God, chapter 13, which introduces it as "that argument also of Epicurus" and then sets out to refute it. No surviving work of Epicurus contains it. An earlier version does survive, without his name. The Sceptic Sextus Empiricus, who lived probably in the second or third century CE, sets out the same four cases in his Outlines of Pyrrhonism (book 3): a god who wills to care for everything but cannot is weak, one who can but will not would seem envious, and one who neither wills nor can is both. He does not attribute it to Epicurus.

Its famous English wording is not a translation of Epicurus either. It comes from David Hume's Dialogues Concerning Natural Religion, published after his death, where Hume has his character Philo say: "EPICURUS's old questions are yet unanswered. Is he willing to prevent evil, but not able? then is he impotent."

Scholars disagree about whether the argument is Epicurus's at all. The 1893 critical edition of Lactantius records one scholar's suggestion that the passage came from a lost part of Cicero's On the Nature of the Gods, and a 1988 article by Reinhold Glei calls it, in its title, "the alleged Epicurus fragment" (our translation). On the other side, O'Keefe counts Epicurus among "the earliest philosophers we know of to have raised the Problem of Evil", because he argued against a world run by a caring god. Both can be true: Epicurus denied that the gods govern the world. But the neat four-way argument is credited to him only by a later writer who was arguing against him and names no source, and an older version survives in a Sceptic's handbook with no mention of him. On that evidence, the argument is better described as used against providence in antiquity than as Epicurus's own words.

6. "Epicurus's four-part cure"

PARTLY TRUE. The ideas are his, but the four-line formula does not appear in any of his own writings that survive. The summary, often called the tetrapharmakos or "fourfold drug", is known from Philodemus, an Epicurean philosopher who lived about two centuries later. His version survives on a charred papyrus from Herculaneum, "buried by the eruption of Vesuvius in 79 CE". In the translation of the scholar David Blank: "god is not fearsome, death not frightening, the good easily got, the bad easily endured". The four lines match the themes of Epicurus's first four Principal Doctrines: the fourth, for example, says that pain "if extreme, is present a very short time".

7. "The Garden was a pleasure commune"

MYTH. It was a household of friends. Diogenes writes that "Friends indeed came to him from all parts and lived with him in his garden", and his Principal Doctrines put friendship first: "by far the most important is the acquisition of friends." He refused to make his followers pool their property, because "such a practice in his opinion implied mistrust, and without confidence there is no friendship."

His will provided for a meeting of the school every month on the twentieth day and freed several of his slaves. On what he called the last day of his life he wrote to his friend Idomeneus that his sufferings were so great "that nothing could augment them; but over against them all I set gladness of mind at the remembrance of our past conversations."

8. "His bad reputation came from later readers who misunderstood him"

PARTLY TRUE. It began in his own lifetime. Epicurus himself complained that he was misread "through ignorance, prejudice, or wilful misrepresentation". Diogenes records the slanders of his enemies: one "adducing fifty scandalous letters as written by Epicurus", another claiming "that Epicurus vomited twice a day from over-indulgence", and, centuries later, Epictetus calling him a "preacher of effeminacy". Diogenes rejects them, and the Stanford Encyclopedia notes that Epicurus's humane character "was the prevailing view, shared even by hostile witnesses".

The image stuck anyway. The Roman poet Horace joked about himself as "a hog of Epicurus' herd", in C. Smart's translation, and by Shakespeare's Macbeth "epicures" was an insult.

Our reading

Epicurus was a hedonist whose idea of pleasure was not being in pain, not being afraid, and eating bread with friends. His enemies turned that into a byword for indulgence, and the word epicure carries their version, not his. The core of what survives in his own words is short enough to read in an afternoon: three letters and a short list of sayings, most of it arguments against fear, of the gods and of death. More survives in pieces: a second collection of sayings, fragments quoted by later writers, and damaged parts of his long work On Nature recovered from the papyri at Herculaneum.

Sources

Diogenes Laertius, Lives of Eminent Philosophers, Book X, trans. R. D. Hicks (Loeb Classical Library, 1925), on Wikisource: the Letter to Menoeceus and the Principal Doctrines.
David Konstan, "Epicurus", Stanford Encyclopedia of Philosophy (revised 2022); David Blank, "Philodemus", Stanford Encyclopedia of Philosophy (revised 2019).
Tim O'Keefe, "Epicurus", Internet Encyclopedia of Philosophy.
Seneca, Moral Letters to Lucilius 21, trans. Richard Mott Gummere (Loeb, 1917).
Lucretius, On the Nature of Things, trans. William Ellery Leonard, Project Gutenberg.
Lactantius, A Treatise on the Anger of God, ch. 13, trans. William Fletcher, Ante-Nicene Fathers vol. 7 (1886); Latin text ed. Samuel Brandt and Georg Laubmann, Corpus Scriptorum Ecclesiasticorum Latinorum 27 (1893).
Reinhold Glei, "Et invidus et inbecillus. Das angebliche Epikurfragment bei Laktanz, De ira dei 13,20–21", Vigiliae Christianae 42 (1988), pp. 47–58 (title and record only).
David Hume, Dialogues Concerning Natural Religion, Part 10, Project Gutenberg; "Hume on Religion", Stanford Encyclopedia of Philosophy.
Sextus Empiricus, Outlines of Pyrrhonism 3, German trans. Eugen Pappenheim, Pyrrhonische Grundzüge (1877); Benjamin Morison, "Sextus Empiricus", Stanford Encyclopedia of Philosophy (revised 2026).
Horace, Epistles 1.4, trans. C. Smart, Project Gutenberg; Shakespeare, Macbeth, Project Gutenberg.
Online Etymology Dictionary and Merriam-Webster, "epicure".

Checked September 2026.

Existentialism, Checked: What Kierkegaard, Sartre and Camus Actually Wrote

Hacks Vitae — Tue, 22 Sep 2026 00:00:00 GMT

If you are struggling or thinking about suicide, please reach out now. findahelpline.com lists free, confidential helplines worldwide. In the US and Canada, call or text 988. In the UK and Ireland, call Samaritans on 116 123. In Australia, call Lifeline on 13 11 14. In India, call Tele-MANAS on 14416. In the Philippines, call the National Center for Mental Health crisis line on 1800-1888-1553 (or 1553 from a Luzon landline). If you are in immediate danger, call your local emergency number.

Existentialism is the philosophy of freedom, choice and meaning, and it is quoted everywhere. Here are nine claims about its three most-quoted names, Søren Kierkegaard, Jean-Paul Sartre and Albert Camus. Sartre and Camus are still in copyright, so we quote them briefly, through the scholars and reference works that quote them: the Stanford Encyclopedia of Philosophy, the Internet Encyclopedia of Philosophy, Oxford Essential Quotations, the Société des Études camusiennes and the publisher Gallimard.

1. "Existentialism was a school, and Kierkegaard, Sartre and Camus belonged to it"

MYTH. The Stanford Encyclopedia says the word was "first introduced by Marcel in 1943" and "is certainly not a reference to a coherent system or philosophical school." Of the thinkers usually grouped under it, "only Sartre and Beauvoir explicitly self-identified as 'existentialists.'" Marcel, who introduced the word, at first accepted "Christian Existentialism" as a name for his own work, the Stanford Encyclopedia's entry on him notes, and "later repudiated" it. Camus was not one of the two: according to the Stanford Encyclopedia, Sartre "accepted the 'existentialist' label that Camus rejected", and the Internet Encyclopedia says "throughout his career Camus repeatedly denied that he was an existentialist". Heidegger, often listed too, answered questions about the movement in a letter published in 1947 as the Letter on Humanism; there, in the Internet Encyclopedia's words, "he repudiates any possible connection of his philosophy with the existentialism of Sartre."

Kierkegaard could not have joined: he lived from 1813 to 1855. The Stanford Encyclopedia treats him as one of the nineteenth-century "pioneers" who laid "the conceptual groundwork" long before the name existed.

2. "Kierkegaard is the father of existentialism"

PARTLY TRUE. He is often called that, and the influence is real. But the Stanford Encyclopedia's entry on him warns: "this label obscures at least as much as it reveals, especially to those who associate existentialism with atheistic figures such as Sartre." Kierkegaard, born in Copenhagen on 5 May 1813, said in his own voice that he was "from first to last a religious author", as the Stanford Encyclopedia reports (a claim the same entry says "has been hotly contested"), and the entry notes that his writings have influenced non-Christian thinkers "despite his own Christian motivations".

3. "Kierkegaard said we must take a leap of faith"

MYTH as a quotation. The idea of a leap is his; the three-word phrase is not. Kierkegaard wrote about a Spring, Danish for "leap". The Stanford Encyclopedia summarises one of his books, The Concept of Anxiety (1844): "a morally free action always comes from a 'leap'".

The phrase itself was searched for by computer. The philosopher Alastair McKinnon, writing in the journal Kierkegaardiana in 1993, searched a machine-readable copy of the third edition of Kierkegaard's collected works and reported that the Danish equivalent "does not occur in any of the thirty five works in that edition". His conclusion: "Kierkegaard never used this phrase". Mariele Nientied (2003) reports the same result: "leap of faith" "does not occur even once in the published writings". Both results cover his published works, not his private journals.

DISPUTED is the fairer verdict on whether it is a good summary. Reference works still use it as a description: the Internet Encyclopedia writes of "the religious leap of faith", and the Stanford Encyclopedia has Abraham "'leaping' (springer) into a paradox". McKinnon grants that Kierkegaard "plainly thought of faith as involving something like a leap", but argues the slogan should go: in his counts the Danish words for faith and leap show "a marked tendency not to co-occur" in the same sentences, the leap belongs with sin and the Fall rather than with faith, and the phrase is "incoherent and meaningless" because it builds faith into the leap that is supposed to reach faith. Nientied allows that the missing wording is not decisive, since the phrase "might be seen as an appropriate summary of Kierkegaardian thinking although it cannot be quoted", then argues that the charges it has been used to support, such as irrationalism, rest on "a misleading interpretation". So: wrong in quotation marks; argued over as a loose description; and McKinnon and Nientied both reject it as evidence that he gave up on reason.

4. "Existence precedes essence"

TRUE. It is Sartre's. The Stanford Encyclopedia's entry on him says he "famously remarks" it "in both Being and Nothingness and Existentialism is a Humanism", the public lecture that "many thousands of Parisians packed into" "towards the end of 1945". Oxford Essential Quotations gives a fuller wording from Being and Nothingness (1943): "Existence precedes and rules essence."

What it means, in the Stanford Encyclopedia's words: "there is no pre-given or essential nature that determines us." The Stanford Encyclopedia's entry on existentialism notes a precursor in Heidegger's Being and Time: "The 'essence' of Dasein lies in its existence".

5. "Hell is other people"

TRUE as a line, but its popular meaning is argued. It is in Sartre's play No Exit (Huis clos), first staged in 1944; Oxford Essential Quotations gives "Hell is other people" and places it in scene 5. The Stanford Encyclopedia reads the line as the result of a "struggle for self-assertion" between people, and the Internet Encyclopedia calls the play a dramatisation of a "'hell' of endlessly circling acts of freedom and objectification".

It is often read as a plain complaint about other people. Sartre's own gloss adds something the slogan leaves out. Gallimard, his publisher, issued a 1964 studio recording of the play, preceded by an interview Sartre gave to Moshé Naïm, and its page for the recording quotes him saying he wanted to show "l'importance, chez nous, de la liberté", "the importance, for us, of freedom", and: "Quel que soit le cercle d'enfer dans lequel nous vivons, je pense que nous sommes libres de le briser." In our translation (both lines): "Whatever circle of hell we live in, I think we are free to break it."

6. "One must imagine Sisyphus happy"

TRUE. It is the last line of Camus's essay The Myth of Sisyphus, finished in February 1941 and published by Gallimard in October 1942, according to the Société des Études camusiennes. In French: "la lutte elle-même vers les sommets suffit à remplir un cœur d'homme. Il faut imaginer Sisyphe heureux." The English translation quoted in the Stanford Encyclopedia reads: "The struggle itself toward the heights is enough to fill a man's heart. One must imagine Sisyphus happy".

The essay opens with its famous first line, which in French reads "Il n'y a qu'un problème philosophique vraiment sérieux : c'est le suicide", "there is only one truly serious philosophical problem: suicide". English translations word it differently, so quote the translation you have. The question is asked in order to be answered, and the Internet Encyclopedia states the answer plainly: "In the end, Camus rejects suicide." On the same page it sums up his view that "the task of living should not merely be accepted, it must be embraced."

7. "Life begins on the other side of despair" — Sartre

PARTLY TRUE. A version of it is real, but the popular wording drops a word and it is not Sartre speaking in his own voice. Oxford Essential Quotations gives it as "Human life begins on the far side of despair", from his play The Flies (Les Mouches, 1943), act 3, scene 2. It is a character's line in a play, not a statement in an essay.

8. "Don't walk behind me; I may not lead. Don't walk in front of me; I may not follow. Just walk beside me and be my friend" — Camus

MYTH. Giovanni Gaetani's study of Camus misquotes in the Journal of Camus Studies (2015) puts it first on his list and concludes that nobody can give a source, "since Camus simply never wrote it". He found that it "belongs to a Jewish song for children". Who first wrote it, and how Camus's name got attached to it, is not known.

9. "In the midst of winter, I found there was, within me, an invincible summer" — Camus

PARTLY TRUE. The sentence is his, but it also circulates inside a longer "letter" that he did not write. Gaetani again: "While Camus is not the author of this honeyed letter, he is surely the author of the isolated sentence". It comes from "Retour à Tipasa" ("Return to Tipasa"), an essay in L'Été (Summer), which Gallimard published in 1954. The French, as Gaetani gives it: "Au milieu de l'hiver, j'apprenais enfin qu'il y avait en moi un été invincible." A closer translation is ours: "In the middle of winter, I was at last learning that there was in me an invincible summer."

Our reading

The lines on the page are harder than the posters. Kierkegaard wrote about a leap, but never "leap of faith", and McKinnon and Nientied both reject the slogan as proof that he gave up on reason. Sartre said his hell was also about freedom, and that its circle can be broken. Camus asks what he called the one truly serious philosophical problem so that he can answer it in favour of living. If a line sounds like a greeting card, the page it came from is worth a look.

Sources

Kevin Aho, "Existentialism"; John Lippitt and C. Stephen Evans, "Søren Kierkegaard"; Jack Reynolds and Pierre-Jean Renaudie, "Jean-Paul Sartre"; Ronald Aronson, "Albert Camus", Stanford Encyclopedia of Philosophy.
Douglas Burnham and George Papandreopoulos, "Existentialism"; David Simpson, "Albert Camus"; William McDonald, "Søren Kierkegaard", Internet Encyclopedia of Philosophy.
"Gabriel Marcel", Stanford Encyclopedia of Philosophy.
Alastair McKinnon, "Kierkegaard and 'The Leap of Faith'", Kierkegaardiana 16 (1993).
Mariele Nientied, "Kierkegaard without 'Leap of Faith'", Papers of the 26th International Wittgenstein Symposium (2003).
Giovanni Gaetani, "The Noble Art of Misquoting Camus – From Its Origins to the Internet Era", Journal of Camus Studies (2015).
Oxford Essential Quotations, 4th ed., ed. Susan Ratcliffe (Oxford University Press, online 2016), "Jean-Paul Sartre".
Société des Études camusiennes, "Le Mythe de Sisyphe (1941)" and "L'Été (1954)".
Gallimard, Huis clos, collection À voix haute (audio CD, 2004).

Checked September 2026.

Diogenes and the Cynics: The Famous Stories, Checked

Hacks Vitae — Tue, 22 Sep 2026 00:00:00 GMT

Diogenes of Sinope is the philosopher people remember for his stunts: the lamp in daylight, the home in a jar, the insult to Alexander the Great. Nearly all of it reaches us through anecdotes. The Internet Encyclopedia of Philosophy puts it plainly: "The evidence regarding the Cynics is limited to apothegms, aphorisms, and ancient hearsay; none of the many Cynic texts have survived." Our main source is Diogenes Laertius, Lives of Eminent Philosophers, Book VI, in R. D. Hicks's 1925 translation. Hicks provisionally dates that work to "the earlier decades of the third century" AD, while Diogenes of Sinope lived from about 404 to 323 BC (dates vary), so the gap is more than five centuries. Here are eight famous claims, checked.

1. "Diogenes walked around with a lamp looking for an honest man"

PARTLY TRUE. The lamp is in the source; the word "honest" is not. In Hicks's translation: "He lit a lamp in broad daylight and said, as he went about, 'I am looking for a man.'" The Greek word is anthrōpos, a human being, and there is no adjective with it in the Greek text. The Internet Encyclopedia renders it "I am searching for a human being". We searched Hicks's full Book VI for "honest": it appears three times, and never in the lamp story.

The same joke is told about someone else in an older written text. The Roman poet Phaedrus, who wrote under the early Roman emperors, has the fable-teller Aesop walking through the forum at midday with a lamp and saying "Hominem … quaero", "I am looking for a human being" (our translation). His English translator, Henry Riley, placed this book of the fables "long after the time of Augustus", noting a suggestion that it was written "as late as the reign of Caligula", which is still long before Diogenes Laertius. Riley added: "The same story is told in Diogenes Laertius, of Diogenes the Cynic."

2. "He told Alexander the Great to get out of his sun"

UNVERIFIED. The story is told in several ancient sources, slightly differently each time: by Cicero in his Tusculan Disputations, where Diogenes asks Alexander to move "a little out of the sun" (our translation), and by Plutarch, Arrian and Diogenes Laertius. In Diogenes Laertius, Diogenes is "sunning himself" when Alexander offers him any favour, and he replies: "Stand out of my light." In Plutarch's Life of Alexander, in the English edition edited by Arthur Hugh Clough, he says: "I would have you stand from between me and the sun", and Alexander remarks that "if he were not Alexander, he would choose to be Diogenes."

Whether it happened, nobody can say. Diogenes Laertius records another tidy coincidence worth keeping in mind: a writer named Demetrius claimed "that on the same day on which Alexander died in Babylon Diogenes died in Corinth."

3. "Diogenes lived in a barrel"

PARTLY TRUE. He lived in a container, but it was not a wooden barrel. Hicks translates: "he took for his abode the tub in the Metron". The Greek word is pithos, which the standard Greek dictionary, Liddell–Scott–Jones, defines as a "large wine-jar", "usu. of earthenware". (The same entry notes that wooden ones, "casks", existed too, so a barrel is not impossible, just not the usual meaning.) Diogenes Laertius adds that when a boy broke it, the Athenians "gave the boy a flogging and presented Diogenes with another."

4. "He mocked Plato's 'featherless biped' with a plucked chicken"

UNVERIFIED, but the target of the joke is real. Diogenes Laertius reports: "Plato had defined Man as an animal, biped and featherless, and was applauded. Diogenes plucked a fowl and brought it into the lecture-room with the words, 'Here is Plato's man.'" According to the same passage, the definition was then patched with "having broad nails". Plato's dialogue Statesman does divide animals this way; in Benjamin Jowett's translation: "we should divide bipeds into those which have feathers and those which have not".

5. "Diogenes invented the idea of being a citizen of the world"

PARTLY TRUE. The line is in the source: "Asked where he came from, he said, 'I am a citizen of the world.'" The Greek is a single word, kosmopolitēs, the root of "cosmopolitan". Hicks himself hedged in a footnote: "If this answer is authentic, it apparently shows that the famous term 'cosmopolitan' originated with Diogenes."

The Stanford Encyclopedia calls him "the first philosopher in the West to give perfectly explicit expression to cosmopolitanism", but not the first to have the idea: it notes that Socrates' ideas "accelerated the development of cosmopolitanism", and that in Plato's Protagoras the Sophist Hippias already calls a mixed audience "fellow-citizens – by nature and not by convention". What he meant is argued. Pauline Kleingeld and Eric Brown, in the Stanford Encyclopedia of Philosophy, read it as a refusal: Diogenes "apparently refused to agree that he owed special service to Sinope", so that "'I am a citizen of the cosmos' is a negative claim". Julie Piering, in the Internet Encyclopedia, argues for a positive reading, noting that "Diogenes does not say that he is apolis, that is, without a polis".

One story about him does have hard evidence behind it. Diogenes Laertius says he and his father were involved in defacing the coinage of their home city, Sinope, and Piering writes: "Thanks to numismatic evidence, the adulteration of Sinopean coinage is one event about which there is certainty."

6. "The Cynics were cynical"

PARTLY TRUE. They were not cynics in the modern sense, but the modern sense grew out of them. Their contempt was aimed at hypocrisy: asked where in Greece he saw good men, Diogenes answered, "Good men nowhere, but good boys at Lacedaemon", and he mocked orators who "make a fuss about justice in their speeches, but never practise it". What the modern word dropped is the way of life behind the contempt. Diogenes Laertius says they held that "'Life according to Virtue' is the End to be sought", that they "should live frugally", and that "Wealth and fame and high birth they despise." Asked what was the most beautiful thing in the world, Diogenes answered "Freedom of speech." Piering sums up the core: "virtue is a life lived in accord with nature", and conventions "are often absurd and worthy of ridicule." The Greek name kunikos means "dog-like", and Diogenes embraced it. Asked why he was called a hound, he said: "I fawn on those who give me anything, I yelp at those who refuse, and I set my teeth in rascals."

In English, the word "cynic" named the ancient school first; Merriam-Webster dates it to 1533 and the Online Etymology Dictionary to the 1550s. The meaning "sneering sarcastic person" follows by the 1590s, according to the Online Etymology Dictionary. As Merriam-Webster puts it: "In English, however, cynic and cynical have more to do with distrust of motives than shamelessness." The adjective cynical first meant "resembling Cynic philosophers" and, according to the Online Etymology Dictionary, by the 1660s had "shaded into" the general sense of doubting "the sincerity or value of social usages or personal character or motives".

7. "Antisthenes founded the Cynic school"

DISPUTED. Ancient tradition says so. Diogenes Laertius makes Antisthenes, a companion of Socrates, the teacher of Diogenes, and writes that he "gave the impulse to the indifference of Diogenes, the continence of Crates, and the hardihood of Zeno". Piering writes that Diogenes Laertius identifies him "controversially as the original Cynic": Antisthenes' contemporaries, "such as Plato and Xenophon", do not call him a dog, and the story of Diogenes as his pupil "has been met with suspicion, especially given the likely dates of Diogenes' arrival in Athens and Antisthenes' death". She adds that this silence "lends some credence to the notion that the term kunikos was applied to Antisthenes posthumously". Even the origin of the name "Cynic" is, in her words, "a point of contention".

8. "The Stoics grew out of the Cynics"

PARTLY TRUE. According to the Stanford Encyclopedia, the Stoic school "was founded around 300 BCE by Zeno of Citium", who "also studied under the Cynic Crates and was influenced by the teachings of Plato's Academy and the Megarian School". The Internet Encyclopedia calls the founding of Stoicism "the first and most direct Cynic influence". Diogenes Laertius even says the Cynics held their view of virtue "exactly like the Stoics", and calls Cynicism "a short cut to virtue". The story of how Zeno met Crates, in a bookshop where the bookseller pointed at Crates and said "Follow yonder man", is one Piering thinks "may strike one as too propitious to be historically accurate".

Crates, whom Diogenes Laertius lists among Diogenes's pupils (one of his sources disagreed), is said to have given away his wealth. The philosopher Hipparchia, whom Piering calls "the first known woman Cynic philosopher", insisted on marrying him and then shared his way of life. When a critic mocked her for leaving the loom, she asked whether he supposed she had been "ill advised about myself, if instead of wasting further time upon the loom I spent it in education?"

How Diogenes died is as uncertain as the rest. Diogenes Laertius says he was "nearly ninety years old" and that "there are several different accounts", including a raw octopus and a dog bite. Piering's view: "it is more likely that he died of old age."

Our reading

The popular Diogenes is a grumpy truth-teller with a lantern. The Diogenes of the sources is stranger and harder to pin down: a man who lived as a public argument against convention, whose stories were collected, polished and probably improved over five centuries. Our verdict on most of them is that they are good stories about a real person, not records of what he did.

Sources

Diogenes Laertius, Lives of Eminent Philosophers, Books VI and VII, trans. R. D. Hicks (Loeb Classical Library, 1925), on Wikisource, with Hicks's introduction (archive.org) and the Greek text on Greek Wikisource.
Julie Piering, "Diogenes of Sinope" and "Cynics", Internet Encyclopedia of Philosophy.
Pauline Kleingeld and Eric Brown, "Cosmopolitanism", and Marion Durand, Simon Shogry and Dirk Baltzly, "Stoicism", Stanford Encyclopedia of Philosophy.
Plutarch, Life of Alexander, ed. Arthur Hugh Clough, Project Gutenberg.
Cicero, Tusculan Disputations 5.92, Latin text (The Latin Library).
Arrian, Anabasis of Alexander 7.2, trans. E. J. Chinnock, on Wikisource.
Phaedrus, Fables III.19, Latin text (The Latin Library) and trans. Henry T. Riley, Project Gutenberg.
Plato, Statesman, trans. Benjamin Jowett, Project Gutenberg.
Liddell–Scott–Jones, Greek–English Lexicon, "πίθος", via Logeion (University of Chicago).
Merriam-Webster and Online Etymology Dictionary, "cynic".

Checked September 2026.

Famous Latin Phrases, Checked: 15 Sayings and Where They Really Come From

Hacks Vitae — Tue, 22 Sep 2026 00:00:00 GMT

Latin phrases carry an air of authority because they sound old. Some are exactly as old as they sound; some are paraphrases; a few are much later than the people they are credited to. Here are fifteen, checked against the ancient texts. English after Latin or Greek is our translation unless a translator is named.

1. Veni, vidi, vici — "I came, I saw, I conquered"

TRUE that it belongs to Julius Caesar and his quick victory at Zela, but two ancient writers report it differently. Suetonius says that in his Pontic triumph Caesar "displayed among the show-pieces of the procession an inscription of but three words, 'I came, I saw, I conquered,'" (Rolfe's translation), "not indicating the events of the war, as the others did, but the speed with which it was finished." Plutarch says it was a letter: "to one of his friends at Rome, Amantius, Caesar wrote three words". Both are reports; neither quotes a document by Caesar that survives. Our longer look is at Veni, vidi, vici.

2. Alea iacta est — "The die is cast"

PARTLY TRUE. The ancient sources disagree on the words and the language, and neither has the familiar word order. Suetonius has Caesar at the Rubicon say "Iacta alea est" ("The die is cast", in Rolfe's translation). Plutarch tells it differently: Caesar was "calling out in Greek to the bystanders these words only, 'Let the die be cast,'" (Perrin's translation, Pompey 60), a phrase "with which men usually prelude their plunge into desperate and daring fortunes" (Caesar 32). The same Greek saying turns up in a comedy by Menander, quoted by Athenaeus, where it is about getting married: "δεδογμένον τὸ πρᾶγμ· ἀνερρίφθω κύβος", "the thing's decided; let the die be cast".

3. Et tu, Brute? — "You too, Brutus?"

MYTH as Caesar's last words. The Latin line is Shakespeare's, in Julius Caesar: "Et tu, Brute?—Then fall, Caesar!" The ancient writers give something else. Suetonius says Caesar died "uttering not a word, but merely a groan at the first stroke, though some have written that when Marcus Brutus rushed at him, he said in Greek, 'You too, my child?'" (Rolfe). Cassius Dio prefers the silent version: "This is the truest account". Plutarch has Caesar cry out in Latin at the first blow, "Accursed Casca, what does thou?" (Perrin), but gives him nothing to say to Brutus: "it is said by some writers" that, seeing Brutus's dagger, "he pulled his toga down over his head and sank". The Greek teknon means "child"; the "my" is the translator's.

4. Carpe diem — "Seize the day"

TRUE as Horace, Odes 1.11: "carpe diem, quam minimum credula postero". The rest of the line changes the meaning; we cover it in Carpe diem.

5. Memento mori — "Remember you will die"

PARTLY TRUE. The idea reaches us through Tertullian, Apologeticus 33, where the emperor riding in a triumph is prompted from behind: "Respice post te! Hominem te memento!" — look behind you, remember you are a man. The two-word tag is not in Tertullian's text. We go into the caveats in Older Than You Think.

6. Si vis pacem, para bellum — "If you want peace, prepare for war"

PARTLY TRUE. Vegetius says something very like it, in the preface to book 3 of his military handbook, but not in these words: "Igitur qui desiderat pacem, praeparet bellum", "so let whoever desires peace prepare for war". His reason follows: "Nemo prouocare, nemo audet offendere quem intellegit superiorem esse, si pugnet", "no one dares provoke or offend someone he knows would be the stronger in a fight". The thought is older than Vegetius. Cicero said, in his seventh Philippic, "si pace frui volumus, bellum gerendum est", "if we want to enjoy peace, we must wage war" (7.19), and Nepos has Epaminondas say "paritur pax bello", "peace is produced by war" (Epaminondas 5.4). The short si vis pacem form does not occur in book 3, and we did not find where it first appears.

7. Cogito, ergo sum — "I think, therefore I am"

PARTLY TRUE. Descartes first published it in French, in the Discours de la méthode: "je pense, donc je suis". The Gutenberg edition's note dates the book: "Le Discours sur la méthode parut le 8 juin 1637." The Latin "Ego cogito, ergo sum" is in his Principia philosophiae, which the same edition dates to 1644. His second Meditation, where the argument is most famous, says something slightly different: "Ego sum, ego existo", "I am, I exist". The words ergo sum do not occur in the Second Meditation.

8. Sic semper tyrannis — "Thus always to tyrants"

UNVERIFIED as something Brutus said. It is not in the three ancient accounts of Caesar's death we checked (Suetonius, Plutarch, Cassius Dio). The earliest use we found is American. The Library of Virginia records that the design for the state's great seal was proposed "on July 5, 1776", and "the Latin Sic Semper Tyrannis ('Thus Always to Tyrants') chosen for the obverse, or front, of the seal became the state motto". It is also what John Wilkes Booth shouted when he shot Abraham Lincoln at Ford's Theatre on 14 April 1865. The sources place it differently. Booth's own journal says: "I shouted Sic Semper before I fired." The National Park Service says "numerous witnesses mention Booth brandishing the dagger while shouting 'Sic Semper Tyrannis' and running out to the back of the stage", which is after the shot. The two need not contradict each other. In the sources we checked, only the shout on the stage is reported by others; the one before the shot rests on Booth's own journal, in an entry the Park Service says was likely written between 17 and 20 April, while he was hiding.

9. Quis custodiet ipsos custodes? — "Who will guard the guards themselves?"

PARTLY TRUE. The line is Juvenal's, from Satire 6, but it is about guarding a wife: "'pone seram, cohibe.' sed quis custodiet ipsos custodes? cauta est et ab illis incipit uxor", "'Put on a bolt, keep her in.' But who will guard the guards themselves? The wife is cunning, and starts with them." It is not about government. The text is also uncertain: the edition we used prints the line twice, once inside square brackets.

10. Panem et circenses — "Bread and circuses"

TRUE as Juvenal, Satire 10.81. In G. G. Ramsay's translation, the people "that once bestowed commands, consulships, legions and all else, now meddles no more and longs eagerly for just two things----Bread and Games!" The lead-in names the reason: "Now that no one buys our votes, the public has long since cast off its cares". Ramsay renders circenses as "Games". In our reading, Juvenal's target is the people who stopped caring, not a ruler distracting them.

11. Mens sana in corpore sano — "A sound mind in a sound body"

PARTLY TRUE. The words are Juvenal's, Satire 10.356, but they are the answer to what to pray for, at the end of a poem arguing that most of what people pray for ruins them. Ramsay: "Nevertheless that you may have something to pray for", "you should pray for a sound mind in a sound body; for a stout heart that has no fear of death". Then Juvenal turns it round: "What I commend to you, you can give to yourself". Most of what it asks for is character, not fitness.

12. In vino veritas — "In wine, truth"

PARTLY TRUE. The idea is old; the three-word Latin is not in the Latin text we checked. Pliny the Elder, in a disapproving passage on drunkenness, says "volgoque veritas iam attributa vino est", "and truth has now commonly been attributed to wine", after warning that drinkers let out secrets and words that will get them killed. In Greek, Athenaeus reports the saying "οἶνος καὶ ἀλήθεια", "wine and truth", and Theocritus opens a poem with it: "οἶνος ὦ φίλε παῖ λέγεται καὶ ἀλάθεα", "'wine, dear boy,' they say, 'and truth'". We did not find the earliest text with the exact Latin.

13. Per aspera ad astra — "Through hardships to the stars"

UNVERIFIED as a classical quotation. The phrase is not in the two classical passages it most resembles. In Seneca's Hercules Furens, Megara says "Non est ad astra mollis e terris via", "there is no soft road from the earth to the stars", meaning Hercules' path to becoming a god. In Vergil's Aeneid 9, Apollo praises the boy Ascanius after his first kill in battle: "macte nova virtute, puer, sic itur ad astra", "blessings on your new courage, boy; this is the way to the stars". Neither text has per aspera, and we did not find where the four-word motto first appears.

14. Festina lente — "Make haste slowly"

PARTLY TRUE. Augustus is reported to have repeated it often, but in Greek, σπεῦδε βραδέως; Suetonius lists it among his "favourite sayings" (Rolfe), which does not make it his own coinage. Suetonius puts it in a passage on generalship: "He thought nothing less becoming in a well-trained leader than haste and rashness" (Rolfe). Aulus Gellius reports it at second hand: "we are told that he used to say in conversation, and write in his letters, σπεῦδε βραδέως, that is, 'make haste slowly,'" and Rolfe's note adds: "Hence the common festina lente". The Latin form does not occur in either ancient text.

15. Errare humanum est — "To err is human"

PARTLY TRUE. The thought is classical; the exact tag is not in the two texts behind it. Cicero, Philippic 12.5: "Cuiusvis hominis est errare, nullius nisi insipientis in errore perseverare", "any man can err; only a fool persists in error". He was admitting that he and the Senate had been wrong. Augustine, in a sermon about the Donatists who would not admit they had been wrong, supplies the second half people now add: "Humanum fuit errare, diabolicum est per animositatem in errore manere", "to err was human; to remain in error out of stubbornness is devilish". Neither has the words errare humanum est or perseverare diabolicum, and we did not find the first text that does.

Our reading

Many of these tags were shortened on the way to us, and the cut part is often the point. Vegetius gave a reason for preparing for war, Juvenal's sound mind was a prayer, Cicero's "to err" came with an admission that he had erred. If a Latin phrase sounds like a slogan, look for the sentence it was cut from.

Sources

Suetonius, Divus Iulius and Divus Augustus, Latin and trans. J. C. Rolfe; Plutarch, Caesar and Pompey, trans. Bernadotte Perrin; Cassius Dio, book 44, trans. Earnest Cary; Pliny the Elder, Natural History 14; Aulus Gellius, Attic Nights 10, trans. J. C. Rolfe — all on LacusCurtius (Bill Thayer).
Juvenal, Satires 6 and 10; Horace, Odes 1; Vergil, Aeneid 9; Vegetius, Epitoma rei militaris 3; Cicero, Philippics 7 and 12; Nepos, Epaminondas; Tertullian, Apologeticus — The Latin Library. Juvenal 10, trans. G. G. Ramsay (1918), tertullian.org.
Athenaeus, Deipnosophists 2.6 and 13.8; Theocritus, Idyll 29 — Scaife Viewer (Perseus).
Seneca, Hercules Furens; Descartes, Principia philosophiae and Meditationes II — Latin Wikisource. Descartes, Discours de la méthode, Project Gutenberg #13846.
Shakespeare, Julius Caesar, Project Gutenberg #1522.
Augustine, Sermo 164, augustinus.it.
Library of Virginia, "The Seal of Virginia"; Code of Virginia § 1-500; National Park Service, Ford's Theatre, Booth's journal and "The Assassin" FAQ.

Checked September 2026.

Google Flow Explained, and Whether Those "Product Ad Commands" Are Real

Hacks Vitae — Tue, 22 Sep 2026 00:00:00 GMT

Lists of "Google Flow commands" for product videos circulate: /product reveal ad, /unboxing ad, /lifestyle ad, /macro detail ad, /360 product spin. Here is what Flow is, and whether those commands exist. Everything below comes from Google's own pages, read in September 2026. AI tools change monthly, so check the app before relying on a detail.

What Flow is

Google launched Flow, an AI filmmaking tool, on 20 May 2025. Google called it "the evolution of VideoFX, a Google Labs experiment that launched last year", and at launch it was for "subscribers of our Google AI Pro and Google AI Ultra plans in the U.S." Its own title now reads "AI Creative Studio for Video, Images & Custom Tools", and the help centre describes it as a way to "create cinematic clips, scenes, and stories with consistency".

Models, now. Flow is "powered by various Veo models, in addition to Gemini Omni (available in standard 720p and draft 360p) and Gemini". Its model page lists Veo 3.1 in Lite, Fast and Quality versions, Gemini Omni Flash 1.1, and Nano Banana models for images. At launch Google named Veo, Imagen and Gemini, with Veo 3 in the headline, so older guides describe a different tool.

Who can use it, and what it costs

Age and place. You must be 18 or over and in a supported country. The list is long and includes the US, the UK, India and the Philippines. "VPNs will not enable access in unsupported regions."
Regional limits. "Editing of uploaded videos is currently not supported" in the European Economic Area, Switzerland and the UK, and extra AI credits cannot be bought in Japan ("AI credits are currently unavailable in Japan", under "Purchase AI credits").
Subscription or not: Google's pages disagree. The "Get started" page says to "be subscribed to Google AI Plus, Google AI Pro, or Google AI Ultra". The credits page says: "With or without a subscription, you receive 50 Google Flow credits per day", and that without one you may not be able to generate videos "during peak hours, roughly 2 PM to 5 PM UTC". Flow's own product page sides with the credits page: "Try Google Flow free of charge", with a tier "Without a Google AI Subscription". Two of the three pages say you can start free, so the "Get started" line looks out of date, but check the app.
Credits. On top of the daily 50, "Google AI Pro subscribers receive an extra 1,000 Google Flow credits each month", and "Google AI Ultra $200 subscribers receive an extra 25,000". A short clip in the cheapest mode costs a handful of credits, and Google warns that "model costs are evolving fast". Flow's own page lists Google AI Plus at $4.99, Pro at $19.99 and Ultra at $99.99 or $199.99 a month, and adds that "Prices may vary by market".
Watermarks. Outputs "include invisible SynthID watermarks", and "a visible watermark will be applied automatically if you reside in India, South Korea, or Vietnam."

What Flow offers

Frames to video: set a start frame, or a start and an end frame, to "create transitions, animate images, or generate a video that starts and ends with specific images".
Ingredients: reference images reused across clips, so you can "consistently use the same images for your character and key objects from one clip to the next".
@ references: "In the prompt box, type @ followed by your character's name", and the @ sign followed by "Voice" for a chosen voice. This is the only special prompt-box syntax Google documents for Flow's video tools.
Scenebuilder, to "arrange multiple clips in a sequence", plus extend for Veo clips ("You can currently only extend Veo generated videos").
Flow Agent, a chat assistant for planning and generating; "by default, the Agent asks for your permission before taking any actions that use your AI credits."
Tools: you can "build custom, reusable tools and mini-apps using natural language", share them by link, and "Google Flow also comes with a preset of Tools supplied by Google".

Are the "product ad commands" real?

We read all 23 articles in Google's Flow help centre (15 on Flow, 8 on the separate Flow Music), Flow's own product page and the launch post, and searched them for slash commands and for each command's words.

MYTH that these are built-in Flow commands. Flow's video help pages document no slash commands. On the keyboard-shortcuts page, "/" is only the key that "opens the search" in the grid. The one slash feature in the help centre belongs to Flow Music, Google's music app, where you can "set up custom slash commands" of your own; none is supplied for product videos. "Unboxing", "lifestyle" and "macro" appear nowhere in the help centre, and "360" appears only as a video resolution ("Generate videos at lower cost with Omni 360p"), not a turntable shot.

TRUE that they can be used as prompt ideas. Nothing in Flow's video help pages gives a slash a special meaning in the prompt box, so "/unboxing ad" is read as a very short description, and Google asks for more: "Be specific about the subject, action, environment, lighting, and style." Google's own tip for product shots is practical: "For the best results, provide subject or product references on a plain or segmented background."

UNVERIFIED whether any Tool inside the app carries names like these, but the six Tools that Flow's own page lists as created by Google do not: Type Overlays, Video Resizer, Image Editor, Storyboard Studio, Shader Effects and Mockup. A Tool is also opened from the Tools panel, not typed into the prompt box, so even a Tool with such a name would not be a slash command.

Our reading

The commands are good recipes for product videos, written to look like built-in features. What makes Flow work is the real set of controls: a clear prompt, a clean product image as an ingredient, start and end frames for a spin or a reveal, and Scenebuilder to put the clips together.

Sources

Google, "Meet Flow: AI-powered filmmaking with Veo 3", The Keyword (20 May 2025).
Google Flow Help: Get started; Models and supported features; Manage your credits; Where you can use Google Flow; Create videos; Scenebuilder and extend; Keyboard shortcuts; Flow Agent; Create and manage Tools.
Google Flow product page (flow.google.com, formerly labs.google/flow); Google Flow Music Help: Get started.

Checked September 2026.

Laozi and the Tao Te Ching: 9 Popular Claims, Checked

Hacks Vitae — Tue, 22 Sep 2026 00:00:00 GMT

The Daodejing, often written Tao Te Ching, is a short book: its earliest biography says it ran to "over five thousand words". Its supposed author, Laozi, has become a name that almost any calm-sounding sentence can be attached to. Here are nine claims, checked against the Chinese text on Chinese Wikisource, James Legge's public-domain English translation of 1891, and the Stanford Encyclopedia of Philosophy. English after a Chinese line is our translation unless it is marked as Legge's.

1. "Laozi was a sage of the 6th century BC who wrote the Tao Te Ching"

DISPUTED. That is the tradition, and the evidence for it is thin. The earliest biography, in the historian Sima Qian's Shiji (Records of the Historian), gives him a name, Li Er, and a job as an archivist of the Zhou court. It says he wrote a book "in two parts, on the meaning of the Way and virtue, over five thousand words" before leaving through a border pass. Then it hedges three times with "或曰", "some say", and ends one passage: "世莫知其然否", "no one in the world knows whether this is so" (our translations).

Scholars still disagree. The philosopher Alan Chan, in the Stanford Encyclopedia, reports that "According to some modern scholars, however, Laozi is entirely legendary; there was never a historical Laozi", and that "Few scholars outside China today would subscribe fully to the Shiji report." His own view is that he finds "little reason not to accept the traditional claim that Laozi was a senior contemporary of Confucius", but he adds that the question "no doubt will continue to attract and divide scholarly opinion." On the book, the evidence leans more clearly. The traditional view, Chan writes, is that it was written by Lao Dan in the sixth or early fifth century B.C.E., and "This seems unlikely, however, if it is assumed that the Laozi was written by a single author." He writes that it may be regarded, as D. C. Lau suggests, as "the product of many hands over a long period of time". The name itself is best taken, he writes, to mean "Old (lao) Master (zi)".

2. "The oldest copies match the book we read today"

PARTLY TRUE. Much of the wording survives, but not the book in the shape we read it. Two archaeological finds changed what we know. In 1973, two silk manuscripts of the text were found at Mawangdui, in Hunan province, in a tomb "sealed in 168 B.C.E."; both begin with the second half of the book as we now read it. In late 1993, a tomb at Guodian, in Hubei province, dated "around 300 B.C.E.", yielded bamboo slips of which "about 2,000 characters, match the Laozi". According to Chan, "they remain the oldest extant Laozi texts to date." He writes that the two Mawangdui manuscripts "contain all the chapters that are found in the current Laozi, although the chapters follow a different order in a few places", and that the Guodian material is "On the whole ... consistent with the received text, although the placement or sequence of the chapters is different and there are numerous variants or archaic characters." Some variants matter: where chapter 19 now says "Cut off benevolence (ren), discard rightness (yi)", the Guodian "A" text says "cut off artificiality, discard deceit."

The picture that emerges is of a text that grew over time. Chan writes that sayings "attributed to Laozi were committed to writing probably from the second half of the fifth century B.C.E.", and that "By the middle of the third century B.C.E., the Laozi probably had reached a relatively stable form." Even "The division into 81 chapters reflects numerological interest", and the title Daodejing "appears not to have been widely used until later".

3. "A journey of a thousand miles begins with a single step"

TRUE, with two things lost in translation. It is in chapter 64. In the received Chinese text: "千里之行，始於足下" — literally, "a journey of a thousand li begins beneath the feet" (our translation). "Miles" is not in the Chinese, which counts in li, and "a single step" is James Legge's English: "the journey of a thousand li commenced with a single step."

The line is the third of three images about big things growing from small ones: "The tree which fills the arms grew from the tiniest sprout; the tower of nine storeys rose from a (small) heap of earth". The chapter's opening paragraph ends with its point, in Legge's words: "Action should be taken before a thing has made its appearance; order should be secured before disorder has begun." It is advice about dealing with things while they are small, not a motivational poster about starting.

The older manuscripts are not all the same here. Chan notes that the Guodian slips "give two different versions of what is now part of chapter 64", and in the Mawangdui silk manuscripts, as collated by the editors on Chinese Wikisource, the line reads "百仞之高", "a height of a hundred ren" (our translation), where the later text has the journey. The two manuscripts themselves write the second character differently, so this is an editors' reading.

4. "Wu wei means doing nothing"

MYTH. Wuwei, usually translated "non-action", is central to the book, but it is not idleness. Chapter 37 says: "道常無為，而無不為", "The Way constantly does not act, and yet nothing is left undone" (our translation). Legge renders it: "The Tao in its regular course does nothing (for the sake of doing it), and so there is nothing which it does not do."

Chan puts it plainly: "Wuwei does not mean total inaction." He notes that "some translators have opted for 'non-assertive action,' 'non-coercive action' or 'effortless action,'" and that in the Laozi the idea "seems to be used more broadly as a contrast against any form of action driven by self-serving desire". The book also addresses rulers: it "condemns aggressive measures such as war (ch. 30), cruel punishment (ch. 74), and heavy taxation (ch. 75)". Chan also notes that naturalness and wuwei "are open to interpretation."

5. "When I let go of what I am, I become what I might be"

MYTH as a Laozi quote. It is not in Legge's complete translation, which we searched in full for "let go", "become what" and "might be". A site that checks fake Laozi quotes says "This is NOT a quote from Tao Te Ching" and that the line is "very loosely based on the first lines of chapter 22". That chapter, in Legge's translation, begins: "The partial becomes complete; the crooked, straight; the empty, full; the worn out, new."

6. "Nature does not hurry, yet everything is accomplished"

UNVERIFIED. No passage in Legge's complete translation says this in any recognisable form, and we could not find where the line comes from. Two chapters come near the idea. Chapter 37 says the Tao "does nothing … and so there is nothing which it does not do". Chapter 73, in Legge's words, says "It is the way of Heaven not to strive, and yet it skilfully overcomes" and "Its demonstrations are quiet, and yet its plans are skilful and effective". Both are relatives in thought, not a source for the wording.

7. "If you are depressed you are living in the past"

MYTH as a Laozi quote. It is not in Legge's complete translation. The same site reports that "The first appearance of the quote on Facebook, already accrediting it to Lao Tzu, is from December 19, 2009."

8. "The butterfly dream is a Laozi story"

MYTH. It is from the Zhuangzi, a different book by a different thinker. The philosopher Chad Hansen, in the Stanford Encyclopedia, describes Zhuangzi as a figure of the "late 4th century BC" and the book as "a compilation of his and others' writings". In chapter 2, Zhuang Zhou dreams he is a butterfly and on waking cannot tell whether he was a man dreaming he was a butterfly or a butterfly dreaming he was a man.

Retellings often stop there. The story itself ends with a qualification, in Legge's translation: "But between Kâu and a butterfly there must be a difference. This is a case of what is called the Transformation of Things." Hansen reads it as an illustration of "mild skepticism".

9. "Laozi founded Taoism"

PARTLY TRUE. He is its central figure, but "founder" is too simple. Chan writes: "In many popular accounts, Laozi is described as the 'founder' or 'father' of 'Daoism.' This begs a number of questions and therefore should not be taken uncritically". Chan reports the view that only in the Han dynasty were Laozi, Zhuangzi and others "classified together under the rubric of a distinct Daoist school of thought."

Daoism is also a living religion. Chan writes that Daoism "encompasses both a philosophical tradition and an organized religion", that "The two are deeply intertwined", and that "In religious Daoism, Laozi is revered as a supreme deity." Scholars disagree about how useful the split between "philosophy" and "religion" is. In the Stanford Encyclopedia's entry "Religious Daoism", Fabrizio Pregadio quotes the scholar Kristofer Schipper: "It has become a sinological dogma to distinguish between the so-called Taoist school (Daojia)" and the Daoist religion, a distinction that, Schipper adds, "The successive Daozang [Daoist Canons] never made".

Our reading

The Daodejing itself is stranger and more political than the quotes it gets credited with. It speaks to rulers about war and taxes, its most famous line is about catching problems early, and its central idea is not doing nothing but not forcing things. Several of the sayings shared under its name are not in it. A good test for any "Laozi quote": ask for the chapter number, then read the chapter.

Sources

Daodejing (received text with Wang Bi's commentary), Shiji chapter 63, Zhuangzi chapter 2, and the Mawangdui silk manuscripts, Chinese texts on Chinese Wikisource.
James Legge, trans., The Tao Teh King (1891), Project Gutenberg; and The Texts of Taoism, Sacred Books of the East vol. 39 (1891), Internet Sacred Text Archive.
Alan Chan, "Laozi", Stanford Encyclopedia of Philosophy (revised 2025).
Chad Hansen, "Zhuangzi", and Fabrizio Pregadio, "Religious Daoism", Stanford Encyclopedia of Philosophy.
"Fake Lao Tzu Quotes", taoistic.com (its verdicts are quoted in sections 5 and 7; the checks themselves rest on the Chinese text).

Checked September 2026.

Machiavelli: What The Prince Actually Says, and 8 Claims Checked

Hacks Vitae — Tue, 22 Sep 2026 00:00:00 GMT

Niccolò Machiavelli wrote The Prince after he was dismissed from his post in the Florentine government, and imprisoned and tortured on a false suspicion of conspiracy. According to Cary Nederman in the Stanford Encyclopedia of Philosophy, it was "Penned at the end of 1513 (and perhaps early 1514), but only published posthumously in 1532", by an author who, "among other things, sought to regain his status in Florentine political affairs." It is short, and its most famous lines are often not in it. Here are eight claims, checked against the Italian text on Italian Wikisource and the public-domain English translations by W. K. Marriott (The Prince) and Ninian Hill Thomson (the Discourses).

1. "Machiavelli wrote that the ends justify the means"

MYTH. The phrase is not in The Prince. A search of Marriott's complete translation for "ends justify", "end justifies" and "justify the means" finds nothing. The nearest passage is in chapter 18, and it says something narrower. In Italian: "e nelle azioni di tutti gli uomini, e massime de' Principi, dove non è giudizio a chi reclamare, si guarda al fine." In our translation: "and in the actions of all men, and above all of princes, where there is no court to appeal to, people look to the end (the outcome)."

He goes on: "Facci adunque un Principe conto di vivere e mantenere lo Stato; i mezzi saranno sempre giudicati onorevoli, e da ciascuno lodati", "so let a prince aim at staying alive and keeping his state; the means will always be judged honourable, and praised by everyone" (our translation), and he says why: "perchè il vulgo ne va sempre preso con quello che pare, e con l'evento della cosa", because the common crowd is always taken in by appearances and by the result. It is a description of how people judge rulers, and practical advice built on it, not a moral rule that a good end makes a bad act good. Marriott's English of the key clause, "which it is not prudent to challenge", does not match the Italian, which is why we translate it ourselves here.

The closest he comes in his two main books is in his Discourses on Livy, book 1, chapter 9, about Romulus killing his brother to found Rome. Thomson translates: "For although the act condemn the doer, the end may justify him; and when, as in the case of Romulus, the end is good, it will always excuse the means". But "the means" is Thomson's word. The Italian reads "lo effetto lo scusi", the result excuses him, the founder. The same chapter warns such a founder not to pass his absolute power on.

2. "The idea that results justify actions started with Machiavelli"

MYTH. It is much older, and an ancient version, in Ovid, already rejects it. In Ovid's Heroides 2, Phyllis quotes the saying "exitus acta probat" ("the outcome proves the deeds") and then curses anyone who "ab eventu facta notanda putat", who thinks deeds should be judged by their result (our translations).

3. "It is better to be feared than loved"

PARTLY TRUE. He asks the question in chapter 17: "s'egli è meglio essere amato che temuto, o temuto che amato" (whether it is better to be loved than feared, or feared than loved). His first answer is both: "si vorrebbe essere l'uno e l'altro". Only because that is hard does he say, in Marriott's translation, "it is much safer to be feared than loved, when, of the two, either must be dispensed with." The Italian word is sicuro, "safer", not "better".

Then comes the limit that usually gets dropped: "Nevertheless a prince ought to inspire fear in such a way that, if he does not win love, he avoids hatred", which, he says, will always be possible "as long as he abstains from the property of his citizens and subjects and from their women."

4. "The Prince is a straightforward handbook for tyrants"

DISPUTED. Scholars have read it very differently for centuries. The Stanford Encyclopedia sets out the range. At one end, "The most extreme versions of this reading find Machiavelli to be a 'teacher of evil', in the famous words of Leo Strauss". A milder version, from Quentin Skinner (1978), is that doing what convention calls vicious is a "last best" option for a ruler. At the other end, "Some scholars, such as Garrett Mattingly (1958), have pronounced Machiavelli the supreme satirist", and Mary Dietz (1986) argued that he wanted to "trap" the prince.

The idea that the book exposes rulers rather than serving them is old. Jean-Jacques Rousseau wrote in The Social Contract (1762): "En feignant de donner des leçons aux Rois il en a donné de grandes aux peuples. Le Prince de Machiavel est le livre des républicains." In G. D. H. Cole's translation: "He professed to teach kings; but it was the people he really taught. His Prince is the book of Republicans." What is not in doubt is the setting Nederman describes: the book was "Originally written for presentation to Giuliano de'Medici" by an author who, "among other things, sought to regain his status" in Florentine politics. How the book was meant to be read is another matter: the texts do not settle it, and this page does not.

5. "Machiavelli believed in rule by strongmen"

PARTLY TRUE, and misleading as a summary of his views. His longer work, the Discourses on Livy, which Nederman writes "perhaps most honestly expresses Machiavelli's personal political beliefs and commitments", argues for republics: "Only in a republic, for which Machiavelli expresses a distinct preference, may this goal be attained." One chapter title reads, in Thomson's translation, "That a People is wiser and more constant than a Prince". How The Prince fits with this is itself argued; Nederman notes that the two works lead to conclusions "quite different from—many scholars have said contradictory to—" each other. The strongman half is not invented, though. In the Romulus chapter (Discourses I.9) he concludes that "a ordinare una republica è necessario essere solo", that one man must act alone to found a republic, and adds that what he founds lasts only when it is left "alla cura di molti", in the care of many.

6. "Machiavellian means what Machiavelli taught"

PARTLY TRUE. The Prince does tell a ruler he must be "a great pretender and dissembler" (chapter 18, Marriott). But the word began as an insult, not a summary. According to the Online Etymology Dictionary, "Machiavellian" meant "cunning, deceitful, habitually duplicitous, unscrupulous" by the 1570s and was "A word of abuse in English well before his works were translated ('The Discourses' in 1636, 'The Prince' in 1640)". Merriam-Webster dates its first known use to 1572. His Art of War had reached English earlier, in Peter Whitehorne's translation of 1560, so the point holds for his two political books rather than for everything he wrote.

The English stage used his name for villains. In Christopher Marlowe's play The Jew of Malta, a character called Machiavel speaks the prologue: "I count religion but a childish toy, And hold there is no sin but ignorance." In Shakespeare's Henry VI, Part 3, Richard of Gloucester boasts that he can "set the murderous Machiavel to school." Both are characters in plays, not quotations of Machiavelli. Whether he was himself "a 'Machiavellian' in the sense commonly ascribed to him" is, in Nederman's words, one of the questions scholars still ask. For the psychology term, see our page on dark psychology.

7. "Keep your friends close and your enemies closer"

UNVERIFIED as Machiavelli's. It is not in Marriott's Prince or Thomson's Discourses: a search of both complete texts for "enemies closer" finds nothing. Another line credited to him is missing from both books too: "Never attempt to win by force what can be won by deception". The nearest real passage is Discourses III.40, "That Fraud is fair in War": "he who gets the better of his enemy by fraud, is as much extolled as he who prevails by force" (Thomson). It is limited to war, and he adds that no fraud is glorious "which leads you to break your plighted word". We did not search his other works, so "not in his two main books" is as far as the evidence goes.

8. "Everyone sees what you appear to be"

TRUE. Some of the sharp lines really are his, in Marriott's translation of The Prince:

"Every one sees what you appear to be, few really know what you are" (chapter 18).
"men are so simple, and so subject to present necessities, that he who seeks to deceive will always find someone who will allow himself to be deceived" (chapter 18).
"men ought either to be well treated or crushed, because they can avenge themselves of lighter injuries, of more serious ones they cannot" (chapter 3).
"where the willingness is great the difficulties cannot be great if you will only follow those men to whom I have directed your attention" (chapter 26).

Chapter 18 also opens with this line: "Every one admits how praiseworthy it is in a prince to keep faith, and to live with integrity and not with craft." His advice is to hold to the good where possible, "but, if compelled, then to know how to set about it."

Our reading

The Machiavelli of the quote cards is a cynic who thinks anything goes if it works. The Machiavelli of the texts is harder to pin down: a republican who wrote a guide for princes, whose best-known lines are descriptions of how people judge power, hedged with conditions that usually get cut. Whether he meant it all is a question scholars still argue.

Sources

Niccolò Machiavelli, Il Principe, chapters 17 and 18, and Discorsi sopra la prima Deca di Tito Livio I.9, Italian text on Italian Wikisource.
Niccolò Machiavelli, The Prince, trans. W. K. Marriott, Project Gutenberg.
Niccolò Machiavelli, Discourses on the First Decade of Titus Livius, trans. Ninian Hill Thomson, Project Gutenberg.
Cary Nederman, "Niccolò Machiavelli", Stanford Encyclopedia of Philosophy (revised 2023).
Ovid, Heroides 2, Latin text, The Latin Library.
Jean-Jacques Rousseau, Du contrat social III.6 (1762 edition, French Wikisource) and The Social Contract, trans. G. D. H. Cole, Project Gutenberg.
Niccolò Machiavelli, The Arte of Warre, trans. Peter Whitehorne (1560), Internet Archive.
Christopher Marlowe, The Jew of Malta, and William Shakespeare, Henry VI, Part 3, Project Gutenberg.
Online Etymology Dictionary and Merriam-Webster, "Machiavellian".

Checked September 2026.

Mental Models Cheat Sheet: 52 Thinking Tools, Where They Come From and What the Original Says

Hacks Vitae — Tue, 22 Sep 2026 00:00:00 GMT

A quick reference to the mental models checked on this site. Each card gives the popular version, a verdict and the key fact from the source. Open a card for the detail; the full evidence is on the five longer pages linked below.

How to read the sheet

The verdict is about the claim as it is usually passed around, not about whether the tool is useful. A partly true tool may be worth using exactly as it is; it usually means the popular version dropped a condition or the original author's caveat. Disputed means researchers disagree; unverified means we could not trace the claim to a source.

Sources

Each of those pages lists its primary sources in full.

Checked September 2026.

Mental Models, Checked: 15 Thinking Tools and Where They Really Come From

Hacks Vitae — Tue, 22 Sep 2026 00:00:00 GMT

A mental model is a small rule of thumb for thinking: invert the problem, look for the second effect, ask what you gave up. They are useful, and most of them are passed around without their origins, which is where the useful detail usually is. Here are fifteen, checked against the texts they came from. English after French or German is our translation.

1. "Charlie Munger invented mental models"

PARTLY TRUE. The phrase has two meanings. In psychology, according to the Mental Models Global Laboratory, a group of researchers who study how people reason with mental models, "the Scottish psychologist Kenneth Craik (1943) proposed a similar idea; he believed that the mind constructs 'small-scale models' of reality", but "cognitive scientists took for granted that people reasoned based on the laws of symbolic logic until Phil Johnson-Laird published the book Mental Models in 1983". In Johnson-Laird's own summary, "reasoning is a simulation of the world fleshed out with our knowledge".

The self-help sense is Munger's. In a talk at the USC business school on 14 April 1994 he said: "You've got to hang experience on a latticework of models in your head", and "80 or 90 important models will carry about 90 percent of the freight in making you a worldly-wise person." The exact phrase "latticework of mental models" comes in a later talk.

2. Circle of competence

PARTLY TRUE as Warren Buffett's alone. It is a Berkshire idea, and the earliest use we found is Charlie Munger's, in the same USC talk of 14 April 1994: "Every person is going to have a circle of competence. And it's going to be very hard to enlarge that circle." He applied it to aptitudes and careers: "you've got to play within your own circle of competence." Buffett set it out for judging businesses in his 1996 letter to Berkshire Hathaway shareholders: "The size of that circle is not very important; knowing its boundaries, however, is vital." The point people drop is that second half: the size matters little, the edge matters a lot.

3. Chesterton's fence

PARTLY TRUE. The popular version is "never remove a fence until you know why it was put up". G. K. Chesterton's image, in "The Drift from Domesticity" in The Thing (first published 1929), is "a fence or gate erected across a road". A reformer says: "I don't see the use of this; let us clear it away." The reply: "If you don't see the use of it, I certainly won't let you clear it away. Go away and think. Then, when you can come back and tell me that you do see the use of it, I may allow you to destroy it." The part usually left out comes a paragraph later: a reformer who knows what an institution was for "may really be able to say that they were bad purposes, or that they have since become bad purposes, or that they are purposes which are no longer served." It is a rule about understanding before you change something, not a rule against change. We did not find who first called it "Chesterton's fence".

4. "The map is not the territory"

TRUE as Alfred Korzybski's, from a paper he gave in New Orleans on 28 December 1931 and his book Science and Sanity (1933). His fuller sentence is not a warning that maps are useless: "A map is not the territory it represents, but, if correct, it has a similar structure to the territory, which accounts for its usefulness." The joke about a perfect map is older. In Lewis Carroll's Sylvie and Bruno Concluded (1893), a character boasts, "We actually made a map of the country, on the scale of a mile to the mile!", and then: "So we now use the country itself, as its own map, and I assure you it does nearly as well."

5. Second-order thinking

PARTLY TRUE as an idea with an old source; the label is modern and we found no single origin for it. Frédéric Bastiat's pamphlet Ce qu'on voit et ce qu'on ne voit pas ("What is seen and what is not seen"), published in July 1850, states it for economics: "Dans la sphère économique, un acte, une habitude, une institution, une loi n'engendrent pas seulement un effet, mais une série d'effets" — in the economic sphere, an act, a habit, an institution or a law produces not one effect but a series of effects. Only the first is seen; "les autres ne se déroulent que successivement, on ne les voit pas; heureux si on les prévoit" — the others unfold one after another, unseen; we are lucky if we foresee them.

6. Opportunity cost

PARTLY TRUE as "Friedrich von Wieser coined it". The English term is in David I. Green, "Pain-Cost and Opportunity-Cost", Quarterly Journal of Economics (January 1894): "Such consideration gives rise to the conception of opportunity-cost", the cost of "the opportunities foregone in accepting a certain line of action". Wieser's claim is to the idea: the History of Economic Thought website says the doctrine "was only explicitly introduced as an all-encompassing theory of cost in a seminar paper by Friedrich von Wieser (1876) and expounded in his later books", and credits Green's 1894 paper with popularising it in English. The idea is older still. Bastiat's broken window (1850): "puisque notre bourgeois a dépensé six francs à une chose, il ne pourra plus les dépenser à une autre" — since our townsman has spent six francs on one thing, he cannot spend them on another.

7. Inversion: "Invert, always invert"

UNVERIFIED as a saying of the mathematician Carl Jacobi. The earliest printed attribution we found is Edward B. Van Vleck, in the Bulletin of the American Mathematical Society for October 1916: "The great mathematician Jacobi is said to have inculcated upon his students the dictum: Man muss immer umkehren." Van Vleck's own gloss is "One must always seek a converse, turn a thought the other end to." Michael Molinsky, in a column republished by the Mathematical Association of America in 2024, found that of the modern sources repeating it, "none of them provide a primary source or any original context for the quotations", and raised "the possibility that it also may be misattributed".

Munger used it as a way of avoiding failure. In his 1986 commencement speech at the Harvard School in Los Angeles he said "many hard problems are best solved only when they are addressed backward", and quoted the rustic who said: "I wish I knew where I was going to die, and then I'd never go there."

8. Occam's razor

PARTLY TRUE. The idea is Ockham's; the famous wording is not. The Stanford Encyclopedia of Philosophy says of "must not multiply entities beyond necessity": "Although the sentiment is certainly Ockham's, this particular formulation is nowhere to be found in his texts." His own fullest version names what can justify positing more: "reason (or reasoning from self-evident premises), experience, and infallible religious authority". It is a rule about needing grounds, not "the simplest explanation is usually right". We trace the popular wording in Older Than You Think, Part 2.

9. Hanlon's razor: "Never attribute to malice that which is adequately explained by stupidity"

PARTLY TRUE as Robert Heinlein's. The earliest close match known to Quote Investigator, with the name, is in Arthur Bloch's 1980 Murphy's Law Book Two. Heinlein's 1941 story "Logic of Empire" has a close ancestor: "You have attributed conditions to villainy that simply result from stupidity." But in the story it is a diagnosis of "the devil theory", and the speaker goes on to blame "an antiquated financial structure" and plain stupidity rather than villains: "You think bankers are scoundrels. They are not".

10. The Pareto principle (80/20)

PARTLY TRUE. Vilfredo Pareto studied the unequal distribution of income and fitted a curve to it (Cours d'économie politique, 1896–97). The universal "vital few and trivial many" principle is Joseph Juran's, and Juran later wrote: "I was forced to confess that I had mistakenly applied the wrong name to the principle." More on this in Older Than You Think, Part 2.

11. Parkinson's law: "Work expands to fill the time available"

PARTLY TRUE. The sentence opens C. Northcote Parkinson's essay in The Economist of 19 November 1955, but as "a commonplace observation that work expands so as to fill the time available for its completion", not as his law. The law he goes on to propose is about staff: "the number of the officials and the quantity of the work to be done are not related to each other at all." The essay is written in a deadpan, mock-scientific style, and ends: "It is not the business of the botanist to eradicate the weeds."

12. Goodhart's law: "When a measure becomes a target, it ceases to be a good measure"

PARTLY TRUE as Charles Goodhart's words. That sentence is the anthropologist Marilyn Strathern's, in a 1997 paper on university audit, where she writes that another scholar "describes this as 'Goodhart's law', after the latter's observation on instruments for monetary control". Goodhart's own 1975 wording, as quoted by Takeda and Ueda, was narrower: "any observed statistical regularity will tend to collapse once pressure is placed upon it for control purposes". We could not open Goodhart's original. See also AI benchmarks explained.

13. The Dunning-Kruger effect

DISPUTED in size, and the famous chart is a MYTH. Justin Kruger and David Dunning (1999) found that people in the bottom quarter on tests of humour, grammar and logic "grossly overestimated their test performance and ability": "Although their test scores put them in the 12th percentile, they estimated themselves to be in the 62nd." They also found the top quarter "tended to underestimate their ability", and they raised the statistical objection themselves, concluding the effect was "more psychological than artifactual". The peak-and-valley curve people share is not in the paper. Critics argue the effect is partly a product of how it is graphed. Nuhfer and colleagues (2016) showed that "graphical conventions common in the self-assessment literature introduce artifacts", and in a study of self-rated intelligence Gignac and Zajenkowski (2020) concluded that "the magnitude of the effect may be much smaller than reported previously". The site's cognitive biases page covers the replication in more detail.

14. First principles

TRUE as Aristotle's. In the Physics he writes that "we do not think that we know a thing until we are acquainted with its primary conditions or first principles, and have carried our analysis as far as its simplest elements." For him first principles are starting points you grasp rather than prove, not a brainstorming technique. A well-known modern example is Elon Musk's, in a 2012 Wired interview: "physics teaches you to reason from first principles rather than by analogy." We go further in first principles thinking.

15. Base rates

DISPUTED in how often people ignore them. The classic experiment is Daniel Kahneman and Amos Tversky's, summarised in their 1974 Science paper: people were told a description was drawn from a group of 100, "30 engineers and 70 lawyers" in one condition and the reverse in another. With no other information, "the subjects used prior probabilities correctly". Given short personality descriptions, "in a sharp violation of Bayes' rule, the subjects in the two conditions produced essentially the same probability judgments". Jonathan Koehler (1996) is one prominent challenge: "We have been oversold on the base rate fallacy", and base rates "play a relatively larger role in tasks where base rates are implicitly learned or can be represented in frequentist terms". The disagreement is over the lab format itself. Tversky and Kahneman read their result as base rates being "effectively ignored" once a description was given; Koehler's review of the literature, "including the famous lawyer–engineer problem", concludes that base rates "are almost always used", and that they count for more when they are learned from experience or given as frequencies. We leave it as disputed: both are reviews of experiments, and they weigh the same studies differently. More in Numbers that mislead.

Our reading

Most of these models came from someone making a careful, limited point: about maps with the right structure, fences you have understood, measures used for control. The versions that travel are shorter and more confident. When a mental model sounds like a law of nature, the page it came from is worth finding.

Sources

The Mental Models Global Laboratory, "What are mental models?"; Philip N. Johnson-Laird, "Mental models and human reasoning", PNAS 107 (2010).
Charles T. Munger, Poor Charlie's Almanack (Stripe Press): Harvard School speech (13 June 1986) and USC talk (14 April 1994).
Warren Buffett, Berkshire Hathaway letter to shareholders, 1996.
G. K. Chesterton, The Thing (1929), "The Drift from Domesticity".
Alfred Korzybski, Science and Sanity (1933); Lewis Carroll, Sylvie and Bruno Concluded (1893), Project Gutenberg.
Frédéric Bastiat, Ce qu'on voit et ce qu'on ne voit pas (1850), French Wikisource.
David I. Green, "Pain-Cost and Opportunity-Cost", Quarterly Journal of Economics 8 (1894); "The Opportunity Cost Doctrine", History of Economic Thought website.
Edward B. Van Vleck, "Current Tendencies of Mathematical Research", Bulletin of the AMS 23 (1916); Michael Molinsky, "Quotations in Context: Jacobi", Convergence (MAA, 2024).
Paul Vincent Spade, Claude Panaccio and Jenny Pelletier, "William of Ockham", Stanford Encyclopedia of Philosophy.
Quote Investigator, "Never Attribute to Malice…" (2016, updated 2025); Robert Heinlein, "Logic of Empire", Astounding Science-Fiction (March 1941).
Vilfredo Pareto, Cours d'économie politique (1896–97); Joseph M. Juran, "The Non-Pareto Principle; Mea Culpa" (1974).
"Parkinson's Law", The Economist (19 November 1955).
Marilyn Strathern, "'Improving ratings': audit in the British University system", European Review 5 (1997); Y. Takeda and A. Ueda, "Uncovering the Goodhart's Law", CIRJE workshop paper.
J. Kruger and D. Dunning, Journal of Personality and Social Psychology 77 (1999); E. Nuhfer et al. (2016); G. E. Gignac and M. Zajenkowski, Intelligence 80 (2020).
Aristotle, Physics I.1, trans. Hardie and Gaye; C. Anderson, "Elon Musk's Mission to Mars", Wired (October 2012).
A. Tversky and D. Kahneman, "Judgment under Uncertainty: Heuristics and Biases", Science 185 (1974); J. J. Koehler, "The base rate fallacy reconsidered", Behavioral and Brain Sciences (1996).

Checked September 2026.

Mental Models from History: 12 Sayings Checked Against the Sources

Hacks Vitae — Tue, 22 Sep 2026 00:00:00 GMT

Plenty of the thinking tools people quote are named after a battle, a ruler or a scandal. The name travels; the source usually does not. Here are twelve, checked against the ancient texts, letters and archives they came from. English after Latin, Greek, Spanish, French or Chinese is our translation unless a translator is named. The companion pages are Mental models, checked and Mental models from philosophy.

1. A Pyrrhic victory

PARTLY TRUE. The remark is reported, not recorded. Plutarch, in his Life of Pyrrhus, writes that "we are told that Pyrrhus said to one who was congratulating him on his victory, 'If we are victorious in one more battle with the Romans, we shall be utterly ruined.'" (tr. Perrin). Diodorus gives a different wording: "If I win a victory in one more battle with the Romans, I shall not have left a single soldier of those who crossed over with me". Plutarch's two sources disagree about the battle itself, and Diodorus says the proverb for such a win was not "Pyrrhic" at all: "A 'Cadmean victory' is a proverbial expression."

Both writers explain it the same way, and it is about replacement more than the dead. Pyrrhus "had lost a great part of the forces with which he came", while the Romans, in Diodorus's words, "though defeated, were in no way humbled, since their dominion was so great". The tool: count what you can replace, not only what you lost.

2. Crossing the Rubicon

PARTLY TRUE as the classic snap point of no return. Caesar did cross; Suetonius has him overtake "his cohorts at the river Rubicon, which was the boundary of his province". But Caesar's own Civil War does not mention the river: book 1 simply says he "set out for Ariminum with that legion" (Ariminum cum ea legione proficiscitur).

The famous scene comes from later writers, and it is a scene of hesitation. Plutarch says Caesar "communed with himself a long time in silence as his resolution wavered back and forth", talked it over with friends "among whom was Asinius Pollio", and crossed at last "as if abandoning calculation and casting himself upon the future". Suetonius needs a supernatural sign to tip him: "a being of wondrous stature and beauty, who sat and played upon a reed". For the words he is supposed to have said, see Latin phrases, checked.

3. The sword of Damocles

PARTLY TRUE as "a danger hanging over you". The story is in Cicero's Tusculan Disputations (5.61–62), and it is about the tyrant, not the guest. Damocles, "one of his flatterers", praises the wealth of Dionysius of Syracuse; Dionysius seats him at a feast and "ordered a bright sword to be let down from the ceiling, suspended by a single horse-hair" (tr. Yonge; the Latin says only saeta equina, a horse-hair). Damocles at last begs "to go, for that now he had no desire to be happy".

Cicero spells out the point: "Does not Dionysius, then, seem to have declared there can be no happiness for one who is under constant apprehensions?" The fear on display is the ruler's own. Cicero names no earlier source for the story. The tool: before envying a position, ask what its holder has to live with.

4. Cutting the Gordian knot

DISPUTED that Alexander cut it with his sword. The ancient accounts split. Arrian's Anabasis (2.3, tr. Chinnock) says some report that "he struck the cord with his sword and cut it through", but "Aristobulus says that he pulled out the pin of the wagon-pole, which was a wooden peg driven right through it, holding the cord together." Aristobulus is one of Arrian's main sources, chosen "because he served under king Alexander in his expedition". Arrian then declines to choose: "How Alexander performed the feat in connection with this cord, I cannot affirm with confidence."

Plutarch says "most writers" tell the sword version, and gives the pin version to Aristobulus too. The prize differs by source: to be "the ruler of Asia" in Arrian, "king of the whole world" in Plutarch. Either way the solution is a sideways one; only the swordstroke is uncertain.

5. Fabian strategy

TRUE that Fabius Maximus avoided pitched battle with Hannibal and wore him down. Polybius says he made "the safety of the army under his command his first and chief aim, adhered steadfastly to this purpose". It was unpopular at the time: "At first, it is true, he was despised for this, and gave people occasion to say that he was playing the coward", and in Plutarch "the soldiers railed at Fabius and scornfully called him Hannibal's pedagogue". The poet Ennius praised him in a line Cicero quotes: "Unus homo nobis cunctando restituit rem" — one man, by delaying, restored the state for us.

The modern version, wait and then strike hard, comes from somewhere else. The Fabian Society, named in January 1884 "in allusion to the victorious policy of Fabius Cunctator", printed on its first tract a motto saying that "when the time comes you must strike hard, as Fabius did". Its own historian, Edward Pease, wrote that the mottoes, "though printed as quotations, are not to be discovered in any history", and noted: "It has been pointed out by Mr. H.G. Wells, and by others before him, that Fabius never did strike hard".

6. Burning the boats

MYTH that Hernán Cortés burned his ships. In his second letter to the emperor he says he decided, "on the pretext that they were unseaworthy", to have "the said ships beached" (tr. MacNutt), so that "everybody lost hope of ever leaving the country". Bernal Díaz, who was there, says the soldiers proposed running the vessels ashore "in order at once to cut off all possibility of further mutiny", adds that "the pilots and sailors would be of greater use to us" as soldiers, and suspects "the idea of destroying our vessels originated with Cortes himself". Díaz's soldiers even compare themselves to Caesar: "as the die had been cast, and we, with Caesar, when he had passed the Rubicon, had now no choice left".

UNVERIFIED that Tariq ibn Ziyad burned his ships on landing in Spain. The story rests on later reports. Pascual de Gayangos, who translated al-Maqqari's history, notes that Tariq "like Cortés, is reported to have set fire to his fleet", and says the claim "may reasonably be questioned".

TRUE for the oldest of the three versions here. Sima Qian's Shiji (chapter 7) says Xiang Yu led his army across a river, and then 皆沈船，破釜甑 — they sank all the boats and smashed the cooking pots — carrying three days' food, "to show the soldiers they must die, with no thought of returning". The Art of War has the same image for a general deep in enemy country: "He burns his boats and breaks his cooking-pots" (chapter 11, tr. Giles). The tool is about commitment, but the case with the fullest record is a warning too: Cortés wrote that the aim was to stop his own men leaving, and Díaz says he arranged it so that, if the owners wanted paying, "he could throw the blame on us".

7. Hobson's choice

PARTLY TRUE. Thomas Hobson was a real "carrier of Cambridge", who died there in January 1631, and Milton wrote comic epitaphs for him. The stable story was printed eighty-one years later, in a letter in The Spectator No. 509 of 14 October 1712. It says Hobson kept "forty good Cattle" and made each customer "take the Horse which stood next to the Stable-Door; so that every Customer was alike well served according to his Chance".

Two details in the letter are worth knowing. It calls him "Tobias Hobson". And it says the phrase was already being used, "by vulgar Error", for a man "reduced to an Extremity"; its own sense was fair turns from plenty. It also glosses the phrase as a choice "forced upon you". The Dictionary of National Biography is cautious: the stable rule "is said to have given rise to the proverb", and the claim that he was the first to hire out horses is "asserted very doubtfully".

8. The cobra effect

UNVERIFIED that British officials in Delhi paid a bounty for dead cobras and people bred cobras to claim it. The phrase is the title of a 2001 German book on economic policy by Horst Siebert, Der Kobra-Effekt. No one we could find has produced a colonial record, a date or an official for the Delhi story. We did not search Indian colonial records on snake bounties in general.

TRUE for the documented case, which is about rats. The historian Michael G. Vann, who found the records in the French colonial archives, says that the city's rat records run from April to July 1902, and that officials in Hanoi put out a bounty "paying a few cents for each rat tail". People responded by "cutting off their tails to collect the bounty, and then letting them go free to breed more rats", and officials "also found a smuggling network bringing rats" into the city. It is the incentive version of Goodhart's law: reward the measure, and people produce the measure.

9. The Thucydides trap

DISPUTED as a reading of Thucydides. The phrase is Graham Allison's, not Thucydides'. In The Atlantic in September 2015 he calls it "the Greek historian's metaphor" and reports that a team of his at Harvard's Belfer Center found that "in 12 of 16 cases over the past 500 years, the result was war", while adding: "War, however, is not inevitable."

The sentence behind it is Thucydides 1.23.6. Richard Crawley's translation gives "made war inevitable"; the Greek verb, anankasai, means forced or compelled, and Thucydides offers the growth of Athens and Sparta's fear as the "truest cause" but the one "least spoken of". Not every classicist reads the passage the way political scientists do: a 2025 article in Classical Antiquity argues that Thucydides "suggests that the war might have started through a process of self-fulfilling prophecy", and that this "challenges not only the idea of the" trap. The case count is Allison's team's own. This page gives no forecast.

10. The Lindy effect

PARTLY TRUE as "the longer something has lasted, the longer it will last". The name comes from Lindy's, a delicatessen, and a 1964 New Republic piece by Albert Goldman listed in the magazine's index as "Lindy's law", which, as Toby Ord summarises it, was "about how the future career length of a comedian might be predicted from their past exposure". In 1982 Benoît Mandelbrot called his rule "Lindy's Law" in honour of Goldman's anecdote, and "the idea was developed by Nassim Taleb (2012) in his book, Antifragile", as Ord puts it. Ord adds: "It is not clear whether Goldman, or the denizens of Lindy's Deli he describes, were really describing the same principle as Mandelbrot."

It is a pattern for some things, not a law for all: Ord notes that "the empirical distribution of human lifespans doesn't exhibit even a broad Lindy effect". A person who has lived ninety years does not have ninety more.

11. Cincinnatus

PARTLY TRUE as "the farmer who left his plough, saved Rome and gave up power". The story is in Livy (book 3), who tells it as a lesson for "those who despise everything human beside riches". Livy does not know what Cincinnatus was doing when the envoys arrived: "whether digging a ditch and leaning on his spade, or ploughing", only that he was, "as is agreed", at farm work. He farmed four iugera across the Tiber, and Livy has already told us why he was poor: money for his son's skipped bail was, in Spillan's translation, "rigorously exacted" from the father, who sold everything and lived "in a solitary cottage on the other side of the Tiber".

He did not resign the moment the war was won. Livy says he "would have laid down his office, had not the assembly for the trial of Marcus Volscius, the false witness, detained him"; Volscius had testified against his son. "Quintius laid down his dictatorship on the sixteenth day, having received it for six months."

12. Potemkin villages

DISPUTED that Potemkin built fake village fronts to fool Catherine II on her journey to the Crimea. The story was already going round during the journey: the Prince de Ligne, who was on it, called the tale of cardboard villages carried along the route "le conte ridicule". It was later printed in a German memoir of Potemkin (English second edition, 1813), which hedges the fake fronts: "It has even been asserted that the most distant buildings were unfinished, and had merely a front." The same book does state as fact that peasants were moved "from one spot to the other (frequently in the night)". It rejects a related story about herds moved at night ("The shores of the Dnieper are, besides, rich enough in cattle not to need such a resource") and concludes that "Envy, which fastens itself upon great men, has magnified what was but show, and diminished what was real".

Two people on the journey describe something in between. The Prince de Ligne wrote that there were often "des villes sans rues, des rues sans maisons et des maisons sans toit, portes ni fenêtres" — towns without streets, streets without houses, and houses without roofs, doors or windows. The French ambassador, the comte de Ségur, says towns and villages were "tellement ornés et déguisés par des arcs de triomphe" — so adorned and disguised with triumphal arches — that the illusion turned them "into superb cities". That is a staged tour of real, partly unfinished settlements. We did not read modern scholarship on the question, so the verdict stays open.

Our reading

The ancient writers were usually more cautious than the sayings named after their stories: "we are told", "I cannot affirm with confidence", "whether digging a ditch… or ploughing". The versions that circulate keep the drama and drop the hedge. When a thinking tool comes with a famous story attached, the story is worth checking before the tool is.

Sources

Plutarch, Lives of Pyrrhus, Caesar, Alexander and Fabius Maximus, tr. B. Perrin (Loeb); Diodorus Siculus, book 22 (Loeb).
Suetonius, Divus Julius, tr. J. C. Rolfe (Loeb); Caesar, Bellum Civile 1 (Latin).
Cicero, Tusculan Disputations 5, Latin and tr. C. D. Yonge; Cicero, De Senectute 10 (quoting Ennius).
Arrian, Anabasis of Alexander 2.3, tr. E. J. Chinnock.
Polybius, Histories 3.89 (Loeb); E. R. Pease, The History of the Fabian Society.
Hernán Cortés, second letter, in Letters of Cortes, tr. F. A. MacNutt; Bernal Díaz del Castillo, Memoirs, tr. J. I. Lockhart; al-Maqqari, tr. P. de Gayangos, The History of the Mohammedan Dynasties in Spain; Sima Qian, Shiji 7.
The Spectator No. 509; T. Cooper, "Hobson, Thomas", Dictionary of National Biography.
Deutsche Nationalbibliothek catalogue, H. Siebert, Der Kobra-Effekt; M. G. Vann, French Colonial History 4 (2003); I. Franceschini, "The Great Hanoi Rat Hunt: A Conversation with Michael G. Vann", Made in China Journal.
Thucydides 1.23.6, Greek text and tr. R. Crawley; G. Allison, The Atlantic, 24 September 2015; D. Sutton, "Thucydides' Ambiguous Trap", Classical Antiquity 44 (2025).
The New Republic index, vol. 150; T. Ord, "The Lindy Effect", arXiv (2023).
Livy, Ab Urbe Condita 3, Latin and tr. D. Spillan.
Memoirs of Prince Potemkin, translated from the German, 2nd ed. (1813); comte de Ségur, Mémoires, vol. 2, with letters of the Prince de Ligne.

Checked September 2026.

Mental Models from Philosophy: 11 Thinking Tools Checked Against the Texts

Hacks Vitae — Tue, 22 Sep 2026 00:00:00 GMT

Many of the thinking tools passed around online are borrowed from philosophy, and they usually travel without the passage they came from. Here are eleven, checked against the texts. English after French or Latin is our translation. The companion page, Mental models, checked, covers Occam's razor, first principles and thirteen others.

1. Hume's is–ought gap

PARTLY TRUE as "Hume proved you can't get an ought from an is". The passage is real: one paragraph at the end of the first section of Book 3 of A Treatise of Human Nature, which he introduces as "an observation, which may, perhaps, be found of some importance". He notices that moral writers move from ordinary statements to ones where "I meet with no proposition that is not connected with an ought, or an ought not", and he asks that the step "should be observed and explained; and at the same time that a reason should be given". He does not say the step is impossible; he says it seems "altogether inconceivable", and that noticing it "would subvert all the vulgar systems of morality".

What the paragraph commits him to is disputed. The Stanford Encyclopedia of Philosophy says: "Few passages in Hume's work have generated more interpretive controversy." The strong reading is "the dominant twentieth-century interpretation", and philosophers "including R. M. Hare" call it "Hume's Law". Some readers take it as support for his argument that moral properties are not discernible by demonstrative reason. Others read it as a point about how we first discover virtue and vice, and "point out that Hume himself makes such inferences frequently in his writings." We leave that open. The tool that survives on every reading is the modest one: when an argument slides from what is to what ought to be, ask for the missing step.

2. Buridan's ass

MYTH that John Buridan wrote it. The Stanford Encyclopedia describes the example, a donkey that "starves to death because it has no reason to choose between two equidistant and equally tempting piles of hay", and says: "This particular example is nowhere to be found in Buridan's writings". Its best explanation of the name is that "it originated as a parody of his account of free choice by later critics". The same entry notes a use of the name in 1597, and Spinoza uses it in the Ethics: "what will happen if the incentives to action are equally balanced, as in the case of Buridan's ass?"

The idea is older. Aristotle, reporting Anaximander's argument that the earth stays put "because of its indifference", compares it to a man "exceedingly hungry and thirsty, and both equally, yet being equidistant from food and drink", and calls the view "ingenious but not true" (On the Heavens II.13). Al-Ghazali, as Averroes quotes him, uses two similar dates to argue the opposite moral: a man who wants them "will take one of them". His point, made in defence of the divine will, is that a will can pick between two things exactly alike. The useful point for us: a perfect tie is rarely the real problem; the cost of not choosing is.

3. Pascal's wager

PARTLY TRUE as "Pascal said believe in God just in case". It is in the Pensées, fragment 233 in Brunschvicg's numbering, headed "Infinite—nothing". Its premise is that "Reason can decide nothing here", so it is not an argument that God exists; it is an argument about what to do when reason cannot settle the question: "you must wager. It is not optional. You are embarked." To the reader who replies "I am so made that I cannot believe", Pascal suggests working not "by increase of proofs of God, but by the abatement of your passions".

The Stanford Encyclopedia finds several wager-style arguments in the one fragment and says "it is only the third of these that is traditionally referred to as 'Pascal's Wager'". It quotes Ian Hacking calling the wager "the first well-understood contribution to decision theory", and gives most attention to "the many Gods objection": if reason cannot decide, other options may be live too. The tool is about structure: compare what each outcome is worth, not just how likely it is, and check that you have listed every option. On whether anyone should wager, this page takes no view.

4. The veil of ignorance

PARTLY TRUE as "Rawls invented it". The phrase is John Rawls's, in A Theory of Justice (1971), where "the principles of justice are chosen behind a 'veil of ignorance'". Behind it you do not know your place in society, and Rawls "even assumes that the parties do not know their values or 'conceptions of the good,' their religious or philosophical convictions".

The device is older. John Harsanyi, "who discussed this problem prior to Rawls", used it in papers of 1953 and 1955: an impartial chooser should decide "as if she had an equal chance of becoming anyone". He drew the opposite conclusion: "Harsanyi (1953) considers this to be an argument in favor of utilitarianism." The difference is how much the chooser knows. Harsanyi assumed the odds of each position are equal; "Rawls relied heavily on the assumption that none of the participants knows anything at all about the probability". Whether the veil leads to Rawls's principles is still argued. The tool: design the rule as if you did not know which seat you would get, and notice that what you assume you know changes the answer.

5. The principle of charity

PARTLY TRUE as "read your opponent's argument in its strongest form". That is a looser use than the original. The principle began as a rule for interpreting what people mean. W. V. O. Quine, in Word and Object (1960), credits the name to N. L. Wilson's 1959 paper "Substances without Substrata", which picks out the thing a speaker is talking about as the one that "will make the largest possible number of... statements true". Quine's own reason: "silliness, beyond a certain point, is less likely than bad translation". Donald Davidson made it central to understanding anyone at all: it "counsels us to interpret speakers as holding true beliefs (true by our lights at least) wherever it is plausible to do" so, and the resulting attributions are "always defeasible".

The tool: before you reject a claim that sounds absurd, check you have understood it; the likelier error is often yours. It is the remedy for the straw man in our logical fallacies cheat sheet.

6. Hitchens's razor

TRUE that the sentence is Christopher Hitchens's. It is in his Slate column of 20 October 2003: "what can be asserted without evidence can also be dismissed without evidence." He did not present it as his own discovery; in the same sentence he calls it, with "extraordinary claims require extraordinary evidence", one of "the elementary rules of logic".

It is a rule about dismissal, not disproof. It lets you decline a claim that comes with no evidence; it does not let you assert the opposite. As our fallacies page puts it, declining to believe something is not the same as claiming to have disproved it. A Latin maxim of the same shape, quod gratis asseritur, gratis negatur, is often given as its ancestor; we could not trace it to a dated source in the Internet Archive's full-text search or Wiktionary, so it stays UNVERIFIED.

7. "Extraordinary claims require extraordinary evidence"

PARTLY TRUE as Carl Sagan's rule. Whatever he said, the idea is two centuries older. Hume, in the Enquiry Concerning Human Understanding (1748): "A wise man, therefore, proportions his belief to the evidence." Laplace, in his Essai philosophique sur les probabilités (1814): "plus un fait est extraordinaire, plus il a besoin d'être appuyé de fortes preuves" — the more extraordinary a fact, the more it needs the support of strong proofs. In 1900 Théodore Flournoy named it the "principe de Laplace": "Le poids des preuves doit être proportionné à l'étrangeté des faits" — the weight of the evidence should match the strangeness of the facts.

In 1978 Marcello Truzzi's journal Zetetic Scholar opened with the rule that "where the claims are extraordinary, the burden of proof increases proportionately", and its second issue carried a contributor's comment, by Laurent Beauregard: "Extraordinary claims require extraordinary proof."

The half people forget is Laplace's own caution, in the same book (quoted here in the 1840 edition's wording): "il serait peu philosophique de nier les phénomènes, uniquement parce qu'ils sont inexplicables dans l'état actuel de nos connaissances" — it would be unphilosophical to deny phenomena only because they cannot be explained in the present state of our knowledge. Truzzi's first issue says the same: "skepticism should not be confused with dogmatic denial." The tool is to scale your demand for evidence to how surprising the claim is, not to refuse anything unexplained.

8. Falsifiability

PARTLY TRUE as "if it can't be disproved it's worthless, and one counter-example kills a theory". Karl Popper took "falsifiability as his criterion for demarcating science from non-science". It is a line between kinds of claim, not a verdict on their worth: the Stanford Encyclopedia says he "is not a positivist and acknowledges that unscientific theories may be enlightening", and his non-science list includes "logic, metaphysics".

Popper separated logic from practice. In logic, "a single genuine counter-instance falsifies the corresponding universal law"; but he "explicitly allows for the fact that in practice a single conflicting or counter-instance is never sufficient methodologically for falsification". The criterion also has critics: Imre Lakatos pointed out that a theory "falsified" in Popper's sense "may still be true", and his emphasis on the logic of falsification was, in the encyclopedia's words, "superseded in the eyes of many by the socio-historical approach taken by Thomas Kuhn". The tool that survives: ask what observation would change your mind. If the answer is none, you are not holding a testable claim.

9. The Socratic method

PARTLY TRUE as "teaching by asking questions until the student finds the answer". In Plato's dialogues, one form of Socrates' questioning is what scholars call the elenchus, and it usually tests a claim to know something. It has rules: the partner "must answer every question according to his own beliefs", and "the partner (not the audience if there is one) judges the outcome". It "usually concludes in the discomfiture of the partner", often in aporia, "an impasse". Socrates also applies it "to challenge views he probably held himself".

In the Meno, Meno complains, in W. R. M. Lamb's translation, that Socrates is like "the flat torpedo sea-fish; for it benumbs anyone who approaches and touches it". Socrates answers, in Benjamin Jowett's: "I perplex others, not because I am clear, but because I am utterly perplexed myself." Whether the method can prove anything is argued: perhaps it "can do no more than convict a partner of confusion", as Hugh Benson has argued, while on Gregory Vlastos's reading Socrates keeps beliefs "that have been well tested and not refuted". The tool: test a belief by asking what else you would have to accept if it were true. For who Socrates was, see Socrates, Plato and Aristotle quotes, checked.

10. The golden mean

PARTLY TRUE as "moderation in everything". Aristotle's mean in the Nicomachean Ethics (Book II, chapter 6) is "the intermediate not in the object but relatively to us". His example is a trainer: if ten pounds of food is too much and two too little, "it does not follow that the trainer will order six pounds", since that may be "too little for Milo, too much for the beginner in athletic exercises". The mean is set "by that principle by which the man of practical wisdom would determine it", which is judgement, not arithmetic.

Two things are usually dropped. "Not every action nor every passion admits of a mean": spite, shamelessness and envy among feelings, and "adultery, theft, murder" among actions, have no right amount. And "with regard to what is best and right" virtue is "an extreme", not a compromise. The phrase "golden mean" does not appear in W. D. Ross's translation of Book II. It is Horace's auream … mediocritatem, "golden moderation", in Odes 2.10, a poem about a safe middle course in life. The tool: the right amount depends on the person and the situation, and some things have none.

11. The ship of Theseus

PARTLY TRUE as "Plutarch's paradox". Plutarch reports that the Athenians kept Theseus's thirty-oared galley "down to the time of Demetrius Phalereus", replacing old timbers with "new and sound ones", so that it "became a standing illustration for the philosophers in the mooted question of growth, some declaring that it remained the same, others that it was not the same vessel." He is describing a debate that already existed, not posing a puzzle of his own.

The famous second half is Thomas Hobbes's, in De Corpore (Part II, chapter 11): suppose "some man had kept the old planks as they were taken out", and put them back together "in the same order"; there would then be two ships with a claim to be the same one. Hobbes also gives an answer: "we must consider by what name anything is called, when we inquire concerning the identity of it", since "it is one thing to ask concerning Socrates, whether he be the same man, and another to ask whether he be the same body". Which ship is really Theseus's is left to the reader. The tool: before arguing whether something is "the same", ask "the same what?"

Our reading

The tools that come from philosophy are usually more careful than their slogans. Hume asked for a reason, Popper allowed that theories with anomalies are often kept, Aristotle's mean depended on the person, and Laplace warned against denying what cannot yet be explained. The popular versions keep the confident half.

Sources

David Hume, A Treatise of Human Nature (1739–40), Book 3, Part 1, Section 1; An Enquiry Concerning Human Understanding (1748), Section X; R. Cohon, "Hume's Moral Philosophy", Stanford Encyclopedia of Philosophy.
J. Zupko, "John Buridan", Stanford Encyclopedia of Philosophy; Aristotle, On the Heavens II.13, tr. J. L. Stocks; Spinoza, Ethics II, Prop. 49, note, tr. R. H. M. Elwes; Averroes, Tahafut al-Tahafut, tr. S. Van den Bergh (1954).
Blaise Pascal, Pensées, fragment 233 (Brunschvicg); A. Hájek, "Pascal's Wager", Stanford Encyclopedia of Philosophy.
S. Freeman, "Original Position"; M. Fleurbaey, "Economic Justice"; S. O. Hansson, "Risk", all Stanford Encyclopedia of Philosophy; J. C. Harsanyi, Journal of Political Economy 61 (1953) and 63 (1955).
W. V. O. Quine, Word and Object (1960), p. 59 n. 2; N. L. Wilson, "Substances without Substrata", Review of Metaphysics 12 (1959); J. Malpas, "Donald Davidson", Stanford Encyclopedia of Philosophy.
Christopher Hitchens, "Mommie Dearest", Slate, 20 October 2003.
P.-S. Laplace, Essai philosophique sur les probabilités (1814; 1840 ed.); T. Flournoy, Des Indes à la planète Mars (1900); Zetetic Scholar 1 and 2 (1978).
S. Thornton, "Karl Popper", Stanford Encyclopedia of Philosophy.
P. Woodruff, "Plato's Shorter Ethical Works", Stanford Encyclopedia of Philosophy; Plato, Meno 80a–c, tr. W. R. M. Lamb and B. Jowett.
Aristotle, Nicomachean Ethics II.6, tr. W. D. Ross; Horace, Odes 2.10.
Plutarch, Life of Theseus 23.1, tr. B. Perrin; Thomas Hobbes, De Corpore, Part II, ch. 11, in English Works, ed. W. Molesworth, vol. 1 (1839).

Checked September 2026.

Mindsets, Checked: 7 Popular Ideas and What the Research Actually Found

Hacks Vitae — Tue, 22 Sep 2026 00:00:00 GMT

"Mindset" posts promise that changing one belief will change your life. Some of the ideas behind them come from real research, some from self-help books, and some from research that did not hold up. Here are seven, checked against the papers. This page describes findings; it does not give health or money advice.

1. "Scarcity mindset vs abundance mindset"

PARTLY TRUE, and it is two different ideas sharing a word. The first is a self-help attitude from the book The 7 Habits of Highly Effective People. FranklinCovey, which runs the 7 Habits courses, describes it under "The Principle of Abundance": when we think in scarcity, "Life becomes a zero-sum game. We start to believe there's only so much to go around". That is an attitude to other people, not a research finding.

The second is a research programme by Sendhil Mullainathan and Eldar Shafir. Its claim is about attention: "scarcity changes how people allocate attention: It leads them to engage more deeply in some problems while neglecting others" (Shah, Mullainathan and Shafir, Science, 2012). In their book Scarcity (2013) they write: "It is not that the poor have less bandwidth as individuals. Rather, it is that the experience of poverty reduces anyone's bandwidth." Nothing in the papers and book excerpt we read says that an "abundance mindset" helps.

DISPUTED is the fair verdict on how robust that research is. A 2013 Science paper by Mani and colleagues found that thinking about finances "reduces cognitive performance among poor but not in well-off participants". A reanalysis by Wicherts and Scholten "fails to corroborate their findings", and the authors replied that the effect "remains reliable across our experiments". A study of low-income US households before and after payday (Carvalho, Meier and Wang, 2016) found no difference in "the performance in cognitive function tasks". And in 2021 O'Donnell and colleagues re-ran 20 studies on scarcity priming, drawn from papers that cited the 2012 study; the set included some of Mullainathan and Shafir's own later work. They reported that "replication effect sizes were smaller than the original effect sizes for 80% of the 20 studies", and "four of our replication efforts yielded significant results". A 2022 correction fixed errors in several of those replications and noted that 18, not 20, of the original effects had been significant. In 2023 Shah, Zhao, Mullainathan and Shafir replied that the audit "deviates from its own a priori sampling criteria"; O'Donnell and colleagues answered that a broad sample was deliberate, while conceding that three of the studies measured scarcity rather than manipulating it.

2. Locus of control

TRUE as to where it comes from. Published work on the idea goes back at least to E. J. Phares in 1957, and it became famous through Julian B. Rotter's 1966 monograph, "Generalized expectancies for internal versus external control of reinforcement". The title says what it measured: a general expectation about whether what happens to you follows from your own actions. We could not read the papers in which Rotter discussed how his idea had been misread, so we say nothing about his verdict on the popular "internal good, external bad" version.

3. Maximizers vs satisficers

TRUE that "satisfice" in this sense is Herbert Simon's term, and DISPUTED that maximizers are less happy. Simon's own dictionary entry says the word, which "appears in the Oxford English Dictionary as a Northumbrian synonym for 'satisfy', was borrowed for this new use by H.A. Simon (1956)": one who chooses an option that "meets or exceeds specified criteria, but that is not guaranteed to be either unique or in any sense the best, is said to satisfice." For Simon it described a way any decision-maker might choose, not a personality type.

The personality version came later. Barry Schwartz and colleagues (2002) built a Maximization Scale and found "negative correlations between maximization and happiness, optimism, self-esteem, and life satisfaction". In a study of job seekers, "students with high maximizing tendencies secured jobs with 20% higher starting salaries", yet "maximizers were less satisfied than satisficers with the jobs they obtained" (Iyengar, Wells and Schwartz, 2006). Diab, Gillespie and Highhouse (2008) built a different scale that "did not correlate with life (dis)satisfaction", and concluded: "the interpretation of maximizers as unhappy may be due to poor measurement of the construct." These are correlations; none shows that maximizing causes unhappiness.

4. Zero-sum thinking

TRUE that it is a measured belief. Różycka-Tran, Boski and Wojciszke (2015) defined it as the belief "that one person's gain is possible only at the expense of other persons", and tested their scale in "a large pancultural project of 37 nations". A 2026 paper by Chinoy, Nunn, Sequeira and Stantcheva, "using a new survey of 20,400 US residents", found that "a more zero-sum mindset is strongly associated with more support for government redistribution, race- and gender-based affirmative action, and more restrictive immigration policies." The authors report that it correlates with stronger alignment with the Democratic Party but is "not primarily a partisan issue"; it helps explain differences within both parties. Those are associations, not causes. Which policies are right is a question of values, and we leave it there.

5. Beginner's mind

PARTLY TRUE as Shunryu Suzuki's sentence. The famous line, "In the beginner's mind there are many possibilities, but in the expert's there are few", comes from Zen Mind, Beginner's Mind (1970). David Chadwick, writing about the book in 2008, reports that a transcript of a lecture dated 10 November 1965 has Suzuki saying: "In beginner's mind we have many possibilities, but in expert mind there is not much possibility." "Only when Trudy polished it did it become" the famous version, Chadwick writes, meaning Trudy Dixon, who edited the book. Suzuki's comment on the finished book, as Chadwick reports: "Good book. I didn't write it, but it looks like a good book." Chadwick describes a book of "Informal Talks on Zen Meditation and Practice", drawn from lectures Suzuki gave in Los Altos that Marian Derby recorded.

6. Growth mindset

DISPUTED in how much it helps. The idea is associated with Carol Dweck, a co-author of the 2019 study below. Large reviews of interventions disagree: a 2018 pair of meta-analyses found "Overall effects were weak for both meta-analyses"; a 2019 national experiment found an intervention "improved grades among lower-achieving students"; and in 2023 one meta-analysis found an effect that "was nonsignificant after correcting for potential publication bias" while another found a small effect on academic achievement for targeted groups in high-fidelity interventions. We cover the evidence in more detail in Growth Mindset.

7. Learned helplessness

PARTLY TRUE as "you learn to stop trying". The original experiments were real: Martin Seligman and Steven Maier's 1967 paper, "Failure to escape traumatic shock", used dogs and electric shock, and compared animals that could stop the shocks with animals given matched shocks (in 1967, the escape group's average shock duration) that they could not control. Their theory was that the animals "learned that outcomes were independent of their responses". In 2016 the same two researchers wrote that "the original theory got it backward", that "Passivity in response to shock is not learned. It is the default, unlearned response to prolonged aversive events", and that what animals learn is that "they can control aversive events". Most of this is animal research; the authors label its use for human depression as speculation.

Our reading

The mindset ideas that last are usually narrower than the posters: a belief about control, a limit on attention, a word for "good enough". The clearest correction here came from the researchers themselves, and that is a good model. If a mindset claim comes without a study, or with one that was never re-run, it is closer to a story than a finding.

Sources

FranklinCovey, "Habit 4: Think Win-Win".
S. Mullainathan and E. Shafir, Scarcity (2013), publisher's excerpt; A. K. Shah, S. Mullainathan and E. Shafir, Science 338 (2012); A. Mani et al., Science 341 (2013); J. M. Wicherts and A. Z. Scholten, and the authors' reply, Science 342 (2013); L. S. Carvalho, S. Meier and S. W. Wang, American Economic Review 106 (2016); M. O'Donnell et al., PNAS 118 (2021), and correction, PNAS 119 (2022); A. K. Shah, J. Zhao, S. Mullainathan and E. Shafir, and M. O'Donnell, L. D. Nelson and D. A. Moore, PNAS 120 (2023).
E. J. Phares, "Expectancy changes in skill and chance situations", Journal of Abnormal and Social Psychology (1957); J. B. Rotter, Psychological Monographs: General and Applied 80 (1966).
H. A. Simon, "Satisficing", The New Palgrave Dictionary of Economics (2008 ed.); B. Schwartz et al., Journal of Personality and Social Psychology 83 (2002); S. S. Iyengar, R. E. Wells and B. Schwartz, Psychological Science 17 (2006); D. L. Diab, M. A. Gillespie and S. Highhouse, Judgment and Decision Making 3 (2008).
J. Różycka-Tran, P. Boski and B. Wojciszke, Journal of Cross-Cultural Psychology 46 (2015); S. Chinoy, N. Nunn, S. Sequeira and S. Stantcheva, American Economic Review 116 (2026).
D. Chadwick, review of Zen Mind, Beginner's Mind, Inquiring Mind (Fall 2008).
V. F. Sisk et al., Psychological Science 29 (2018); D. S. Yeager et al., Nature 573 (2019); B. N. Macnamara and A. P. Burgoyne, and J. L. Burnette et al., Psychological Bulletin 149 (2023).
M. E. Seligman and S. F. Maier, Journal of Experimental Psychology 74 (1967); S. F. Maier and M. E. Seligman, "Learned helplessness at fifty", Psychological Review 123 (2016).

Checked September 2026.

Nietzsche Quotes, Checked: What He Wrote and What He Never Did

Hacks Vitae — Tue, 22 Sep 2026 00:00:00 GMT

Friedrich Nietzsche wrote in short, sharp sentences that travel well on their own, so they often get cut loose from the paragraph that gave them their meaning. Here are eight famous lines and claims, checked against his German text in the digital critical edition, Nietzsche Source, and against the public-domain English translations edited by Oscar Levy.

1. "What does not kill me makes me stronger"

TRUE, but trimmed. It is maxim 8 in the section "Maxims and Missiles" of Twilight of the Idols, whose writing the Stanford Encyclopedia of Philosophy dates to "August-September 1888". In German: "Aus der Kriegsschule des Lebens. — Was mich nicht umbringt, macht mich stärker." That is: "From life's school of war. — What does not kill me makes me stronger" (our translation). Anthony Ludovici's English reads "From the military school of life."

Two things get lost. The heading frames it as a lesson in hardening, and the pronoun is "me", not "you": it reads as a remark about himself, not a promise about everyone. It also sits in a run of barbed one-liners; the next one reads "Help thyself, then everyone will help thee. A principle of neighbour-love."

2. "He who has a why to live can bear almost any how"

TRUE. The idea is maxim 12 of the same section: "Hat man sein warum? des Lebens, so verträgt man sich fast mit jedem wie?" ("If one has one's why? of life, one can put up with almost any how?", our translation). The popular "why … how" wording is actually closer to his German than Ludovici's "If a man knows the wherefore of his existence, then the manner of it can take care of itself."

The psychiatrist Viktor Frankl quotes it in Man's Search for Meaning. The popular version leaves out the second half of the maxim: "Der Mensch strebt nicht nach Glück; nur der Engländer thut das" ("Man does not strive for happiness; only the Englishman does that", our translation).

3. "Without music, life would be a mistake"

TRUE, and he wrote it more than once. It is maxim 33 of Twilight of the Idols: "Wie wenig gehört zum Glücke! Der Ton eines Dudelsacks. — Ohne Musik wäre das Leben ein Irrthum." That is: "How little is needed for happiness! The sound of a bagpipe. — Without music, life would be an error" (our translation). The same maxim ends with a joke: "The German imagines even God as a songster."

He had already written it in letters. To his friend Heinrich Köselitz, in January 1888: "Das Leben ohne Musik ist einfach ein Irrthum, eine Strapatze, ein Exil" ("Life without music is simply an error, a hardship, an exile", our translation). To Georg Brandes, in March 1888: "Ohne Musik wäre mir das Leben ein Irrthum" ("Without music, life would be an error to me", our translation).

4. "God is dead"

TRUE as a quotation, but it is not an argument that atheism is true. The phrase appears in The Gay Science (in the old translation, The Joyful Wisdom), most famously in section 125, "The Madman", in Thomas Common's translation: "God is dead! God remains dead! And we have killed him!" The madman brings this news to a crowd of people "who did not believe in God", and he only amuses them. He answers: "I come too early". The passage ends with him singing a requiem for God in churches and asking: "What are these churches now, if they are not the tombs and monuments of God?"

R. Lanier Anderson, writing in the Stanford Encyclopedia, puts it this way: "The idea is not so much that atheism is true—in GS 125, he depicts this pronouncement arriving as fresh news to a group of atheists". In a section added to the book's 1887 edition, Nietzsche explained the phrase himself: "the belief in the Christian God has become unworthy of belief". Anderson describes the response it calls for as "mourning and deep disorientation". The same 1887 section strikes a second note, though: it is titled "What our Cheerfulness Signifies", and in it Nietzsche writes that "we philosophers and 'free spirits' feel ourselves irradiated as by a new dawn" by the news. Whether Nietzsche was right about where European belief was heading is a question of interpretation, and this page does not take a side.

5. "When you gaze long into an abyss, the abyss gazes also into you"

TRUE, but it is the second half of a two-sentence aphorism, Beyond Good and Evil section 146. In Helen Zimmern's translation the whole of it reads: "He who fights with monsters should be careful lest he thereby become a monster. And if thou gaze long into an abyss, the abyss will also gaze into thee." The first sentence suggests the meaning: it reads as a warning about what a long struggle does to the one who is struggling.

6. "There are no facts, only interpretations"

TRUE, but from an unpublished notebook. The words are his, but he never put them in a book he finished. The editors of his notebooks date the notebook to the end of 1886 and spring 1887. The fragment begins as an argument against positivism, which, in our translation, "stops at the phenomenon" and says "there are only facts": "nein, gerade Thatsachen giebt es nicht, nur Interpretationen" ("no — facts are precisely what there are not, only interpretations", our translation).

The short English line is, if anything, closer to his German than the old translation of the same fragment, which became section 481 of The Will to Power: "facts are precisely what is lacking, all that exists consists of interpretations." In the same note he calls his view "Perspektivismus".

7. "And those who were seen dancing were thought to be insane by those who could not hear the music"

MYTH. Neither we nor the researchers at Quote Investigator have found it in his writings. We searched complete public-domain English translations of his published books and of The Will to Power, and the German of Zarathustra, Twilight of the Idols, The Gay Science, Beyond Good and Evil and the notebook in section 6, and found no match. His full notebooks were not searched, so strictly the finding is "not found", not "never said". The MYTH label is for the attribution: the earliest attribution to him that anyone has reported is from 2003, and the saying was already in print as "the old proverb" in 1927.

The quotation researchers at Quote Investigator report that they have "not yet located substantive evidence that Nietzsche wrote or said the statement", and that the earliest attribution to him they found is from 2003, in an internet discussion group, more than a century after his death. Among the earlier versions they list is one by Germaine de Staël, who in De l'Allemagne (1813) compared a moment of spiritual detachment to being "comme dans un bal dont on n'entendrait pas la musique" ("as at a ball where one could not hear the music", our translation), where the dancing would look "insensé" (senseless, our translation). In her version the dancers look mad; the modern saying turns it into praise of the dancers.

He did write about dancing. In Common's translation of Zarathustra: "I should only believe in a God that would know how to dance," and "one must still have chaos in one, to give birth to a dancing star."

8. "The Will to Power is Nietzsche's last book"

MYTH. He never finished it. Nietzsche had a mental breakdown in Turin, according to the Stanford Encyclopedia, "On the morning of January 3, 1889", and his sister Elisabeth "assumed control of his literary legacy". The book known as The Will to Power is "a selection of writing from his notebooks", and Anderson writes that the editing "was not well founded in Nietzsche's surviving plans for the book" and "was also marred by Elisabeth's strong anti-Semitic commitments, which had been extremely distressing to Nietzsche himself." Its own English translator, Anthony Ludovici, wrote in his preface: "Unfortunately, The Will to Power was never completed by its author."

Elisabeth moved her brother and his manuscripts to a house in Weimar, which became the home of the Nietzsche Archive. Robert Wicks, also in the Stanford Encyclopedia, writes that aspects of Nietzsche's thought were taken up by the Nazis and Italian Fascists, "partly due to the encouragement of Elisabeth Förster-Nietzsche", and that "It was possible for the Nazi interpreters to assemble, quite selectively, various passages from Nietzsche's writings". Nietzsche himself wrote in a letter of 1887 that he felt too foreign to the "German spirit" of his day to watch its quirks patiently, and that among them he counted "den Antisemitismus" (antisemitism) in particular (our translation). The record is not simple, though: Anderson also notes that in On the Genealogy of Morals "he freely deploys offensive anti-Semitic tropes", which he turns back against antisemitic Christians. He died in 1900.

His most famous coinage, the Übermensch, was misread too. It is "Superman" in the old translations and "overman" in the Stanford Encyclopedia. In Ecce Homo he complained that the word was misunderstood, and that "Other learned cattle have suspected me of Darwinism on account of this word". What he meant by it is still argued. Wicks describes it as a "higher mode of being"; Anderson reports, among other readings, the view that Nietzsche's position on the overman is "riven by internal contradiction".

Our reading

The real Nietzsche lines are sharper than the poster versions and usually stranger: a maxim about himself, a joke about bagpipes, a madman grieving in a crowd of atheists, a warning about monsters. The fake one is the gentlest of the lot. A good rule for any Nietzsche quote: find the section number. If nobody can give one, treat the attribution as unproven.

Sources

Friedrich Nietzsche, German texts in the Digitale Kritische Gesamtausgabe, ed. Paolo D'Iorio, Nietzsche Source (nietzschesource.org): Götzen-Dämmerung, Die fröhliche Wissenschaft, Jenseits von Gut und Böse, notebook fragment 7[60], and letters of 1887 and 1888.
The Complete Works of Friedrich Nietzsche, ed. Oscar Levy, Project Gutenberg: Twilight of the Idols and Ecce Homo, trans. Anthony M. Ludovici; The Joyful Wisdom and Thus Spake Zarathustra, trans. Thomas Common; Beyond Good and Evil, trans. Helen Zimmern; The Will to Power, trans. Anthony M. Ludovici.
R. Lanier Anderson, "Friedrich Nietzsche", and Robert Wicks, "Nietzsche's Life and Works", Stanford Encyclopedia of Philosophy.
Job Smit, Carmen Schuhmann and Annelieke Damen, "Responding to Life Itself", Journal of Pastoral Care & Counseling (2024), on Frankl's use of the line.
Quote Investigator, "Those Who Dance Are Considered Insane by Those Who Can't Hear the Music" (2012, updated 2025).
Germaine de Staël, De l'Allemagne, Project Gutenberg.

Checked September 2026.

Socrates, Plato and Aristotle Quotes, Checked: What They Actually Said

Hacks Vitae — Tue, 22 Sep 2026 00:00:00 GMT

Socrates, Plato and Aristotle are among the names most often attached to ancient wisdom, and all three collect lines they never wrote. Socrates in particular wrote no philosophy. Here are ten claims, checked against public-domain English translations: Benjamin Jowett's Plato, W. D. Ross's and D. P. Chase's translations of Aristotle's Nicomachean Ethics, William Ellis's Politics, and R. D. Hicks's translation of Diogenes Laertius, with the Greek where it matters. When we say a line is not in a work, we mean we searched the complete translations named in that section.

1. "Socrates wrote down his philosophy"

MYTH. He wrote no philosophy. (Plato's Phaedo says that in prison he put some of Aesop's fables into verse.) The Stanford Encyclopedia of Philosophy describes him as famous "despite having written nothing", and the Internet Encyclopedia of Philosophy explains the consequence: "Because he wrote nothing, what we know of his ideas and methods comes to us mainly from his contemporaries and disciples." Those portraits differ. The Stanford Encyclopedia calls the result the "Socratic problem", "a rat's nest of complexities arising from the fact that various people wrote about Socrates whose accounts differ in crucial respects". So "Socrates said" usually means "Plato has Socrates say".

2. "The unexamined life is not worth living"

TRUE. It is in Plato's Apology 38a, his version of Socrates' defence speech. The Stanford Encyclopedia renders it "the unexamined life is not worth living for a human being", said "at his trial". In Jowett's translation, Socrates has just said that to discourse daily "about virtue, and of those other things about which you hear me examining myself and others, is the greatest good of man", and "the unexamined life is not worth living". He is explaining why he cannot simply hold his tongue and go to live in another city.

3. "I know that I know nothing"

PARTLY TRUE. The idea, known as Socratic ignorance, is real; the tidy paradox is not in the Apology. At 21d, Socrates compares himself with a man who thinks he is wise. In Jowett's translation: "for he knows nothing, and thinks that he knows; I neither know nor think that I know". The Greek of that sentence is narrower: "ὅτι ἃ μὴ οἶδα οὐδὲ οἴομαι εἰδέναι", "that what I do not know, I do not think I know" (our translation). The point is about not pretending. But the Apology does come close to the famous line elsewhere: at 21b Socrates says "I know that I have no wisdom, small or great", and at 22d, among the craftsmen, "I was conscious that I knew nothing at all, as I may say". What it never does is turn this into the neat paradox; even at 22d he hedges with "as I may say". The tidy English version does appear in Jowett's own introduction to his translation: "He knew nothing, and knew that he knew nothing". That is the translator summarising, not Plato.

Other texts come close too. At the end of the first book of Plato's Republic, Socrates says: "the result of the whole discussion has been that I know nothing at all. For I know not what justice is". And the biographer Diogenes Laertius writes that Socrates said "he knew nothing except just the fact of his ignorance".

4. "Wisdom begins in wonder"

PARTLY TRUE. The word is philosophy, not wisdom. In Plato's Theaetetus, in Jowett's translation, Socrates says: "wonder is the feeling of a philosopher, and philosophy begins in wonder". Aristotle says much the same in his Metaphysics, in Ross's translation: "For it is owing to their wonder that men both now begin and at first began to philosophize".

5. "Be kind, for everyone you meet is fighting a hard battle"

MYTH as Plato. It is not in the eight Jowett dialogues we searched, and its traceable origin is modern. The quotation researchers at Quote Investigator trace it to a Christmas message around the end of 1897 by Ian Maclaren, "the pseudonym or pen name of Rev. John Watson", in the form: "Be pitiful, for every man is fighting a hard battle." As they note, "The original aphorism did not use the word 'kind'."

6. "The measure of a man is what he does with power"

MYTH as Plato, though a real Greek proverb lies behind it. It is not in the Jowett dialogues we searched, we could not find where the English wording comes from, and none of the pages we checked that repeat it gives a dialogue and passage. The ancient proverb is about office, and even the ancient sources disagree about who said it. Diogenes Laertius gives it to the sage Pittacus: "Office shows the man." Aristotle, in the Nicomachean Ethics, gives it to another sage, Bias, and explains why it matters, in Ross's translation: "the saying of Bias is thought to be true, that 'rule will show the man'; for a ruler is necessarily in relation to other men and a member of a society."

7. "It is the mark of an educated mind to be able to entertain a thought without accepting it"

MYTH as Aristotle. It is not in the Nicomachean Ethics, where the real "mark of an educated man" line is, and it says something quite different. In Ross's translation of Book I, chapter 3: "it is the mark of an educated man to look for precision in each class of things just so far as the nature of the subject admits". Aristotle is saying you should not demand the same certainty from ethics as from mathematics: "it is evidently equally foolish to accept probable reasoning from a mathematician and to demand from a rhetorician scientific proofs".

8. "Knowing yourself is the beginning of all wisdom"

MYTH as Aristotle. We did not find it in the Nicomachean Ethics, the Politics or the Metaphysics (all fourteen books, in the Ross and Smith translation on Wikisource), and none of the pages we checked that repeat it gives a passage. "Know thyself" is real, but it is older than Aristotle: it was an inscription at Delphi. In Plato's Protagoras, in Jowett's translation, it is one of "the far-famed inscriptions, which are in all men's mouths--'Know thyself,' and 'Nothing too much.'" In the Phaedrus, Socrates says: "I must first know myself, as the Delphian inscription says".

9. "Man is by nature a political animal"

TRUE. It is Aristotle, Politics Book I. In William Ellis's translation: "man is naturally a political animal".

10. "Plato's noble lie" and "the cave"

TRUE, both are in Plato's Republic. In Book III, Socrates proposes a founding myth for his ideal city. Jowett translates it as a "royal lie": "How then may we devise one of those needful falsehoods of which we lately spoke—just one royal lie which may deceive the rulers, if that be possible, and at any rate the rest of the city?" "Noble lie" is another translation of the same Greek word, gennaion. Eric Brown, in the Stanford Encyclopedia, writes that the rulers "advance a 'noble lie' to convince citizens of their unequal standing and deep tie to the city", and judges: "This propagandistic control plainly represents a totalitarian concern". He is careful to separate it from the city's laws, of which he writes: "There is nothing especially totalitarian about the rule of law pervasive in Kallipolis". That is his reading.

The cave, in Book VII, is an allegory Socrates tells to Glaucon in the dialogue. Jowett: "Behold! human beings living in a underground den, which has a mouth open towards the light", whose prisoners "have their legs and necks chained so that they cannot move".

Our reading

The misquotes here share a pattern: they turn a specific, often uncomfortable remark into a general piece of advice. Socrates' ignorance becomes a clever paradox, Aristotle's point about precision becomes a slogan about open-mindedness, and a proverb about what office reveals becomes a line about power. The originals are harder, and we think better. The quickest check is to ask for the book and the passage, then read the paragraph around it.

Sources

Plato, Apology, Theaetetus, Republic, Protagoras, Charmides, Phaedrus, Gorgias and Laws, trans. Benjamin Jowett, Project Gutenberg; Greek text ed. John Burnet (Oxford, 1903), Perseus Digital Library.
Aristotle, Nicomachean Ethics, trans. W. D. Ross, Wikisource, and trans. D. P. Chase, Project Gutenberg; Metaphysics, trans. W. D. Ross and J. A. Smith (1908), Wikisource; Politics, trans. William Ellis, Project Gutenberg.
Diogenes Laertius, Lives of Eminent Philosophers, Books I and II, trans. R. D. Hicks, Wikisource.
Debra Nails and S. Sara Monoson, "Socrates", and Eric Brown, "Plato's Ethics and Politics in The Republic", Stanford Encyclopedia of Philosophy; James M. Ambury, "Socrates", Internet Encyclopedia of Philosophy.
Quote Investigator, "Be Kind; Everyone You Meet Is Fighting a Hard Battle" (2010, updated 2025).

Checked September 2026.

Where the Trolley Problem Comes From: 10 Claims, Checked

Hacks Vitae — Tue, 22 Sep 2026 00:00:00 GMT

The trolley problem is a thought experiment in ethics: a runaway vehicle will kill five people unless you divert it onto a track where it will kill one. Here are ten common claims about where it came from and what it shows, checked against the papers themselves. Whether you should pull the lever is a question of values, and this page does not answer it.

1. "Philippa Foot invented the trolley problem"

PARTLY TRUE. She wrote the case everything descends from, in "The Problem of Abortion and the Doctrine of the Double Effect" (Oxford Review, 1967). In the text as later reprinted in her collection Virtues and Vices, the chooser "is the driver of a runaway tram which he can only steer from one narrow track on to another"; "five men are working on one track and one man on the other". But she did not give it a name, and she built it from an example already in circulation. After a judge case that, she says, "the controversy has raged around", she writes: "Beside this example is placed another in which a pilot whose airplane is about to crash is deciding whether to steer from a more to a less inhabited area. To make the parallel as close as possible it may rather be supposed that he is the driver of a runaway tram". The tram is her adjustment of the pilot case. The name came later (see 4).

2. "The original paper was about trolleys"

MYTH. It was about abortion. Foot's stated method was "setting up parallels involving adults or children once born", and her target was the doctrine of double effect, which she defines as "the thesis that it is sometimes permissible to bring about by oblique intention what one may not directly intend". The tram is one of those parallels. She pairs it with a judge, faced with rioters, who can prevent the bloodshed "only by framing some innocent person and having him executed", and asks why we would say "without hesitation, that the driver should steer for the less occupied track, while most of us would be appalled at the idea that the innocent man could be framed". Her answer is that the driver faces "a conflict of negative duties", while the judge is "weighing the duty of not inflicting injury against the duty of bringing aid". Then she writes: "Let us now return to the problem of abortion". And she wrote "tram", not "trolley".

3. "The dilemma began in 1967"

MYTH. A version was put to students at the University of Wisconsin in December 1905, and it was not new then: Sharp writes that "one hundred and forty-five persons ... answered IX of Series I in an inquiry made in 1896 and 1897", and that "the returns of 1905 were practically identical." Frank Chapman Sharp published the results in 1908 in A Study of the Influence of Custom on the Moral Judgment. His Question IX: a man sees that "a careless switchman had left a switch open which would mean death or injury to the several hundred people on the" train, while "he saw his only child playing upon the track in front of the approaching engine. He had time only to turn the switch and save the train or else to save the child." It is not the modern case: the one person is the chooser's own child, and Sharp was studying custom and moral judgment.

German criminal law has its own "switchman" (Weichensteller) case, discussed by Karl Engisch in 1930 and Hans Welzel in 1951, according to a 2019 article by Liane Wörner. In Welzel's version a railway official throws the switch at the last minute and, as he foresaw, three workers are killed. German criminal law treats it differently from most survey respondents: according to Wörner, under the overwhelmingly prevailing view today, throwing the switch "kann … nicht gerechtfertigt werden" (cannot be justified; our translation), though it may be excused on other grounds. Neither Foot's 1967 paper nor Thomson's 1985 paper cites any of these.

4. "Judith Jarvis Thomson gave it its name"

TRUE. Her 1976 paper in The Monist is titled "Killing, Letting Die, and the Trolley Problem". In her 1985 Yale Law Journal paper, "The Trolley Problem", she opens by crediting Foot ("Some years ago, Philippa Foot drew attention to an extraordinarily interesting problem"), turns Foot's tram into a trolley, and changes the chooser from the driver to "a case I will call Bystander at the Switch". Then: "Since I find it particularly puzzling that the bystander may turn his trolley, I am inclined to call this The Trolley Problem." She adds: "Those who find it particularly puzzling that the surgeon may not operate are cordially invited to call it The Transplant Problem instead."

5. "The footbridge version is called 'the fat man'"

TRUE, and the label is older than the footbridge. Thomson's 1985 paper sets out "a case … which I shall call Fat Man", in which "you are standing on a footbridge over the trolley track". But Foot had already used a fat man in 1967, in a different story she calls "well known to philosophers": "the fat man stuck in the mouth of the cave", whom the people trapped behind him could blast out with dynamite. Foot does not say who first told it. The 2020 study in 6 calls it "Footbridge".

6. "Most people would pull the lever"

PARTLY TRUE, for the switch, and much less so for the footbridge. A large study, by Awad and colleagues in PNAS (2020), analysed "three sacrificial dilemmas by 70,000 participants in 10 languages and 42 countries". Participants endorsed the sacrifice in the switch case at a "country-level average: 81%", in a loop variant at 72%, and in the footbridge case at 51%. The ordering was the same everywhere, but "the quantitative acceptability of each sacrifice, however, showed substantial country-level variations". Two cautions come from the authors themselves: "our sample shows clear signs of self-selection", and "we estimate that a third of our participants were young, college-educated men".

Thomson's own "Everybody to whom I have put this hypothetical case says, Yes, it is", said of Foot's driver case, was a remark about the people she had asked, not a survey.

7. "What people answer in a trolley dilemma predicts what they would do"

DISPUTED. Psychologists took the dilemmas up. Greene and colleagues used them in brain scans in 2001, arguing that "moral dilemmas vary systematically in the extent to which they engage emotional processing". Critics question what the answers show. Bauman and colleagues (2014) worried that such studies "may lack experimental, mundane, and psychological realism", partly because "they are amusing rather than sobering". FeldmanHall and colleagues (2012) found that "real moral decisions can dramatically contradict moral choices made in hypothetical scenarios", though their task was not a trolley case. In a 2018 study in which participants chose whether to "administer an electroshock (that they did not know was bogus) to a single mouse or allow five other mice to receive the shock", Bostyn and colleagues reported that "responses to hypothetical dilemmas are not predictive of real-life dilemma behavior, but they are predictive of affective and cognitive aspects of the real-life decision". Five later pieces in the same journal discuss it, and it is one study. On the other side, in a virtual-reality simulation of the switch case, Navarrete and colleagues (2012) found that "the majority of research participants behaved as 'moral utilitarians'", the choice most people also make on paper. The fair summary: the dilemmas reliably show how people judge; how well they show what people do is argued over.

8. "Thomson always held that you should pull the lever"

MYTH. She changed her mind. According to the Stanford Encyclopedia of Philosophy, "Thomson (2008) changed her mind and argued that the consensus that it is permissible for the bystander to turn the trolley was mistaken." Her argument adds a third option: "turning the trolley onto and killing oneself". If you would not do that, the entry summarises, "we are not entitled to turn the trolley onto a stranger."

9. "It was invented to study self-driving cars"

MYTH. It is decades older (see 1 to 3). The self-driving framing arrived with work such as Bonnefon and colleagues in Science (2016), and the Moral Machine experiment in Nature (2018), "an online experimental platform designed to explore the moral dilemmas faced by autonomous vehicles", which gathered "40 million decisions in ten languages from millions of people in 233 countries and territories". The 2020 figures in 6 came from that site's "Classic" mode, "where visitors made decisions about the traditional Switch, Loop, and Footbridge variants".

10. "Self-driving cars will face real trolley problems"

DISPUTED. There are three positions, each with published support.

The question is real. Bonnefon's team wrote that such cars "will sometimes have to choose between two evils", and found that participants "approved of utilitarian AVs (that is, AVs that sacrifice their passengers for the greater good) and would like others to buy them, but they would themselves prefer to ride in AVs that protect their passengers at all costs".
The format distorts the answer. Yochanan Bigman and Kurt Gray argued in Nature (2020) that the Moral Machine's apparent preferences for some lives over others were "driven by the specific 'trolley-type' paradigm", and that "people overwhelmingly want autonomous vehicles to treat different human lives equally". The Moral Machine authors replied that Bigman and Gray "make the important point that the way we measure these preferences can affect the results we obtain".
The analogy is weak. Sven Nyholm and Jilles Smids (2016) identified "three important ways" the two differ: the decision situation, "moral and legal responsibility", and "decision-making in the face of risks and uncertainty".

The evidence does not settle which of the three is right.

Germany's Ethics Commission on Automated and Connected Driving (June 2017) recommended that in unavoidable accidents "any distinction based on personal features (age, gender, physical or mental constitution) is strictly prohibited", that it is "also prohibited to offset victims against one another", and that "general programming to reduce the number of personal injuries may be justifiable". It added: "Those parties involved in the generation of mobility risks must not sacrifice non-involved parties." It also noted a difference from the classic case: in a real crash "the identity of the injured or killed parties is not yet known (unlike in the trolley dilemma)". Those are a report's recommendations; what a car ought to do is a question of values, and we leave it there.

Our reading

The trolley problem started as a side example in an argument about abortion, used to test the doctrine of double effect rather than to be answered for its own sake. Foot used it to ask why two cases with the same numbers feel so different. That question has lasted longer than any answer to it.

Sources

Philippa Foot, "The Problem of Abortion and the Doctrine of the Double Effect", Oxford Review 5 (1967), as reprinted in Virtues and Vices (PhilArchive copy via the Internet Archive).
Judith Jarvis Thomson, "Killing, Letting Die, and the Trolley Problem", The Monist 59(2) (1976), bibliographic record only; "The Trolley Problem", Yale Law Journal 94 (1985), Yale Law School open repository.
Frank Chapman Sharp, A Study of the Influence of Custom on the Moral Judgment, Bulletin of the University of Wisconsin (1908), Internet Archive.
Liane Wörner, "Der Weichensteller 4.0", Zeitschrift für Internationale Strafrechtsdogmatik 1/2019.
Alison McIntyre, "Doctrine of Double Effect"; Fiona Woollard and Frances Howard-Snyder, "Doing vs. Allowing Harm", Stanford Encyclopedia of Philosophy.
Awad et al., "Universals and variations in moral decisions made in 42 countries by 70,000 participants", PNAS 117 (2020); Awad et al., "The Moral Machine experiment", Nature 563 (2018); Bonnefon, Shariff and Rahwan, "The social dilemma of autonomous vehicles", Science 352 (2016).
Greene et al., Science 293 (2001); Navarrete et al., Emotion 12 (2012); Bauman et al., Social and Personality Psychology Compass 8 (2014); FeldmanHall et al., Cognition 123 (2012); Bostyn, Sevenhant and Roets, Psychological Science 29 (2018).
Bigman and Gray, "Life and death decisions of autonomous vehicles", and the authors' reply, Nature 579 (2020); Nyholm and Smids, Ethical Theory and Moral Practice 19 (2016).
Ethics Commission on Automated and Connected Driving, Report (June 2017), German Federal Ministry of Transport.

Checked September 2026.

The Aztec and Inca Empires: 9 Popular Claims, Checked

Hacks Vitae — Fri, 18 Sep 2026 00:00:00 GMT

The Aztec and Inca empires fell to Spanish invaders within about a decade of each other, in the 1520s and 1530s. Most of what the world repeats about them comes from the conquerors, from friars writing decades later, or from 19th-century retellings. Here are nine common claims, checked against the conquest-era sources and against archaeology.

1. "The Aztecs thought Cortés was the god Quetzalcoatl"

DISPUTED. The story is old, but it is not found in the one account written at the time. In his second letter to Charles V, written in 1520, Hernán Cortés reports Moctezuma telling him "that our race was brought to these parts by a lord, whose vassals they all were, and who returned to his native country", and that the lord's descendants would "come to subjugate this country and us, as his vassals". Later in the speech, as Cortés reports it, Moctezuma says: "we believe, and hold for certain, that he is our rightful sovereign", meaning Charles V. It names a lord, not a god, and never mentions Quetzalcoatl. The link to Quetzalcoatl in the English edition comes from its 1908 editor, Francis MacNutt, in a note.

The god story appears in print decades later. The historian Camilla Townsend (American Historical Review, 2003) names a history published in 1552 by Francisco López de Gómara, Cortés's chaplain and secretary, who had never been to Mexico, as one of the first to claim in print that the Spaniards were taken for gods. Its fullest version is in the work of the Franciscan friar Bernardino de Sahagún, compiled with Indigenous informants from the 1550s. There the coast-watchers who meet an earlier expedition, led, the text says, by Juan de Grijalva in 1518, kiss the prows of the ships in adoration, "pensant que c'était le dieu Quetzalcoatl qui revenait": thinking it was the god Quetzalcoatl returning (our translation, from the French edition of 1880). When Cortés's ships appear the following year, Moctezuma sends envoys to receive Quetzalcoatl, "pensant que c'était lui qui venait": thinking it was he who was coming.

Scholars have divided over this. Townsend lists David Carrasco (1982) and H. B. Nicholson (2001) among scholars who accepted the story. She argues that it took shape after the conquest, and Matthew Restall's 2018 book When Montezuma Met Cortés challenges the long habit, in one reviewer's words, of casting Moctezuma's actions "as a surrender to prophecy". The evidence leans toward a later origin: as a summary of Townsend's work in JSTOR Daily notes, "Cortés own letters during the conquest make no mention of being mistaken for or interpreted as a god." The record cannot show what Moctezuma privately believed.

2. "A few hundred Spaniards conquered the Aztec Empire"

MYTH. Cortés counted his own forces when the siege of Tenochtitlan began in 1521: "eighty-six horsemen, a hundred and eighteen bowmen and musketeers, seven hundred and odd foot soldiers". By our arithmetic that is about 900 Spaniards. Beside them, in his own words, "there were more than fifty thousand warriors" from allied peoples on one occasion, and "more than twenty-five thousand warriors of Tascaltecal", that is Tlaxcala, in one division alone. These are round figures from a man with reasons to make his campaign look both bold and well supported, but they are his.

Disease struck before the siege. Cortés writes of "the sickness of small-pox", from which one of the lords of Tlaxcala "had died". In Sahagún's list of great epidemics, "La première eut lieu en 1520": the first was in 1520 (our translation). Historians now describe a conquest fought inside a world of Mesoamerican alliances. In a review of the scholarly volume Indian Conquistadors (2007), the historian Raphael Folsom summarises: "the Spaniards stumbled into a Mesoamerican world of constantly shifting alliances and coalitions."

3. "The Aztecs sacrificed 80,400 people in four days in 1487"

UNVERIFIED. Human sacrifice was real, and archaeology confirms it. The figure is another matter. It is best known from the friar Diego Durán, writing in 1581, almost a century later, who says the temple of Mexico was rededicated in 1487 with eighty thousand four hundred men (our translation). Durán says he took it from the history he was translating ("Dice la historia"). He added that the number "se me hizo tan increíble" (seemed to me so incredible, in modern spelling) that he would not have dared set it down, for fear of being taken for a man who wrote fables, had his source not compelled him and had he not found it "en otros muchos lugares", in many other places, written and painted (our translation). Another colonial record disagrees: the Codex Telleriano-Remensis pictures 20,000 victims, and the Spanish note beside the picture says 4,000 men, according to the historian Xavier Noguez (Arqueología Mexicana, 2011), who adds that most scholars doubt such large totals.

Between 2015 and 2017, Mexico's National Institute of Anthropology and History (INAH) excavated the great skull rack, the Huei Tzompantli, beside the Templo Mayor. By February 2023 it reported "655 cráneos humanos", 655 human skulls, identified so far (our translation). Two of the bioarchaeologists who worked on the site, Ximena Chávez Balderas and Lorena Vázquez Vallín, write that the finds "prove both the existence of human sacrifice and the fact that Spanish sources exaggerated the numbers of people sacrificed in Tenochtitlan." Archaeology counts what survived, so it cannot give a total either.

4. "They called themselves Aztecs"

PARTLY TRUE. The people of Tenochtitlan called themselves Mexica. Durán records both names from their migration story: they came from a land called Aztlan, "y así les llamaban á estas naciones Azteca" (and so these nations were called Azteca), and "por otro nombre Mecitin, que quiere decir Mexicanos" (by another name Mecitin, which means Mexicans) (our translations). The historian Susan Kellogg explains that the Mexica "called themselves 'Azteca' early in their history but then were commanded by a patron deity to change that name," and that "Late colonial and nineteenth-century writers like Francisco Javier Clavijero and William Prescott popularized the use of 'Aztec'". She also notes that "those we call Aztecs never saw themselves as a single people." Kellogg still uses "Aztec" in her book because, she writes, it "remains widely recognized".

5. "The Inca were a people"

PARTLY TRUE. "Inca" was a title. The chronicler Garcilaso de la Vega wrote in 1609 that the name, "when applied to the prince, signified lord, king, or emperor", and that more broadly "it may be said to indicate a person of the blood royal." The empire had its own name, which modern writers spell Tawantinsuyu: Garcilaso says it "means the four quarters of the world". Two of the first Spanish reports did not use the word at all. Pizarro's secretary, Francisco de Xerez, called Atahualpa's late father, the emperor Huayna Capac, "Old Cuzco", and Hernando Pizarro did the same; the 1872 editor Clements Markham notes that "Xeres never seems to have heard the word Ynca", "mistaking the name of the capital city for the name of the sovereign."

6. "The Inca had no writing"

DISPUTED. The Inca had no writing on paper, but they kept records on knotted cords called khipu. Whether those cords could also record words is still open. Garcilaso says "they recorded every thing relating to numbers by means of the knots", but that "neither the words nor the reasoning, nor any historical event could be expressed by the knots."

Two studies, from 2017 and 2018, argue that some khipus did record names. The anthropologist Sabine Hyland (Current Anthropology, 2017) studied khipu letters from the village of Collata and concluded that khipus "could constitute an intelligible writing system, possibly logosyllabic". She adds that the Collata khipus are 18th-century and that "It is unclear how similar the 18th century khipu missives may have been to the earlier Inka khipu letters." Manuel Medrano and Gary Urton (2018) matched six colonial khipus to a written census and argue they record villagers' names. As Hyland also writes, "no specific khipu has ever been reliably identified as a narrative text."

7. "Pizarro captured the Inca emperor with 168 men and took a room full of gold"

TRUE, with context. In November 1532, "a group of 168 Spaniards seized the Inca emperor Atahuallpa in the town of Cajamarca", as the historian James Lockhart's study The Men of Cajamarca (1972) is described by its publisher. Xerez's own count at the last muster before Cajamarca is slightly lower: "sixty-two horsemen and one hundred and two foot soldiers", 164 in all. The empire was already split by a civil war between Atahualpa and his brother Huascar. Xerez reports Atahualpa's offer: "I will give gold enough to fill a room twenty-two feet long and seventeen wide, up to a white line which is half way up the wall", and silver to fill it "twice over". The ransom was paid. Atahualpa was then sentenced to burning "unless he became a Christian", and, Xerez writes, "strangled" instead, in 1533.

8. "Machu Picchu was a lost city discovered in 1911"

MYTH. Hiram Bingham, who reached it in 1911, recorded that it was not empty. Two farmers had made the terraces their home, and one "told us that they had been living here four years." Bingham also learned "that the ruins were visited in 1902 by Lizarraga", a local tenant farmer who had written his name on the granite walls.

The site's original name is also in question. A 2022 study by Donato Amado Gonzales and Brian Bauer argues, in the University of Illinois Chicago's summary, that "the Incas originally called it Huayna Picchu". Radiocarbon dating published in Antiquity in 2021 found the site "was occupied from c. AD 1420–1532", about two decades earlier than the dates in the Spanish-era chronicles.

9. "The Aztecs sacrificed people; the Inca did not"

MYTH. The Inca state carried out child sacrifice in a ceremony called capacocha. The priest Cristóbal de Molina, of the hospital in Cuzco, writing between 1570 and 1584, described it using "one or two male and female children aged about ten years." In 1999, archaeologists found three children buried near the summit of Llullaillaco, a volcano in Argentina. A 2013 study in PNAS identified them as "a 13-y-old girl and a girl and boy aged 4 to 5 y" and found evidence of "escalating coca and alcohol ingestion in the lead-up to death." The museum in Salta, Argentina, where they are kept gives their ages as about 15, six and seven. The authors relate their findings to "questions of consent, coercion, and/or compliance", and write that for such young victims "any kind of informed consent to their own deaths cannot be unproblematically presumed".

Our reading

Both empires are usually remembered through the eyes of the people who destroyed them. Spanish sources inflated the conquerors' own daring and the scale of the rites they condemned; later writers added gods, lost cities and tidy numbers. The record that survives is stranger and more human: alliances of Indigenous peoples deciding the fate of Tenochtitlan, a friar who found his own source's number incredible, knotted cords that no one can yet fully read, and children buried on a mountain more than 6,700 metres high.

Sources

Letters of Cortes, trans. and ed. Francis Augustus MacNutt, 2 vols (Putnam, 1908): Second and Third Letters.
Bernardino de Sahagún, Histoire générale des choses de la Nouvelle-Espagne, trans. D. Jourdanet and Rémi Siméon (G. Masson, 1880).
Jorge Cañizares-Esguerra, review of Matthew Restall, When Montezuma Met Cortés (2018), Not Even Past, University of Texas at Austin (2019).
Matthew Wills, "The Mexica Didn't Believe the Conquistadors Were Gods", JSTOR Daily (2020), on Camilla Townsend, "Burying the White Gods", American Historical Review 108(3) (2003).
Raphael Folsom, review of Laura E. Matthew and Michel R. Oudijk (eds.), Indian Conquistadors (2007), H-Net Reviews (2008).
Camilla Townsend, "Burying the White Gods: New Perspectives on the Conquest of Mexico", American Historical Review 108(3) (2003), pp. 659–687.
Diego Durán, Historia de las Indias de Nueva-España, ed. José F. Ramírez, vol. 1 (Mexico, 1867).
Xavier Noguez, "Sacrificios en el Templo Mayor", Arqueología Mexicana 108 (2011), online extract.
Ximena Chávez Balderas and Lorena Vázquez Vallín, "Mexica (Aztec) Human Sacrifice: New Perspectives", Mexicolore.
INAH, "Huei Tzompantli" (Lugares INAH) and Boletín, 25 February 2023.
Susan Kellogg, "Aztecs: Image and Reality", Fifteen Eighty Four, Cambridge University Press (2024).
Inca Garcilaso de la Vega, First Part of the Royal Commentaries of the Yncas, trans. Clements R. Markham, 2 vols (Hakluyt Society, 1869–71).
Reports on the Discovery of Peru, trans. and ed. Clements R. Markham (Hakluyt Society, 1872): Francisco de Xerez.
Sabine Hyland, "Writing with Twisted Cords", Current Anthropology 58(3) (2017), author's accepted manuscript.
Duke University Press, on Manuel Medrano and Gary Urton, Ethnohistory (2018).
University of Texas Press, catalogue page for James Lockhart, The Men of Cajamarca (1972).
Hiram Bingham, Inca Land (Houghton Mifflin, 1922).
University of Illinois Chicago, "Study reconsiders name of Peru's Machu Picchu" (2022).
Richard L. Burger et al., "New AMS dates for Machu Picchu", Antiquity 95 (2021), abstract.
Cristóbal de Molina, An Account of the Fables and Rites of the Yncas, trans. Clements R. Markham (Hakluyt Society, 1873).
Andrew S. Wilson et al., "Archaeological, radiological, and biological evidence offer insight into Inca child sacrifice", PNAS 110 (2013), abstract and full text; Museo de Arqueología de Alta Montaña, Salta.

Checked September 2026.

AI Benchmarks Explained: What the Scores Measure and Why They Can Mislead

Hacks Vitae — Thu, 17 Sep 2026 00:00:00 GMT

When a company releases an AI model, it usually publishes a table of benchmark scores to show how the model compares with others. Those numbers are useful, but they are easy to over-read. This page explains what the best-known benchmarks test and the specific, documented ways a score can say less than it seems. Scores and policies change quickly; this is a snapshot as of September 2026.

What a benchmark is

A benchmark is a fixed set of tasks with known answers, scored automatically, so that different models can be measured on exactly the same questions.

A classic example is MMLU. Its authors wrote: "The test covers 57 tasks including elementary mathematics, US history, computer science, law, and more." When it was published, they found that "most recent models have near random-chance accuracy".

Tests like this stop being useful once the best models score near the top. The authors of a newer benchmark, Humanity's Last Exam, wrote that "benchmarks are not keeping pace in difficulty", noting that models now score over 90% on popular benchmarks like MMLU. This is called saturation: when nearly every leading model passes, the test can no longer tell them apart.

The benchmarks you will see most often

Benchmark	What it tests	A number worth knowing
GPQA	"448 multiple-choice questions written by domain experts in biology, physics, and chemistry"	Experts with or pursuing PhDs reached 65% (74% after discounting mistakes they spotted afterwards); skilled non-experts reached 34% even with web access.
Humanity's Last Exam (HLE)	Hard academic questions; "HLE consists of 2,500 questions across dozens of subjects, including mathematics, humanities, and the natural sciences."	Each question "has a known solution that is unambiguous and easily verifiable, but cannot be quickly answered via internet retrieval."
SWE-bench	Real software problems: "Given a codebase along with a description of an issue to be resolved, a language model is tasked with editing the codebase to address the issue."	2,294 problems from 12 popular Python repositories; at launch the best model solved 1.96%.
SWE-bench Verified	A smaller set of SWE-bench problems checked by people: "a subset of the original test set from SWE-bench, consisting of 500 samples verified to be non-problematic by our human annotators"	Released by OpenAI with the SWE-bench authors in August 2024. See below for why OpenAI later stopped using it.
ARC-AGI-3	Interactive puzzles: "There are no instructions, no rules, and no stated goals."	At its launch, the organisers reported: "Humans score 100%. Frontier AI scores 0.51%."
METR time horizon	How long a task a model can handle: "the time humans typically take to complete tasks that AI models can complete with 50% success rate"	METR's paper reports this "has been doubling approximately every seven months since 2019, though the trend may have accelerated in 2024". A January 2026 update estimated a doubling time of about 131 days for models since 2023.

Why a score can mislead

1. The test questions leak into training data

AI models are trained on huge amounts of text gathered from the internet, and benchmark questions are often published there too. Researchers call this data contamination: "the unintended overlap between training and test datasets", which can lead to "an overestimation of the models' true generalization capabilities."

The clearest example comes from a benchmark's own co-creator. In February 2026 OpenAI explained why it had dropped SWE-bench Verified. It said that "all frontier models we tested were able to reproduce the original, human-written bug fix used as the ground-truth reference, known as the gold patch, or verbatim problem statement specifics for certain tasks" — in other words, each model had seen at least some of the problems and their solutions during training. It concluded: "This is why we have stopped reporting SWE-bench Verified scores, and we recommend that other model developers do so too."

2. The answer key can be wrong

The same OpenAI analysis found a second problem. It "audited a 27.6% subset of the dataset that models often failed to solve and found that at least 59.4% of the audited problems have flawed test cases that reject functionally correct submissions".

That figure applies to the audited problems, not to the whole benchmark. But it shows how a model can be marked wrong for a correct answer — and why a score near the top of a test may say more about the test than about the model.

3. The way the test is run changes the result

OpenAI put it plainly in July 2026: "Benchmarks rarely measure AI models in isolation." A model runs inside a harness — the software that feeds it the task, keeps track of its progress and collects its answer — and often at a chosen level of reasoning effort.

On 3 September 2026, ARC Prize published its results for OpenAI's GPT-6 Astra on the Semi-Private test set of ARC-AGI-3. With its Standard harness the model scored 62.7% at maximum effort and 54.8% at high effort. With a Provider Adapter harness — which "preserves opaque reasoning state between requests and uses compaction for longer conversations, allowing the model to reuse prior work" — it scored 99.9% at high effort and 98.6% at maximum effort. The organisers' summary put two of those runs side by side: "62.7% for $26K" and "99.9% for $19K".

Same model, same benchmark, very different scores. A benchmark number without its setup is only half a fact.

4. The tasks may not be the tasks you care about

METR, which publishes the time-horizon measurements, lists the limits of its own benchmark. It says the time horizon is "a measure of the difficulty of a task, rather than the time an AI spends to complete the task." It says: "Our task distribution is primarily composed of software engineering, machine learning, or cybersecurity tasks." It warns: "In other words, our tasks are much “cleaner” than real economically valuable labor." And its chart notes: "Measurements above 16 hrs are unreliable with our current task suite".

A high score on a clean, well-specified software task does not show how a model will do on a messy task in a different field.

5. Who gets to test before release

Some leaderboards rank models by public votes rather than fixed questions. The best known is Arena (formerly Chatbot Arena, then LMArena), where people compare answers from two anonymous models and pick the better one. A 2025 paper, "The Leaderboard Illusion", argued that its rules gave some developers an advantage. Its operators, then called LMArena, published a response disputing several of its claims.

DISPUTED. The two sides agree that developers tested private versions of models before release. They disagree about whether that practice was undisclosed or favoured some developers, and about how much it distorted the rankings.

The paper reported that "undisclosed private testing practices benefit a handful of providers who are able to test multiple variants before public release and retract scores if desired". It estimated that two large providers received 19.2% and 20.4% of all the arena's data, while "a combined 83 open-weight models have only received an estimated 29.7% of the total data", and reported "relative performance gains of up to 112% on ArenaHard, a test set from the arena distribution", from training on arena data.
LMArena replied that its testing policy had been public since March 2024 and that "any model provider can submit as many public and private variants as they would like, as long as we have capacity for it", while also saying that a future policy release would "explicitly state that model providers are all allowed to test multiple variants of their models pre-release"; that ArenaHard is "a static benchmark with 500 data points that uses an LLM judge, and no human labels", and so not representative of the arena itself; and that its analysis put the effect of pre-release testing at "around +11 Elo after 50 tests and 3000 votes", falling towards zero as new votes arrive.

Arena's current policy, last updated in September 2026, says that if a model was tested before release, its score is marked "as preliminary until enough fresh votes have been collected after the model's public release".

The evidence does not settle these questions. The safe reading is that a public-vote ranking is one signal among several, not a final verdict.

6. A single "intelligence" number is a choice of ingredients

Some trackers combine many benchmarks into one index. Artificial Analysis says its "Intelligence Index v4.3 incorporates 10 evaluations", and adds: "Like all evaluation metrics, it has limitations and may not apply directly to every use case." Its methodology page now lists GPQA Diamond and MMLU-Pro under "Legacy Evaluations".

Change the ingredients and the ranking can change. A combined score is useful for a quick comparison, but it reflects the tracker's choice of which tests count.

The old rule behind all of this

The social anthropologist Marilyn Strathern wrote in 1997: "When a measure becomes a target, it ceases to be a good measure." She drew on the work of an educationalist, Hoskin: "Hoskin describes this as 'Goodhart's law', after the latter's observation on instruments for monetary control". Strathern was writing about exam grades in British universities, not AI — but the pattern is the same. Once a benchmark becomes the number everyone competes on, it becomes easier to improve the number than the ability it was meant to measure.

How to read a benchmark claim in two minutes

Which benchmark, and which version? SWE-bench and SWE-bench Verified are different tests; so are ARC-AGI-2 and ARC-AGI-3.
Who ran it? The model's developer, the benchmark's organisers, or an independent tracker. Results gathered by others are worth comparing: Epoch AI's hub, for example, "includes results from benchmarks administered internally by Epoch AI as well as data collected from external sources."
With what setup? The harness and the reasoning effort can change the result.
At what cost? Two scores reached at very different cost are not the same achievement.
Is the test saturated? If every leading model scores near the top, small differences mean little.
Could it be contaminated? OpenAI noted that SWE-bench Verified was "open-source and broadly used and discussed, which makes avoiding contamination difficult". Widely published tests carry that risk.
Are the tasks like yours? A coding benchmark tells you about coding, not about writing or research.

Our reading: benchmarks are the best shared measuring stick the field has, and they do track real progress. But a single score is a result under particular conditions, not a general measure of how capable a model is. Read the conditions before the number.

Sources

Hendrycks et al., "Measuring Massive Multitask Language Understanding" (MMLU), arXiv:2009.03300.
Rein et al., "GPQA: A Graduate-Level Google-Proof Q&A Benchmark", arXiv:2311.12022.
Phan et al., "Humanity's Last Exam", arXiv:2501.14249.
Jimenez et al., "SWE-bench: Can Language Models Resolve Real-World GitHub Issues?", arXiv:2310.06770.
OpenAI, "Introducing SWE-bench Verified" (13 August 2024); "Why SWE-bench Verified no longer measures frontier coding capabilities" (23 February 2026).
ARC Prize, "Announcing ARC-AGI-3" (25 March 2026); "OpenAI's GPT-6 Astra on ARC-AGI-3" (3 September 2026).
Kwa et al., "Measuring AI Ability to Complete Long Software Tasks", arXiv:2503.14499; METR, "Time horizons" page; METR, "Time Horizon 1.1" (January 2026).
"A Survey on Data Contamination for Large Language Models", arXiv:2502.14425.
Singh et al., "The Leaderboard Illusion", arXiv:2504.20879; LMArena, response to "The Leaderboard Illusion", arena.ai.
OpenAI, "How enabling two settings tripled our scores on the ARC-AGI-3 benchmark" (29 July 2026).
Arena, Leaderboard Policy (last updated 1 September 2026), arena.ai.
Artificial Analysis, Intelligence Index methodology; Epoch AI, Benchmarking Hub.
Strathern, M., "'Improving ratings': audit in the British University system", European Review 5(3), 1997, p. 308.

Checked September 2026.

AI Music Generators and the Lawsuits, Explained: What Was Filed, Settled and Still Open

Hacks Vitae — Thu, 17 Sep 2026 00:00:00 GMT

AI music generators such as Suno and Udio can produce a full song from a text prompt. In 2024 the major record labels took both companies to court. Since then some labels have settled and signed licensing deals, other claims are still being fought, and new cases have started in the US, in Europe and in Canada.

This page sets out what happened, in order, using the court filings and the companies' own announcements. It is an explainer, not legal advice. A complaint contains allegations, an answer contains a company's position, and only orders and judgments are the court's own words. Court records were read on 25 September 2026; the online docket service used warns that its copies "may not be up to date".

The short version

Case	Where	Filed	Status (25 Sep 2026)
Record labels v. Suno	US federal court, Massachusetts	24 June 2024	Open; Warner settled
Record labels v. Udio	US federal court, New York	24 June 2024	Open for Sony; Universal and Warner settled
Sony and affiliated labels v. Udio (second case)	US federal court, New York	20 July 2026	Open
Universal, Sony and affiliated labels v. Suno (second case)	US federal court, Massachusetts	18 September 2026	Open
GEMA v. Suno	Munich Regional Court, Germany	21 January 2025	GEMA largely won at first instance; not final
Koda v. Suno	Copenhagen City Court, Denmark	4 November 2025	No later update found
SOCAN v. Suno	Federal Court of Canada	2 September 2026	Filed; no later update found

1. The 2024 lawsuits

On 24 June 2024 the Recording Industry Association of America announced two suits brought by record companies: one against Suno, "filed in the United States District Court for the District of Massachusetts", and one against Udio (the company is called Uncharted Labs) in New York. The plaintiffs were labels belonging to Universal Music Group, Sony Music and Warner Music Group.

The complaints allege that the services copied the labels' recordings to train their AI models. The Suno complaint opens its case this way: "There is nothing that exempts AI technology from copyright law or that excuses AI companies from playing by the rules."

Both complaints ask for damages "including in an amount up to $150,000 per work infringed". That is the most US law allows in statutory damages per work, and only where a court finds the infringement was wilful; otherwise the limit is $30,000. The complaints do not state a total, and no US court has awarded anything.

2. What Suno and Udio said

Both companies filed their answers on 1 August 2024, and both argue that what they did is fair use under US copyright law.

Suno said its program "was constructed by showing the program tens of millions of instances of different kinds of recordings", and that those recordings "presumably included" recordings owned by the labels. That is a statement about training data, not an admission of infringement. Suno's answer adds: "But what the major record labels really don’t want is competition." In a blog post the same day, Suno wrote: "learning is not infringing. It never has been, and it is not now."
Udio said it "will prevail in this litigation because decades of judicial precedent establishes that no company controls a genre or style of music", and called its use "a quintessential “fair use” under copyright law".

3. The YouTube question

During the cases, a second issue emerged: where the training audio came from.

Both companies now say in their court filings that they took audio from YouTube. Suno's 2026 answer states: "Suno further admits that it obtained audio data from YouTube for use as training data." Udio's says: "Udio admits that it obtained audio data from YouTube for use as training data".

The labels argue this broke a separate law against getting around technical protections. The Suno court described what the labels learned in discovery: "plaintiffs learned that Suno obtained the recordings by downloading them from YouTube." The labels' amended complaint calls it "stream ripping".

The US courts have allowed those claims to go ahead, but that is not a finding that anyone broke the law. In the Udio case, the judge wrote that whether YouTube's measures count as protections under the law "requires a greater factual record than the pleadings contain". In the Suno case, the court found the labels had made "a plausible claim" and let them add it.

4. The settlements and licensing deals

Universal Music Group and Udio announced on 29 October 2025 that they had "settled copyright infringement litigation" and would launch a new platform "trained on authorized and licensed music". The same day Udio told its users: "Starting today, downloads from the platform will be unavailable".

Warner Music Group and Udio announced on 19 November 2025 a deal that "resolves the companies' copyright infringement litigation". Udio added that the labels did not buy the company: "Did the labels buy Udio / do they own Udio? No, they did not."

Warner Music Group and Suno announced on 25 November 2025 a partnership that "also settles previous litigation between the companies". Warner said that when Suno's new licensed models launch, "the current models will be deprecated", that "songs made on the free tier will not be downloadable", and that "Suno has also acquired Songkick", the concert-discovery service.

None of the settlements disclosed their financial terms.

Suno has since announced a partnership with the music company BMG (12 August 2026) and, on 9 September 2026, its v6 models, "developed with our industry partners, including Warner Music Group, BMG and Believe". Suno's announcement does not say whether v6 was trained only on licensed music. Universal and Sony's second suit (section 5) alleges that Suno developed v6 using their recordings.

We found no announced deal between Sony Music and either company.

5. What is still open

Universal, Capitol and Sony v. Suno. Warner's claims were dismissed after its deal. The remaining labels filed an amended complaint on 25 August 2026, and Suno answered on 1 September 2026, again arguing fair use: "This is quintessential fair use." On 18 August 2026 the judge denied, without prejudice, a request to add 61,026 more recordings to this case, noting that fact discovery had been extended to "September 30, 2026". The judge also wrote that summary judgment on Suno's fair use defence "will likely resolve the predominant issue" in the case. On 22 September 2026 the judge granted a joint request to extend discovery on the labels' claim about getting around YouTube's protections (section 3) to 20 November 2026; the amended schedule lists 22 June 2027 for dispositive motions.

Universal, Sony and affiliated labels v. Suno (a second case). On 18 September 2026 the labels filed a new suit in the same court over "60,202 sound recordings", describing it as "the second action" against Suno. The complaint alleges that the infringement continues in Suno's v6 models: "On information and belief, Suno developed v6 using Plaintiffs' copyrighted sound recordings". Suno had not answered when the docket was read.

Sony v. Udio (the original case). After Universal and Warner settled, the court noted that "the number of Plaintiffs has been reduced from 12 to 3, and the number of works at issue from 1,670 to 333." The court also refused to add about 30,000 more recordings, saying "there is no requirement that it be done in this lawsuit."

Sony v. Udio (a second case). Sony and affiliated labels then filed a new suit on 20 July 2026 over "30,117 sound recordings". Udio answered on 27 August 2026: "This is quintessential fair use." (The earlier court order had described Sony's proposal as "adding 30,442 works"; the documents do not explain the difference.)

Independent artists, publishers and others have also filed separate suits against Suno or Udio since 2025. We have not read those complaints, so we do not describe them here.

6. Outside the US

Germany: GEMA v. Suno. GEMA, the German music-rights society, sued Suno in Munich on 21 January 2025. On 31 July 2026 the court "largely granted" (überwiegend stattgegeben) GEMA's claims for an injunction, information and damages over six well-known songs (for one, only its chorus). The claims covered copies made during training in the US as well as the model and its outputs in Germany. Applying US law to the training copies, the court found they were not covered by fair use, because the works reappeared in the outputs (nicht von der fair use Doktrin nach 17 U.S.C. § 107 gedeckt). The court's press office noted that the case did not cover song lyrics, and that "The judgment is not final" (Das Urteil ist nicht rechtskräftig).

Germany: GEMA v. OpenAI. The same Munich court ruled on 11 November 2025 in a separate case about song lyrics reproduced by a chatbot, largely in GEMA's favour. That case is about lyrics, not an AI music generator. According to GEMA, OpenAI has appealed to the Higher Regional Court in Munich (Oberlandesgericht München).

Denmark: Koda v. Suno. Koda, the Danish music-rights society, filed suit against Suno on 4 November 2025, describing it as "the first time a Danish rights organization has filed a lawsuit against an AI service." We found no later ruling.

Canada: SOCAN v. Suno. SOCAN, Canada's largest member-owned music rights organization, filed a statement of claim against Suno in the Federal Court of Canada on 2 September 2026. According to SOCAN, the claim alleges that by making the platform available in Canada and generating and streaming outputs that replicate songs in its repertoire, "Suno has infringed SOCAN’s performing rights in the underlying songs", and it lists "a sample of 150 publicly available Suno outputs".

7. The US Copyright Office's view

On 9 May 2025 the US Copyright Office released a pre-publication version of a report on generative AI training. It concluded that "Various uses of copyrighted works in AI training are likely to be transformative", but that "making commercial use of vast troves of copyrighted works to produce expressive content that competes with them in existing markets, especially where this is accomplished through illegal access, goes beyond established fair use boundaries."

The report is the Office's view. It does not bind the courts, and it does not analyse or decide these cases (it cites the Suno complaint once, in a footnote).

8. What the terms say for users now

Suno (terms effective 3 September 2026): for paid plans, Suno "assigns to you all of its right, title and interest" in outputs it owns, but "makes no representation or warranty to you that any copyright will vest in any Output." Free-tier users may use outputs only "for your lawful, personal and non-commercial purposes". Suno's policy post lists "Free: Up to 7 (lifetime) trial downloads", "Pro: 20 downloads per month" and "Premier: 60 downloads per month".

Udio (terms last revised 12 November 2025): "You may not download copies of any Output", and "you may use Output solely for your personal and non-commercial purposes". Udio's terms say the company and its licensors own the output.

Terms change. If you use either service, read the current terms on its own site.

What to watch

A summary-judgment ruling on fair use in the Suno case, which the judge said "will likely resolve the predominant issue" in the case.
The next hearing in the original Udio case, a status conference set for 23 October 2026.
Suno's response to the labels' second suit and to SOCAN's claim in Canada.
Whether Udio launches the licensed platform it announced for 2026, and whether Sony reaches a deal.
Whether Suno appeals the Munich judgment, and the outcome of OpenAI's appeal in the lyrics case.

Sources

RIAA, announcement of the Suno and Udio suits (24 June 2024).
UMG Recordings, Inc. v. Suno, Inc., D. Mass. No. 1:24-cv-11611: complaint; Suno's answers (1 August 2024; 1 September 2026); orders of 18 August 2026 and 22 September 2026; amended complaint (25 August 2026). Via CourtListener.
UMG Recordings, Inc. v. Uncharted Labs, Inc. (now Sony Music Entertainment v. Uncharted Labs), S.D.N.Y. No. 1:24-cv-04777: complaint; Udio's answers (1 August 2024; 29 April 2026); orders of April, 29 June and 24 August 2026.
Sony Music Entertainment v. Uncharted Labs, Inc., S.D.N.Y. No. 1:26-cv-06120: complaint (20 July 2026); answer (27 August 2026).
UMG Recordings, Inc. v. Suno, Inc., D. Mass. No. 1:26-cv-14275: complaint (18 September 2026).
Suno, "The Future of Music" (1 August 2024); WMG partnership (25 November 2025); BMG partnership (12 August 2026); "Introducing v6" (9 September 2026); terms of service (effective 3 September 2026) and policy update (August 2026).
Universal Music Group and Udio (29 October 2025); Warner Music Group and Udio (19 November 2025); Warner Music Group and Suno (25 November 2025), press releases. Udio blog posts (29 October and 19 November 2025); Udio terms of service.
Landgericht München I, press releases no. 11 (11 November 2025) and no. 16 (31 July 2026); GEMA (31 July 2026).
Koda, "Koda sues US tech company Suno" (4 November 2025).
SOCAN, announcement of its lawsuit against Suno (2 September 2026) and statement of claim, Federal Court of Canada file T-3803-26.
US Copyright Office, Copyright and Artificial Intelligence, Part 3: Generative AI Training, pre-publication version (9 May 2025).

Checked September 2026.

AI Myths Checked: 10 Things People Believe About AI Chatbots

Hacks Vitae — Thu, 17 Sep 2026 00:00:00 GMT

A lot of what circulates about AI chatbots comes from headlines and social posts rather than from the companies' documentation or the research papers. Below are ten common beliefs, each checked against the original source. Some are wrong, some are half right, and a few are genuinely unsettled. This is a snapshot as of September 2026.

How to read the verdicts: MYTH means the sources contradict the belief. PARTLY TRUE means there is something to it, but the usual version overstates it. UNVERIFIED means no primary source settles it. DISPUTED means credible sources disagree.

1. "AI chatbots learn from your conversation in real time"

PARTLY TRUE. The model itself does not change while you talk to it. Its knowledge comes from training that happened before you started chatting, and that takes a long time: Anthropic wrote in August 2025 that "models released today began development 18 to 24 months ago".

But many chat apps now have a memory feature that carries details from your chats into later conversations. OpenAI says memory, when enabled, "helps ChatGPT automatically remember useful context from your chats, files, and connected apps to personalize your experience". Anthropic says Claude "can also remember context from your chats and carry it into new conversations", and that "Memory is on by default for Free, Pro, and Max plans". That is saved context, not retraining.

Separately, your chats may be used to train later models, depending on the company and your settings.

Anthropic (consumer accounts, August 2025): "We will train new models using data from Free, Pro, and Max accounts when this setting is on". It said the change does not apply to services under its Commercial Terms, including API use.
OpenAI (policy page updated March 2026): "ChatGPT, for instance, improves by further training on the conversations people have with it, unless you opt out". For business products it says: "By default, we do not train on any inputs or outputs from our products for business users, including ChatGPT Team, ChatGPT Enterprise, and the API."

Check the data settings of the app you use.

2. "GPT-4 has 1.8 trillion parameters"

UNVERIFIED. OpenAI never published GPT-4's size. Its technical report says it "contains no further details about the architecture (including model size), hardware, training compute, dataset construction, training method, or similar". We found no primary source for the 1.8 trillion figure.

3. "OpenAI invented the Transformer"

MYTH. The Transformer — the design behind today's large language models — was introduced in the 2017 paper "Attention Is All You Need". Its authors list affiliations with Google Brain and Google Research, and one with the University of Toronto. OpenAI built on it: its GPT-4 report describes GPT-4 as "a Transformer-based model".

4. "AI chatbots look things up on the web when they answer"

PARTLY TRUE. It depends on the product. OpenAI's help centre says: "ChatGPT may search the web automatically when your question would benefit from current information." In developer APIs, search is a tool that has to be switched on. OpenAI's API guide says: "To enable this, use the web search tool in the Responses API or, in some cases, Chat Completions." Anthropic describes its version as a tool that "gives Claude direct access to real-time web content, allowing it to answer questions with up-to-date information beyond its knowledge cutoff."

When no search happens, the answer comes from what the model learned in training and from what is already in the conversation.

5. "AI is conscious"

UNVERIFIED. There is no established evidence that current AI systems are conscious, and no agreed test. A 2023 report by a group of researchers assessed AI systems against "indicator properties" drawn from scientific theories of consciousness. It concluded: "Our analysis suggests that no current AI systems are conscious". It also said "there are no obvious technical barriers to building AI systems which satisfy these indicators".

That assessment covered systems available in 2023. A 2025 follow-up led by the same first author noted that for tests of consciousness "it remains unclear how they should (or even could) be validated".

6. "AI detectors can reliably tell if text was written by AI"

PARTLY TRUE. OpenAI released its own AI-text classifier in January 2023 and then withdrew it: "As of July 20, 2023, the AI classifier is no longer available due to its low rate of accuracy." In its own evaluation, it "correctly identifies 26% of AI-written text (true positives) as “likely AI-written,” while incorrectly labeling human-written text as AI-written 9% of the time".

A 2023 study tested "seven widely-used GPT detectors on a corpus of 91 human-authored TOEFL essays" from a Chinese educational forum — essays written by people taking an English test. The detectors "consistently misclassify non-native English writing samples as AI-generated", with an average false positive rate of 61.22%, and "89 of the 91 TOEFL essays (97.80%) are flagged as AI-generated by at least one detector".

Newer detectors do much better. Nature reported in August 2026 that current leading tools "correctly flag solely human-written content as human almost all the time", but that "they still sometimes make mistakes, so the tools can be used only as starting points for investigation", and that their results are less illuminating for AI-edited writing. A detector's verdict alone is still not proof that someone used AI.

7. "The Turing test has been passed"

PARTLY TRUE. In a study published in PNAS in May 2026 (first posted as a preprint in March 2025), two researchers ran a three-party version of the test. "Participants had 5 min conversations simultaneously with another human participant and one of these systems". "When prompted to adopt a humanlike persona, GPT-4.5 was judged to be the human 73% of the time", while the baseline systems, ELIZA and GPT-4o, were judged to be the human only 23% and 21% of the time. The published paper concludes: "The results constitute empirical evidence that artificial systems can pass a standard three-party Turing test."

The limits matter. Without the persona prompt, the same models "performed significantly worse (38% and 36%)". A third, 15-minute study, using LLaMa-3.1 and GPT-5 because GPT-4.5 had been retired, found both persona-prompted models still passed (56% and 59%). And passing shows only that a system can be mistaken for a person in conversation. The authors themselves say the results bear on debates about "what kind of intelligence is exhibited by large language models" — they do not claim the question is settled.

8. "Every AI prompt uses a lot of water"

DISPUTED. Published estimates differ by more than a hundredfold, because they measure different things:

Estimate	What it covers	Figure
Academic study (Li et al.)	GPT-3 in Microsoft data centres in the US and other countries (estimate); includes water used to generate the electricity	"a 500ml bottle of water for roughly 10 – 50 medium-length responses, depending on when and where it is deployed"
Google (own measurement, 2025)	Median text prompt in its Gemini app; water used for cooling in its data centres	"the equivalent of five drops of water (0.26 mL)"
Mistral AI (own life-cycle analysis, 2025)	A 400-token response from its Le Chat assistant	"45 mL of water", including "upstream emissions" such as manufacturing servers

The models, the years and the boundaries all differ, and the company figures are self-reported. In short, water use per prompt depends heavily on what is counted, and these numbers cannot be compared directly.

9. "Bigger AI models are always better"

MYTH. Size helps, but it is not the whole story. OpenAI's InstructGPT paper stated: "Making language models bigger does not inherently make them better at following a user's intent." In its human evaluations, "outputs from the 1.3B parameter InstructGPT model are preferred to outputs from the 175B GPT-3, despite having 100x fewer parameters".

A 2022 study found that "current large language models are significantly undertrained", and its smaller, better-trained model, Chinchilla, "uniformly and significantly outperforms Gopher (280B), GPT-3 (175B), Jurassic-1 (178B), and Megatron-Turing NLG (530B)". The authors concluded that, for compute-optimal training, model size and the amount of training data should be scaled equally.

10. "Being polite to a chatbot gets you better answers"

DISPUTED. The studies point in different directions:

A 2024 study across English, Chinese and Japanese found "impolite prompts often result in poor performance, but overly polite language does not guarantee better outcomes" and that "The best politeness level is different according to the language."
A 2025 report found "sometimes being polite to the LLM helps performance, and sometimes it lowers performance".
A short 2025 paper using one model and 50 base questions found "impolite prompts consistently outperformed polite ones, with accuracy ranging from 80.8% for Very Polite prompts to 84.8% for Very Rude prompts".

Tone can change results on some tasks, but the effect is inconsistent across studies, models and languages.

What these have in common

The product is not the model. Apps add search, settings and data policies on top of the model, so "the AI does X" often depends on which app and which settings.
Dates matter. Several of these answers describe 2023–2025 systems and may change with new models.
The original source is usually more careful than the headline. Most of the beliefs above are a simplified version of something a paper or policy page says with conditions attached.

Sources

Anthropic, "Updates to Consumer Terms and Privacy Policy" (28 August 2025).
OpenAI, "How your data is used to improve model performance" (updated 13 March 2026).
OpenAI, "GPT-4 Technical Report", arXiv:2303.08774.
Vaswani et al., "Attention Is All You Need", arXiv:1706.03762.
OpenAI Help Center, "Searching the web with ChatGPT" and "Memory FAQ"; Claude Help Center, chat search and memory; OpenAI API docs, "Web search"; Anthropic docs, "Web search tool".
Butlin, Long et al., "Consciousness in Artificial Intelligence: Insights from the Science of Consciousness", arXiv:2308.08708; Butlin et al., "Identifying indicators of consciousness in AI systems", Trends in Cognitive Sciences (2025).
OpenAI, "New AI classifier for indicating AI-written text" (31 January 2023, with July 2023 update).
Liang et al., "GPT detectors are biased against non-native English writers", arXiv:2304.02819; Nature, "AI-detection tools have made huge leaps forward — how good are they?" (25 August 2026).
Jones and Bergen, "Large language models pass a standard three-party Turing test", PNAS 123(21), e2524472123 (2026); preprint arXiv:2503.23674.
Li et al., "Making AI Less 'Thirsty'", arXiv:2304.03271; Elsworth et al., "Measuring the environmental impact of delivering AI at Google Scale", arXiv:2508.15734; Mistral AI, "Our contribution to a global environmental standard for AI" (22 July 2025).
Ouyang et al., "Training language models to follow instructions with human feedback", arXiv:2203.02155; Hoffmann et al., "Training Compute-Optimal Large Language Models", arXiv:2203.15556.
Yin et al., "Should We Respect LLMs?", arXiv:2402.14531; Meincke et al., "Prompting Science Report 1", arXiv:2503.04818; Dobariya and Kumar, "Mind Your Tone", arXiv:2510.04950.

Checked September 2026.

AI Glossary: 22 Terms Explained in Plain Words

Hacks Vitae — Thu, 17 Sep 2026 00:00:00 GMT

AI news is full of jargon. This glossary explains 22 common terms in plain words. Most entries quote a short definition from a named source — usually Google's machine learning glossary, an AI company's own documentation, a research paper or the Open Source Initiative — followed by what it means in practice. Usage changes quickly; this reflects September 2026.

For the bigger picture, see how large language models work.

The basics

1. Large language model (LLM)

Google's glossary: "At a minimum, a language model having a very high number of parameters." It adds that, more informally, the term covers "any Transformer-based language model, such as Gemini or GPT".

In practice: the AI system behind a chatbot. The chat app is the product; the model is the engine inside it. Anthropic, for example, describes its assistant Claude as one that "is based on a large language model that has been fine-tuned and trained using RLHF".

2. Token

Google's glossary: "In a language model, the atomic unit that the model is training on and making predictions on."

In practice: models read and write in tokens, and usage limits and prices are counted in them. How much text fits in a token varies. OpenAI's help centre says "1 token is approximately 4 characters" for English text, while Anthropic's glossary says that for Claude "a token approximately represents 3.5 English characters, though the exact number can vary depending on the language used." Anthropic has also said that Claude Opus 4.7's updated tokenizer can turn the same text into "roughly 1.0–1.35×" as many tokens as Claude Opus 4.6, depending on the content type, so any such figure is approximate.

3. Context window

Google's glossary: "The number of tokens a model can process in a given prompt."

In practice: how much text — your question, pasted documents, the conversation so far and the model's own reply — the model can handle at once.

4. Parameters

Google's glossary: "The weights and biases that a model learns during training."

In practice: the internal numbers that training adjusts. A bigger parameter count means a bigger model, but not automatically a better one.

5. Mixture of experts (MoE)

Google's glossary: "A scheme to increase neural network efficiency by using only a subset of its parameters (known as an expert) to process a given input token or example."

In practice: this is why some models quote two sizes. DeepSeek's V3 report, for example, describes a model "with 671B total parameters with 37B activated for each token".

6. Training compute (FLOP)

Epoch AI: "Compute usage is commonly measured as the number of floating point operations (FLOP) required to train the final version of the system".

In practice: a count of the calculations used to train a model. Some laws use it as a threshold — New York's definition of a "frontier model" uses 10^26 operations; see what is a frontier AI model?

Training and using a model

7. Pretraining

Google's glossary: "The initial training of a model on a large dataset."

In practice: the first stage of training, on a very large amount of text. Google adds that "Some pre-trained models are clumsy giants and must typically be refined through additional training."

8. Fine-tuning

Google's glossary: "A second, task-specific training pass performed on a pre-trained model to refine its parameters for a specific use case."

In practice: extra training that shapes a general model for a purpose. One common kind is instruction tuning, which Google describes as "A form of fine-tuning that improves a generative AI model's ability to follow instructions."

9. RLHF (reinforcement learning from human feedback)

Google's glossary: "Using feedback from human raters to improve the quality of a model's responses."

In practice: people compare or rank model answers, and the model is trained towards the preferred ones. OpenAI's InstructGPT paper, which combined supervised fine-tuning with RLHF, reported that "outputs from the 1.3B parameter InstructGPT model are preferred to outputs from the 175B GPT-3, despite having 100x fewer parameters" in its human evaluations.

10. Inference

Google's glossary: "In large language models, inference is the process of using a trained model to generate a response to an input prompt."

In practice: using the model, as opposed to training it. Google notes the word "has a somewhat different meaning in statistics".

11. Temperature

Google's glossary: "A hyperparameter that controls the degree of randomness of a model's output."

In practice: one reason the same question can get different answers. Some newer models no longer let developers set it.

12. Reasoning model

OpenAI: "Reasoning models use internal reasoning tokens before producing a response." Google uses the word "thinking": "When you use a thinking model, Gemini reasons internally before responding."

In practice: models that work through a problem before answering. Companies use different names for the same idea, and the extra thinking uses tokens.

What models can do

13. Multimodal

Google's glossary: "A model whose inputs, outputs, or both include more than one modality."

In practice: a model that handles more than text, such as images, audio or video.

14. Tool use (function calling)

Anthropic: "Tool use (also called function calling) lets Claude call functions that you define or that Anthropic provides." The model "returns a structured call that your application executes".

In practice: the model does not run the tool itself. It asks for a tool — a search, a calculator, a database lookup — and software carries out the request.

15. Agent

There is no single definition. Google's glossary: "Software that can reason about user inputs in order to plan and execute actions on behalf of the user." Anthropic draws a line between workflows, where "LLMs and tools are orchestrated through predefined code paths", and agents, "systems where LLMs dynamically direct their own processes and tool usage, maintaining control over how they accomplish tasks".

In practice: a system that takes several steps on its own, often using tools, rather than giving one answer.

16. Hallucination

Google's glossary: "The production of plausible-seeming but factually incorrect output by a generative AI model that purports to be making an assertion about the real world."

In practice: a plausible-sounding answer that is wrong. The word is debated — see why AI makes things up.

17. Knowledge cutoff

In practice: the point after which a model's training data stops, so it may not know about later events unless it can search. The term can refer to more than one date. Anthropic, for example, gave two dates for its now-retired Claude Opus 4 and Claude Sonnet 4: a "training data cutoff date of March 2025" and "a reliable knowledge cutoff date of January 2025".

Openness and documentation

18. Open weights

The Open Source Initiative: "Open Weights refer to the final weights and biases of a trained neural network."

In practice: the trained weights are published, so you can download and run the model yourself, under whatever licence comes with them. The same page notes that "Open Weights differ significantly from Open Source AI".

19. Open-source AI

The Open Source Initiative's definition: "An Open Source AI is an AI system made available under terms and in a way that grant the freedoms" to use, study, modify and share it. It requires detailed information about the training data, "The complete source code used to train and run the system", and the model parameters.

In practice: a much higher bar than downloadable weights. See open-weight is not open-source.

20. Model card

From the 2018 paper that proposed them: "Model cards are short documents accompanying trained machine learning models that provide benchmarked evaluation in a variety of conditions".

In practice: a document describing a model, how it was evaluated and the context it is intended for.

21. System card

There is no standard definition. Anthropic describes its own: "System cards document the capabilities, safety evaluations, and responsible deployment decisions for Claude models."

In practice: AI companies use the term for a longer report published with a model, often focused on safety testing. The name has also meant other things: Meta used it in February 2022 for "a new resource for understanding how AI systems work".

22. Benchmark

Google's glossary defines LLM evaluations as "A set of metrics and benchmarks for assessing the performance of large language models".

In practice: a standard set of questions or tasks, scored the same way for every model, used to compare models on the same test. Scores depend on how the test is run — see AI benchmarks explained.

Sources

Google for Developers, Machine Learning Glossary. Definitions quoted from it are © Google, licensed under CC BY 4.0.
Anthropic, Claude Platform Docs: Glossary; Tool use overview; Anthropic, "Building effective agents" (December 2024); Transparency Hub; System cards.
OpenAI, API docs: Reasoning models; Help Center, "Understanding and counting tokens".
Anthropic, "Introducing Claude Opus 4.7"; Meta, "System Cards, a new resource for understanding how AI systems work" (23 February 2022).
New York Senate Bill S8828 (2026), frontier model definition.
Google, Gemini API docs: Thinking.
DeepSeek-AI, "DeepSeek-V3 Technical Report", arXiv:2412.19437.
Epoch AI, "Estimating Training Compute of Deep Learning Models" (2022).
Ouyang et al., "Training language models to follow instructions with human feedback", arXiv:2203.02155.
Open Source Initiative, "Open Weights"; "The Open Source AI Definition – 1.0".
Mitchell et al., "Model Cards for Model Reporting", arXiv:1810.03993.

Checked September 2026.

Algebra, Zero and the House of Wisdom: 8 Claims About the Islamic Golden Age, Checked

Hacks Vitae — Thu, 17 Sep 2026 00:00:00 GMT

From the 8th century on, scholars working in Arabic under Islamic rule translated Greek, Persian and Indian science and added a great deal of their own. That much is not in doubt. But the popular version of the story has grown some claims the evidence does not carry. Here are eight, checked against the medieval texts and the historians who study them.

1. "The word algebra comes from al-Khwarizmi's book"

TRUE. Muhammad ibn Musa al-Khwarizmi (born about 780, died about 850, according to the MacTutor History of Mathematics at the University of St Andrews) wrote in Baghdad under the caliph al-Ma'mun. In his own preface, in Frederic Rosen's 1831 translation, he says the caliph's support "has encouraged me to compose a short work on Calculating by (the rules of) Completion and Reduction", covering what people "constantly require in cases of inheritance, legacies, partition, law-suits, and trade".

"Completion and Reduction" translates al-jabr wa'l-muqabala, and "algebra" comes from the first word. MacTutor explains both terms: al-jabr "means "completion" and is the process of removing negative terms from an equation", and al-muqabala, "balancing", "is the process of reducing positive terms of the same power when they occur on both sides of an equation." When Robert of Chester put the book into Latin in Segovia in 1145, he called it "liber Restaurationis et Oppositionis numeri": the book of restoration and opposition of number (our translation).

2. "The word algorithm comes from his name"

TRUE. MacTutor says simply that "The word algorithm derives from his name." It came through a different book, his arithmetic. According to MacTutor, "The Arabic text is lost", but a medieval Latin version survives. Its opening sentence reads "Dixit algoritmi", "Algoritmi says" (our translation), with Algoritmi standing for al-Khwarizmi's name. Louis Karpinski, who edited the Latin algebra in 1915, traced the word "algorism" to "the use of the name Al-Khowarizmi in the opening sentence of the arithmetic", adding that "the term came into use through Al-Khowarizmi's arithmetic and not his algebra." For centuries "algorism" was the word for arithmetic with Indian numerals.

The Arabic original has still not been found. A 1991 study by André Allard says the oldest Latin adaptations "are the only evidence" for it.

3. "Al-Khwarizmi invented algebra"

PARTLY TRUE. He gave the subject its name and a systematic treatment. He did not invent the solving of equations.

Mathematicians in Babylon worked problems we would now write as quadratic equations around 1800 BC, though MacTutor cautions that "the Babylonians had no notion of 'equation'".
Diophantus of Alexandria, "often known as the 'father of algebra'", used "an abbreviation for the unknown and for the powers of the unknown".
In India, Brahmagupta "gives an, almost modern, method which admits negative quantities."

What was new in al-Khwarizmi was the method. He reduces every problem "to one of six standard forms", and his "mathematics is done entirely in words with no symbols being used." His translator saw the point in 1831: "That he was not the inventor of the Art, is now well established".

How original he was has long been argued. Two views quoted on MacTutor show the range: Rashed writes that "It is impossible to overstress the originality of the conception and style of al-Khwarizmi's algebra", while Crossley says he "may not have been very original".

4. "Arabic numerals were invented by Arabs"

MYTH. The digits came from India. MacTutor says they "were not transmitted directly from India to Europe but rather came first to the Arabic/Islamic peoples and from them to Europe."

In 662, Severus Sebokht, a bishop who lived on the Euphrates, praised the Indians' "valuable methods of calculation", writing, in the English translation quoted on MacTutor, "that this computation is done by means of nine signs."
The digits appear in a European manuscript for the first time in the Codex Vigilanus, "copied by a monk in Spain in 976".
Leonardo of Pisa, known as Fibonacci, learned them as a boy in Bugia, now Bejaia in Algeria. In Liber abaci (1202) he writes, in the translation quoted on MacTutor, of being "introduced to the art of the Indians' nine symbols through remarkable teaching".

Whether al-Khwarizmi wrote the first Arabic book on the Indian numerals is uncertain. MacTutor says "Some degree of mystery still remains."

5. "Zero was invented in the Arab world"

MYTH. The Indian mathematician Brahmagupta, writing in 628, already treated zero as a number. He "defined zero as the result of subtracting a number from itself" and gave rules for it: "a number multiplied by zero becomes zero."

The words, though, did pass through Arabic. David Eugene Smith and Louis Karpinski traced them in 1911. In Sanskrit, zero was sunya, "void", which became Arabic sifr. Fibonacci wrote that the sign 0 "arabice zephirum appellatur", is called zephirum in Arabic (our translation), and Italian forms of that word were "a step from zephirum to zero." And "the English cipher, French chiffre, is derived from the same Arabic word".

6. "The House of Wisdom was a great research academy"

DISPUTED. Older accounts describe the House of Wisdom (Bayt al-Hikma) in Baghdad as an academy where scholars met, taught and translated. Historians who have gone back to the sources are much more cautious.

Dimitri Gutas, in Greek Thought, Arabic Culture (1998), writes: "The fact is that we have exceedingly little historical information about the bayt al-hikma". He points out that the name is "the translation of the Sasanian designation for a library", argues that it was not an academy for teaching, and concludes that "a minimalist interpretation would fit the historical record better." Ekmeleddin İhsanoğlu's The Abbasid House of Wisdom: Between Myth and Reality (2023) treats it as the caliphs' "palace library" and, in its publisher's words, "shows how misconceptions arose because of incorrect translations of Arabic manuscripts".

The evidence leans towards a palace library. Gutas finds that the only translation the sources connect with it was from Persian, and that "the Graeco-Arabic translation movement was completely unrelated to any of the activities of the bayt al-hikma"; under al-Ma'mun, al-Khwarizmi was employed there. The translation movement itself was real and lasted over two centuries, according to Gutas. It did not need a single research institute.

7. "Al-Ghazali killed Islamic science"

MYTH as stated. The theologian al-Ghazali (died 1111) is often blamed for ending scientific inquiry. His own writings are more mixed than the story suggests. In his autobiography, in Claud Field's 1909 translation, he says mathematics "rests on a foundation of proofs which, once known and understood, cannot be refuted", and that "It is therefore a great injury to religion to suppose that the defence of Islam involves the condemnation of the exact sciences." In the same passage he also warns that "those who study mathematics should be checked from going too far in their researches."

His famous condemnation was narrower than science. At the end of The Incoherence of the Philosophers, the Stanford Encyclopedia of Philosophy explains, he declared that "everybody who teaches these three positions publicly is an unbeliever" and "an apostate from Islam". The three positions come from the philosophy of Ibn Sina (Avicenna): that the world has no beginning in time, that God knows only universals and not particulars, and that souls never return to bodies.

Science also carried on after him. The historian George Saliba finds "a very flourishing activity in almost every scientific discipline" in the centuries after him, including the astronomers al-Tusi (died 1274) and Ibn al-Shatir (died 1375), and the physician Ibn al-Nafis (died 1288). He calls this period "the golden age of Islamic astronomy".

The wider question, why the scientific revolution happened in Europe and not in the Islamic world, has long been argued. The physicist Pervez Hoodbhoy, in Islam and Science (1991), argues that al-Ghazali "led the orthodox to final victory". He grants that al-Ghazali's position on mathematics "was not one of blind opposition", but points to another passage in which, in Hoodbhoy's words, al-Ghazali "condemns mathematics with vigour and without reservation". He also lists several other causes, from education and law to economics and politics.

8. "The Mongols ended it all in 1258, and the Tigris ran black with ink"

PARTLY TRUE. The sack of Baghdad was real and terrible. According to the Encyclopaedia Iranica, "The city was taken on 13 February 1258 and subjected to a week of looting and massacre", and "The caliph was executed, together with most of his family."

It did not end science in the region. The conqueror himself, Hülegü, patronised the astronomer al-Tusi, "for whom he built an observatory" at Maragha, and al-Tusi's work there is part of what Saliba calls the golden age of astronomy.

The famous image of the Tigris running black with ink is unverified. We found no medieval source for the ink; whether, and how many, books went into the river is a separate question we did not settle.

Who did the work

The label can mislead in another way. Not all of these scholars were Arab, and not all were Muslim. Hunayn ibn Ishaq was "a Nestorian Christian" and "most important as a translator, making Greek works available"; Thabit ibn Qurra "was a native of Harran and a member of the Sabian sect", and his native language was Syriac; al-Khwarizmi's own name may point to Khwarizm, "south of the Aral Sea in central Asia", though MacTutor notes that the evidence is thin. What they shared was a language of scholarship, Arabic, and patrons who paid for the work: in Gutas's account, "the entire elite of 'Abbasid society", from caliphs and princes to merchants and bankers.

The phrase "golden age of Islam" has a history of its own. A British travel guidebook of 1868 already uses it, as part of a story of rise and decline that historians such as Saliba now question.

Our reading

The history holds up without the legend. A caliph's patronage, a practical book for inheritance and trade, Indian digits passed on through Baghdad and North Africa, and centuries of astronomy after the dates when science was supposed to have died: none of it needs a grand academy or an ink-black river to be remarkable.

Sources

al-Khwarizmi, The Algebra of Mohammed ben Musa, ed. and trans. Frederic Rosen (Oriental Translation Fund, 1831).
Louis Charles Karpinski, Robert of Chester's Latin Translation of the Algebra of Al-Khowarizmi (Macmillan, 1915).
J. J. O'Connor and E. F. Robertson, MacTutor History of Mathematics (University of St Andrews): "Al-Khwarizmi", "The Arabic numeral system", "Brahmagupta", "Fibonacci", "Diophantus of Alexandria", "Quadratic, cubic and quartic equations", "Hunayn ibn Ishaq", "Thabit ibn Qurra".
André Allard, "The Arabic Origins and Development of Latin Algorisms in the Twelfth Century", Arabic Sciences and Philosophy 1.2 (1991), abstract.
David Eugene Smith and Louis Charles Karpinski, The Hindu-Arabic Numerals (Ginn, 1911).
Dimitri Gutas, Greek Thought, Arabic Culture (Routledge, 1998).
Ekmeleddin İhsanoğlu, The Abbasid House of Wisdom: Between Myth and Reality (Routledge, 2023), publisher's description.
al-Ghazali, The Confessions of Al Ghazzali, trans. Claud Field (John Murray, 1909).
Frank Griffel, "al-Ghazali", Stanford Encyclopedia of Philosophy (revised 2026).
George Saliba, Islamic Science and the Making of the European Renaissance (MIT Press, 2007).
Pervez Hoodbhoy, Islam and Science: Religious Orthodoxy and the Battle for Rationality (Zed Books, 1991).
Reuven Amitai, "Hulāgu Khan", Encyclopaedia Iranica.
A Handbook for Travellers in Syria and Palestine (John Murray, 1868).

Checked September 2026.

Ashoka: What His Own Edicts Say, and 8 Popular Claims Checked

Hacks Vitae — Thu, 17 Sep 2026 00:00:00 GMT

Ashoka, the Mauryan emperor of the 3rd century BCE, is one of the few ancient rulers who can still speak to us in his own words. He had edicts carved on rocks and pillars across his empire, many of them in his own voice. He is also the hero of Buddhist chronicles and legends written long after his death: the Sri Lankan Mahavamsa was composed around the start of the sixth century CE, and the Sanskrit Ashokavadana had reached China in translation by about AD 300. The popular story blends the two. Here are eight claims, checked against the edicts, in E. Hultzsch's 1925 edition and translation, and against the legends.

One note on words: Hultzsch translates the key term dhamma as "morality".

1. "The Kalinga war killed 100,000 people"

PARTLY TRUE. The figure is real, but it is the king's own. In Rock Edict XIII he writes: "When king Devanampriya Priyadarsin had been anointed eight years, (the country of) the Kalingas was conquered by (him)." He goes on: "One hundred and fifty thousand in number were the men who were deported thence, one hundred thousand in number were those who were slain there, and many times as many those who died". The popular figure matches the edict's count of the slain; in Hultzsch's translation, the edict claims many more deaths besides, and nothing outside it can check either number.

2. "Seeing Kalinga turned Ashoka into a Buddhist"

PARTLY TRUE. The remorse is in the edict: "This is the repentance of Devanampriya on account of his conquest of (the country of) the Kalingas". But the edict says it turned him to "morality", not that it made him a Buddhist, and it names no battlefield scene. A separate edict, found at Rupnath and several other sites, gives its own timeline: "Two and a half years and somewhat more (have passed) since I am openly a Sakya", a lay follower of the Buddha, and at first "(I had) not been very zealous." Hultzsch reads the Kalinga war as "the turning-point in his religious career". That reading rests on how the Rupnath edict is dated, and, in Hultzsch's words, those edicts "do not specify the years of Asoka's reign in which his conversion ... took place". Ashoka himself never links the war to the Buddha, and not every source puts the war first.

The Sri Lankan chronicle, the Mahavamsa, tells a different story. It places his conversion in the fourth year of his reign, before the war, and does not tell the Kalinga story at all. Hultzsch's verdict on its dates: "These dates cannot be reconciled with the epigraphical ones and must be erroneous."

A striking detail: in Kalinga itself, at Dhauli and Jaugada, the remorse edict was not carved. Hultzsch notes that "two separate edicts were substituted" there, because the sites "formed part of the newly conquered country of Kalinga and required different treatment".

3. "After Kalinga, Ashoka renounced violence"

MYTH as stated. The same edict that expresses remorse also warns the peoples of the forests within his lands: "they are told of the power (to punish them) which Devanampriya (possesses) in spite of (his) repentance, in order that they may be ashamed (of their crimes) and may not be killed." It asks his descendants not to think "that a fresh conquest ought to be made", but adds that "if a conquest does please them, they should take pleasure in mercy and light punishments". His own preferred conquest was "the conquest by morality". What the edict does set aside is further conquest by arms.

He also kept the death penalty. Pillar Edict IV grants "a respite of three days" to prisoners "who have been condemned to death", time for appeals and preparation, not abolition. Rock Edict I shows the same step-by-step reform in the royal kitchen: where "many hundred thousands of animals were killed daily for the sake of curry", only three were now killed, and "even these three animals shall not be killed (in future)."

Even the Buddhist legend does not make him a pacifist. In a 1911 English version of the Ashokavadana, made from Eugène Burnouf's French, the king, already known for his piety, orders a massacre after an insult to the Buddha: "in one single day eighteen thousand inhabitants suffered death." It is a story, not evidence of a real massacre. It shows that the gentle pacifist is not the picture even the legend paints.

4. "Ashoka built a Buddhist empire"

MYTH. Ashoka was openly Buddhist, and a few short edicts deal with the Buddhist order itself, but his major edicts address everyone. Rock Edict XII says the king "is honouring all sects; both ascetics and householders", and that "other sects ought to be duly honoured in every case." Twelve years after his consecration, after Kalinga and after he became a lay Buddhist, he gave a cave in the Barabar Hills to the Ajivikas, a non-Buddhist ascetic order: "this Banyan-cave was given to the Ajivikas."

What the edicts ask of people is everyday ethics: "proper courtesy to slaves and servants, obedience to mother (and) father, liberality to friends, acquaintances, and relatives", gifts to Brahmins and ascetics alike, and "abstention from killing animals". Hultzsch observed that the edicts "do not yet know anything of the doctrine of Nirvana".

5. "Ashoka killed 99 of his brothers to take the throne"

UNVERIFIED. The story comes from the Mahavamsa, which says that "when he had slain his ninety-nine brothers born of different mothers" he "won the undivided sovereignty over all Jambudipa". No edict mentions it. Rock Edict V points the other way: it has his officials at work "in the harems of our brothers, of (our) sisters", that is, in his brothers' royal households, in the present tense. Hultzsch took this as showing that the tradition that he "had killed all his brothers but one before ascending the throne ... is opposed to facts." That is Hultzsch's inference: a brother's household could in principle outlive the brother.

6. "He was 'Ashoka the Fierce', ran a prison like Hell, and built 84,000 stupas"

UNVERIFIED. All three come from the legends, not the edicts. The Mahavamsa says he was called "Candasoka (the wicked Asoka)... by reason of his evil deeds" before becoming "Dhammasoka (the pious Asoka)". In the Ashokavadana, as rendered in 1911, "he is Chandasoka, Asoka the Furious", a monk tells the king he has "constructed this dwelling, which is like unto Hell, wherein thousands are put to death", and after his conversion he "on the same day erected throughout the world eighty-four thousand Stupas".

These are edifying religious stories, built on a before-and-after shape, and they were written long after him. The edicts give their own before-and-after, a war and a public repentance, and none of these details.

7. "Ashoka's message reached the Greek kings"

PARTLY TRUE. Rock Edict XIII claims that his "conquest by morality" had been won as far as the lands "where the Yona king named Antiyoka (is ruling)" and four more kings. Hultzsch identifies them as Antiochus II of Syria, Ptolemy II of Egypt, Antigonus Gonatas of Macedonia, Magas of Cyrene, and an Alexander of Epirus or of Corinth. Those names help fix Ashoka's reign in time: Hultzsch gives, for example, "Antiochus II Theos 261-246". But this is Ashoka's claim. We did not find a Greek source recording his envoys. What does survive is his message in Greek: inscriptions found at Kandahar, in Afghanistan, and published in 1958 and 1964 include Greek versions of parts of Rock Edicts XII and XIII, this very edict, made for the Greek community living there, inside his own empire.

8. "India's national emblem comes from Ashoka"

TRUE, with two details. According to the Government of India, "The State Emblem is an adaptation of the Lion Capital of Asoka at Sarnath", adopted on 26 January 1950. The national flag's wheel, "a navy-blue wheel with 24 equally spaced spokes", represents the Ashoka Chakra. The motto Satyameva Jayate, written below the lions, is part of the modern emblem, not of the ancient capital, and "The bell-shaped lotus was omitted."

The stone Ashoka also had to be rediscovered. His name survived in Buddhist literature and the Puranas, but the edicts themselves had to be deciphered again: according to Hultzsch, the Delhi-Topra pillar was "the first record of Asoka that was read and translated in 1837 by Prinsep".

Our reading

The Ashoka of the edicts is more complicated than the pacifist saint. He admits to a war that killed and uprooted hundreds of thousands, says he regrets it, keeps the power to punish, reforms his own kitchen one step at a time, and honours religious groups other than his own. The legends tell a story of transformation. The stone tells a story of a ruler trying to govern by persuasion without giving up force. Both are worth reading; only one is the record.

Sources

E. Hultzsch, Inscriptions of Asoka, Corpus Inscriptionum Indicarum vol. I, new edition (Clarendon Press, for the Government of India, 1925).
The Mahavamsa, or the Great Chronicle of Ceylon, trans. Wilhelm Geiger (Pali Text Society, 1912).
Legends of Indian Buddhism, translated from Eugène Burnouf, with an introduction by Winifred Stephens (London, 1911).
Jean Przyluski, The Legend of Emperor Asoka in Indian and Chinese Texts, trans. D. K. Biswas (Calcutta, 1967).
Encyclopaedia Iranica, "Aśoka" i (J. G. de Casparis), ii (G. Fussman) and iii (P. O. Skjærvø), 1987.
Daniel Schlumberger, "Une nouvelle inscription grecque d'Açoka", Comptes rendus de l'Académie des Inscriptions et Belles-Lettres 108 (1964), pp. 126-140.
Government of India, National Portal "Know India": "State Emblem" and "National Flag" (archived copies, 2026).

Checked September 2026.

Cognitive Biases, Checked: 14 Famous Findings and Whether They Held Up

Hacks Vitae — Thu, 17 Sep 2026 00:00:00 GMT

Cognitive biases are often presented as settled science. They are not all equally settled.

Psychologists have re-run many of their field's most famous experiments — often with larger samples, in many labs at once, and sometimes with the analysis plan registered in advance. At least one came back stronger than first reported. Several came back smaller. A few have not come back convincingly at all.

This page checks fourteen of them. The cheat sheet on this page ("What we found") has a verdict and the key study for each; the sections below explain how to read those verdicts.

Why this matters: the replication projects

In 2015, the Open Science Collaboration published an attempt to repeat 100 studies from three psychology journals. By one of its measures, "Thirty-six percent of replications had statistically significant results", and the replication effects "were half the magnitude of original effects, representing a substantial decline."

It is tempting to read that as "only 36% of psychology replicates". That goes too far. The project used several measures of success, and 36% is only one of them. A fairer summary is narrower: many published effects were weaker than first reported, and some did not reappear.

The Many Labs projects took a different approach — testing a set of well-known effects across dozens of labs at once. In the second of them, covering 28 findings, the authors reported that 15, or 54%, "provided evidence of a statistically significant effect in the same direction as the original finding."

How to read the verdicts

TRUE — the finding replicated in large, independent tests.
PARTLY TRUE — something real was found, but it is smaller, narrower, or different from the popular version.
DISPUTED — credible studies point in different directions. We say which way the evidence currently leans and why.
UNVERIFIED — the original finding has not been confirmed by a convincing replication.

Two cautions apply to every one of them. A single failed replication is not a disproof: it may have tested the idea differently, or been unlucky. And a single success is not the end of the argument either. What matters is the weight of evidence across several good tests — which is also why these verdicts could change.

What the fourteen show — our reading

The biases about judgement mostly held up, sometimes smaller. Anchoring and hindsight bias replicated clearly; framing replicated at about half size; the sunk cost effect replicated as a small effect in a hypothetical scenario.

The findings about subtle manipulations mostly shrank or failed. Ego depletion, elderly priming, power posing's hormone effects and the pen-in-mouth smile all claimed that a small manipulation changes how people feel, perform or act. Those are the claims that struggled most when tested again.

Several famous findings are real but routinely misdescribed. The bystander effect is real for each individual, yet in real conflicts caught on camera someone usually helps. The marshmallow test found something, but much less than the popular story. The Dunning–Kruger effect has a real core, but how much of its classic pattern is real is still argued over.

Using this without over-correcting

It is tempting to swing from "psychology proves everything" to "psychology proves nothing". Neither is right. The replication projects are themselves psychology working as it should — checking its own claims and changing its mind.

The practical rule for any study you hear about: ask whether it has been independently repeated, and how big the effect was the second time.

For the method of checking a study yourself, see how to check something online and our guide to numbers that mislead. The Dunning–Kruger chart is covered in Older Than You Think, Part 2.

Sources

Open Science Collaboration, "Estimating the reproducibility of psychological science", Science (2015).
Klein and colleagues, Many Labs 1, Social Psychology 45 (2014).
Klein and colleagues, Many Labs 2, Advances in Methods and Practices in Psychological Science (2018).
Tversky and Kahneman, "Judgment under Uncertainty: Heuristics and Biases", Science 185 (1974), and "The framing of decisions and the psychology of choice", Science (1981).
Chen and colleagues, Journal of Experimental Social Psychology 96 (2021).
Ditto and colleagues, "At Least Bias Is Bipartisan", Perspectives on Psychological Science (2019).
Watts, Duncan and Quan, Psychological Science (2018).
Fischer and colleagues, Psychological Bulletin (2011).
Philpot, Liebst, Levine, Bernasco and Lindegaard, American Psychologist (2020).
Ranehill, Dreber, Johannesson, Leiberg, Sul and Weber, Psychological Science 26 (2015); Dana Carney, "My position on Power Poses" (statement hosted on her faculty page).
Wagenmakers and colleagues, Perspectives on Psychological Science (2016); Coles, Larsen and Lench, Psychological Bulletin (2019); Coles and colleagues, Many Smiles Collaboration, Nature Human Behaviour (2022).
Gal and Rucker, Journal of Consumer Psychology 28 (2018); Mrkva, Johnson, Gächter and Herrmann, Journal of Consumer Psychology 30 (2020); Brown, Imai, Vieider and Camerer, "Meta-analysis of Empirical Estimates of Loss Aversion", Journal of Economic Literature (2024); Walasek, Mullett and Stewart, "A meta-analysis of loss aversion in risky contexts", Journal of Economic Psychology 103 (2024): https://wrap.warwick.ac.uk/id/eprint/185745/13/1-s2.0-S0167487024000485-main.pdf
Kruger and Dunning, Journal of Personality and Social Psychology 77 (1999); Gignac and Zajenkowski, "The Dunning-Kruger effect is (mostly) a statistical artefact", Intelligence 80 (2020); Jansen, Rafferty and Griffiths, Nature Human Behaviour (2021).
Flore and Wicherts, Journal of School Psychology (2015); Flore, Mulder and Wicherts, Comprehensive Results in Social Psychology 3 (2018); Shewach, Sackett and Quint, Journal of Applied Psychology (2019).
Hagger, Chatzisarantis and colleagues, Perspectives on Psychological Science (2016); Vohs, Schmeichel, Lohmann and colleagues, Psychological Science (2021); Dang and colleagues, "A Multilab Replication of the Ego Depletion Effect", Social Psychological and Personality Science (2020).
Doyen, Klein, Pichon and Cleeremans, PLoS ONE (2012).

Checked September 2026.

DAWs Compared: Free and Paid Music Production Software, With Real Limits and Prices

Hacks Vitae — Thu, 17 Sep 2026 00:00:00 GMT

A DAW (digital audio workstation) is the software people use to record, arrange and mix music. There are dozens, and their "free" offers are easy to misread. This guide sorts more than twenty of them by what you actually get without paying, then lists the paid editions. Everything was read from each company's own pages on 17 September 2026.

This is not a ranking. The right DAW depends on your computer, the kind of music you make and how you like to work.

How to read this guide

Free (no time limit) — you can keep using it, but tracks or features may be capped.
Trial — full or near-full access for a set number of days.
Free, but key features blocked — something essential (like reopening projects) is switched off.

About prices: they are shown exactly as each vendor's page displayed them. Some sites served euros because of where the pages were read; others showed US dollars. We have not converted them. The same page can change currency: Ableton's shop showed Live 12 Intro at EUR 79 in one session and USD 99 in another on the same day. Prices differ by country, may or may not include tax, and change often — check the vendor's page for your region.

Free, with no time limit

DAW	Platforms	What the free version limits
GarageBand (Apple)	Mac, iPhone, iPad	Free on the App Store
Pro Tools Intro (Avid)	Mac, Windows	"up to eight audio tracks, eight instrument tracks, and eight MIDI tracks"
Reason Free (Reason Studios)	Windows, macOS	"16 audio and/or MIDI tracks, the same set of 15 Devices"
Waveform Free (Tracktion)	macOS 13+, Windows 10/11, Linux	Tracktion calls it "completely unlimited"; it includes 14 built-in audio effects and 4 instruments (Waveform Pro: 38 and 10), and more features are sold as expansion packs
Cakewalk Sonar, free tier (BandLab)	Windows (11 or higher suggested)	Some features are membership-only
BandLab	Web, iOS, Android	Paid membership gives "a doubled limit of 32 tracks"
Serato Studio, Free Mode	macOS, Windows	1 audio track, 4 decks, 4 scenes
MPC Beats (Akai)	Windows, macOS	"2 stereo audio tracks"
Fender Studio (not Studio Pro)	iOS, Android, macOS, Windows, Linux	Free registration unlocks "a higher track limit of 16"
LMMS	Linux, Windows, macOS	Free and open source
Audacity	Windows, macOS, Linux	Free and open source; it describes itself as an "audio editing and recording app"
Ardour	Linux, macOS, Windows	Free to build from source; ready-made downloads are pay-what-you-want from US$1. Its FAQ also mentions a "free/demo version", and says that when a paid copy "keeps going silent after 10 minutes" the usual cause is that the free/demo version is still the one running

A few notes on these:

Pro Tools Intro: Avid says "Pro Tools Intro is free forever."
Reason Free is new: Reason Studios announced it on 27 August 2026, promising "No trial or demo, no watermark, no expiry."
Cakewalk Sonar: its free licence "is currently indefinite, but as with the premium license, it requires periodic renewal through online or offline activation", and "some premium capabilities are reserved for BandLab Membership".
Ardour: subscriptions are optional — "Nothing about Ardour's functionality is connected to your subscription."
Audacity describes itself as "The world's most popular audio editing and recording app".

Free, but key features blocked

FL Studio: "The trial is fully functional and never expires — you can use all features, but saved projects can't be reopened."
Renoise: the demo has "No ASIO support on Windows" and "Rendering to .wav is disabled"; rendering selections and freezing plugin instruments are also disabled, and it shows "Nag screens".

Trials

DAW	Trial	What happens after
Ableton Live	30 days of Live 12 Suite	"you can still open Live Sets created using the Trial, but you will no longer be able to save or export your work."
Pro Tools	30 days of Pro Tools Ultimate	"all expiring subscriptions and trials will automatically default to Pro Tools Intro"
Cubase	60 days (Pro or Elements)	"you'll need a Cubase license to continue using the software"
REAPER	"60 days of evaluation free, with full functionality"	Not stated on the purchase page
Bitwig Studio	"without limitations for 30 days"	"You'll still be able to start Bitwig Studio in demo mode and open your projects, but save and export will be disabled."
Logic Pro (via Apple Creator Studio)	"free for 30 days"	Subscription starts
Waveform Pro	"30 day unlimited trial"	Not stated on the product page
Reason+	"free 7-day trial"	Subscription starts
BandLab Pro	"3-day free trial"	Membership starts

Paid editions and prices

DAW	Current version	Platforms	Editions and prices (as shown)	Licence
Ableton Live	12.4.6 (15 Sep 2026)	Mac, Windows	Intro EUR 79 (16 tracks) · Standard EUR 279 · Suite EUR 599	One-time; Suite also rent-to-own
FL Studio	26.1.6	Windows, macOS	From $99 (Fruity) to $449 (All Plugins)	One-time; "all future updates to your Edition: Free, forever"
Logic Pro	12.3.1	Mac (and iPad via Creator Studio)	$199.99 one-time for Mac, or Apple Creator Studio $12.99/month or $129/year	One-time or subscription
Pro Tools	2026.4.1 (July 2026)	Mac, Windows	Artist $9.99 · Studio $34.99 · Ultimate $99 "USD /month", excluding VAT and applicable taxes; new 1-year paid-upfront plans discounted for the first year until 30 September 2026	Subscription; perpetual via resellers
Cubase	15.0.30 (3 Jun 2026)	macOS, Windows	Pro €579 · Artist €329 · Elements €99.99, shown reduced to €347.40 · €197.40 · €59.99 in a "Save up to 40%" sale (incl. 21% VAT)	One-time
Fender Studio Pro	8.1	macOS 13+, Windows 10/11 (Linux in public beta)	$199.99 one-time, or $19.99/month (Pro+)	One-time or subscription
Bitwig Studio	6.1.1 (2 Sep 2026)	macOS, Windows, Linux	Essentials EUR 99 · Producer EUR 199 · Studio EUR 399	One-time; "updates in the following 12 months are FREE"
REAPER	7.80 (13 Sep 2026)	Windows, macOS, Linux	$60 discounted · $225 commercial	One-time; "free upgrades through REAPER version 8.99"
Reason	14	Windows, macOS	Reason 14 €239 · Reason Rack 14 plugin €159	One-time or Reason+ subscription
Waveform Pro	14	macOS, Windows, Linux	$249 one-time (upgrade $149)	One-time
Serato Studio	2.5.0 (2 Dec 2025)	macOS, Windows	$149 one-time · Producer Suite $9.99/month	One-time or subscription
Renoise	3.5.4	Windows, macOS, Linux	EUR 76 / USD 88 plus VAT, "license from 3.5.4 - 4.5.4"	Licence spans versions
Maschine 3 (Native Instruments)	3.5 (per NI support; the product page still says 3.4)	macOS, Windows	€109	Purchase (term not stated on the page)
BandLab Membership	—	Web, iOS, Android	Pro $14.99/month, or $99 first year then $149/year	Subscription

Details worth knowing:

REAPER's cheaper licence is for personal use, or for commercial users whose "yearly gross revenue does not exceed USD $20,000".
Ableton's rent-to-own is for Suite only: "Rent-to-own lets you rent Live 12 Suite on a monthly basis."
FL Studio prices come from product data on Image-Line's pricing page; the page itself says "From $99".
Pro Tools: Avid's page does not say in its text whether the monthly prices are billed monthly or annually.
Waveform Pro on Linux: "Waveform Pro features are unlocked automatically while running on Linux."
Soundtrap (web-based) has a free version and several paid plans; we could not read its prices from its site.

Renames and recent changes

Studio One Pro is now Fender Studio Pro. Fender announced on 13 January 2026: "Starting this year, Studio One Pro becomes Fender Studio Pro". The free Fender Studio app is a different product.
Cakewalk by BandLab has ended; Cakewalk Sonar replaces it. The Sonar page calls this "the official end of development and support for Cakewalk by BandLab, coinciding with the introduction of a free tier of Cakewalk Sonar." It says "Activation renewals for Cakewalk by BandLab ceases December 31, 2025." (Another answer on the same page gives 1 August 2025 and does not explain the difference; both dates have now passed.)
Logic Pro in a subscription. Apple Creator Studio, a bundle that includes Logic Pro, became available on 28 January 2026 "for $12.99 (U.S.) per month or $129 (U.S.) per year". Logic Pro for Mac is still sold as a one-time purchase.
Reason Free launched on 27 August 2026.

Questions to ask before you choose

Does it run on your computer? Some run only on Apple devices (Logic Pro, GarageBand) or only on Windows (Cakewalk Sonar).
Is the free version enough for your projects? Check the track limit against what you make.
What happens when the trial ends — can you still open and export your work?
One-time or subscription? And for one-time licences, how long are updates included?
Do you need a full DAW or an editor? Audacity describes itself as an audio editing and recording app.

Sources

Each vendor's own product, pricing, trial, download and release-notes pages, read on 17 September 2026: Ableton (Live 12 release notes, shop, trial); Image-Line (FL Studio download and pricing); Apple (Logic Pro page and release notes, App Store listings for Logic Pro and GarageBand, Apple Creator Studio newsroom announcement of 13 January 2026); Avid (Pro Tools, Pro Tools Intro, perpetual licences, What's New); Steinberg (Cubase product, trial and downloads pages); Fender (press release of 13 January 2026, Fender Studio Pro store page and Quick Start Guide); Bitwig (buy, download, previous releases, trial support); Cockos (REAPER home, purchase and download pages); Reason Studios (shop, "Reason 14 is Out Now" and "Announcing Reason Free" posts, Get Reason FAQ, Reason+ page); Cakewalk (Sonar page); BandLab (Membership FAQ, Payments FAQ); Soundtrap (support: subscription plans; free version); Tracktion (Waveform Free and Waveform Pro); LMMS (download page); Ardour (home page, FAQ); Renoise (home and downloads); Audacity (home); Serato (Studio pricing and download archive); Native Instruments (Maschine 3 product page and support); Akai Professional (MPC Beats).

Checked September 2026.

The Free Music Production Toolkit: Synths, Effects, Sample Licences, Notation and Stem Separation

Hacks Vitae — Thu, 17 Sep 2026 00:00:00 GMT

You can make a complete record without paying for software. This guide covers the tools around the DAW (for the DAWs themselves, see our DAWs compared guide): instruments, effects, sample licences, notation and stem separation. Everything was read from each project's own site, repository or licence on 17 September 2026.

This is not legal advice. Where a licence matters, we quote it and name where it comes from (see Sources); read the full terms for your own use.

Four kinds of "free"

Free and open source: you can use it and read or change the code, under a licence such as the GPL or MIT.
Free but closed: free to download and use, but the code is not published.
A free tier of a paid product: the free version works; paying adds sounds, presets or features.
Content licences: the rules for the samples and loops you use, separate from the software that plays them.

Some tools fit more than one: Vital has a free tier and publishes its source code under the GPLv3, and TDR Nova is free but closed, with a paid edition.

1. Free synths

Synth	Type of free	Current version	Platforms
Vital	Free tier of a paid product; source code GPLv3	Not stated on its site	Windows 10+, macOS 10.15+, Ubuntu 18.04+
Surge XT	Open source (GPL-3.0)	1.3.4 (11 August 2024)	Windows, macOS, Linux
Dexed	Open source (GPL v3)	1.0.1 (29 November 2025)	macOS, Windows, Linux
TAL-NoiseMaker	Free; no licence stated on its page	5.0.6 (16 August 2023)	Windows 7+, macOS 10.9+, Linux
Splice INSTRUMENT	Free tier of a paid product	Not stated	macOS 11+, Windows 10+

Vital: the free "Basic" version is "The full synth with all features", with "75 presets" and "25 wavetables". Its source code is public too: "The source code is licensed under the GPLv3." But the repository adds: "Do not distribute the presets that come with the free version of Vital", and "This repository is updated on a delay after binary releases".
Surge XT began as a commercial product: "In September 2018, Claes decided to release a partially completed version of Surge 1.6 under GPL3". Its site also offers nightly test builds.
Dexed is modelled on Yamaha's DX7 FM synth: "Dexed is free software and is licensed on the GPL v3."
TAL-NoiseMaker is listed among TAL's "FREE Plug-Ins", and its downloads include "VST VST3 AU AAX CLAP" formats. Its page links no source code. Source code for an older version (copyright 2005–2010) is kept in the DISTRHO Ports repository, whose file headers say it may be licensed under the "GNU General Public License Version 2".
Splice INSTRUMENT now includes what was Spitfire Audio's LABS: "LABS is becoming part of Splice INSTRUMENT", and the old LABS address redirects there. "Splice INSTRUMENT is free to download and use for anyone", with some free presets; the INSTRUMENT plan is "$12.99/mo", and the page says premium packs also come "with an INSTRUMENT, Creator, or Creator+ plan". If you cancel, "Any DAW projects with INSTRUMENT loaded will continue to playback", and "any Free Drops you claim are yours to keep, and play, for good."

2. Free effects

Valhalla Supermassive (reverb and echo, Mac and Windows) is free: "No strings attached." Its latest update: "We have just released the 5.0.0 update of Valhalla Supermassive", posted on 25 November 2025, which added a mode called Sirius.
TDR Nova (Tokyo Dawn Labs): "NOVA is a parallel dynamic equalizer." It is a free download for Windows and Mac; "Latest version: 2.2.2". A paid "Gentleman's Edition" is sold for €60.
Airwindows: "All this is free and open source under the MIT license". "These plugins come in Mac AU, and Mac, Windows and Linux VST", and new plug-ins are posted regularly on the site.

A note on plug-in formats

VST: "Virtual Studio Technology (VST®) is an audio plug-in software interface that facilitates the integration of software synthesizers and effects in digital audio workstations (DAW)". Steinberg's VST 3 SDK is now MIT-licensed; its repository says licensing under GPLv3 and the Steinberg proprietary licence "is no longer available".
CLAP ("CLever Audio Plug-in") is an open format: "It is an audio plug-in ABI which defines a standard for Digital Audio Workstations and audio plug-ins". Its code is MIT-licensed.
AU (Audio Units) is Apple's format. Apple's developer documentation for Audio Unit v3 says developers can "Deliver custom audio effects, instruments, and other audio behaviors using an Audio Unit v3 app extension."

3. Sample and loop licences

The software may be free, but the sounds you put in a track have their own rules.

Licence	On commercial use, the deed says	On credit, the deed says
CC0	"even for commercial purposes"	No credit term; "you should not imply endorsement by the author or the affirmer"
CC BY 4.0	"for any purpose, even commercially"	"You must give appropriate credit"
CC BY-NC 4.0	"You may not use the material for commercial purposes"	"You must give appropriate credit"
Sampling+ 1.0 (retired)	Transforming it: "for commercial or noncommercial purposes"; copies of the whole work: "for noncommercial purposes"	"You must give appropriate credit"

CC0: "You can copy, modify, distribute and perform the work, even for commercial purposes, all without asking permission."
CC BY 4.0 allows use "for any purpose, even commercially", but "You must give appropriate credit, provide a link to the license, and indicate if changes were made."
CC BY-NC 4.0: "NonCommercial — You may not use the material for commercial purposes."
Sampling+ 1.0 is retired: "Creative Commons has retired this legal tool and does not recommend that it be applied to works." It allowed people "To sample, mash-up, or otherwise creatively transform this work for commercial or noncommercial purposes", but only "To perform, display, and distribute copies of this whole work for noncommercial purposes", and "You may not use this work to advertise for or promote anything but the work you create from it." Its legal code requires derivative works to be "highly transformative of the original", and Freesound's FAQ gives "Our interpretation of this license": "you can do pretty much what you could do with the attribution noncommercial license, but additionally you can't make commercial advertisement with the sound."

Freesound, a free sound library, "lets the user select one of three licenses for their sounds. And, we used to have a 4th license" (Sampling+). Its FAQ explains the NC licence as "works like attribution, but you can't earn any money with the piece of work you create!", and adds: "We aren't lawyers so this isn't legal advice".

Splice sounds come with a licence, not ownership: "Sounds are licensed, not sold, to you." Among its prohibited uses: you may not "redistribute Sounds in new sample packs". If you cancel a plan you "Keep everything you download, even if you cancel", and "Your remaining credits will expire 28 days after your final billing period ends". The terms were last updated on 24 July 2026.

Looperman's terms say that a user who uploads loops grants registered users "the non-exclusive, royalty free right to copy, reproduce, edit, adapt, modify, remix, use, incorporate and otherwise exploit the specified Loops in new and original recordings of new and original musical and/or lyrical compositions". Uploaded loops "must be 100% owned by you, 100% your own original work", and the terms add: "It is your responsibility to ensure that you have the appropriate rights and licences to exploit Content for your intended purposes."

4. Free notation software

MuseScore Studio: "The latest stable version is 4.7.5 Released September 8, 2026". "MuseScore Studio is a GNU GPL licensed open-source program." It runs on Windows 10 or higher, macOS 11 or higher and "Linux distros released in 2022 or later".
LilyPond works differently: you type the music as text, and "it is more similar to a programming language than a graphical score editing program." "LilyPond is free software and part of the GNU Project." The current stable release is 2.26.0 (21 April 2026).
Dorico SE: Steinberg describes it as "the free notation software, Dorico SE", and its latest update, 6.2.30 (3 June 2026), "is a free update for existing Dorico Pro 6, Dorico Elements 6 and Dorico SE 6 users." On iPad, "Dorico is free to use, allowing you to write for ensembles of up to four players", and you can "Register with your free Steinberg ID to increase the player limit to eight".
Sibelius First (Avid): "Download Sibelius First for free Includes 30-day trial to Sibelius Ultimate." Avid's reference guide for version 2026.5 says "It supports up to 8 staves in a score", and "you are limited to two voices with Sibelius First".

5. Stem separation

Stem-separation tools split a finished mix into parts such as vocals, drums and bass.

Demucs's code is MIT-licensed. Meta's original repository is archived and says "this repository is not maintained anymore". Its author's copy now says: "This is the officially maintained Demucs now that I (Alexandre Défossez) have left Meta", but also: "Note that I'm not actively working on Demucs anymore, so expect slow replies and no new feature for now." Version 4.1.0 was released on 11 July 2026.
Ultimate Vocal Remover (UVR) is a free desktop app for Windows 10 or higher and macOS Big Sur and above. Its latest GitHub release is version 5.6, from 26 September 2023. Its README says the code is MIT-licensed.
Spleeter (Deezer): "The code of Spleeter is MIT-licensed"; its latest release on PyPI is 2.4.2 (3 April 2025). Its README carries this disclaimer: "If you plan to use Spleeter on copyrighted material, make sure you get proper authorization from right owners beforehand."

6. Recording and editing: Audacity

Audacity is a free, open-source audio editor. "Audacity is released under the GNU General Public License version 3 (GPLv3)." Its "Current version: 4.0.0" was released on 3 September 2026, and it runs on Windows, macOS and Linux. Its site's copyright line reads "Muse Group & contributors."

Questions to ask before you choose

Is it open source, free but closed, or a free tier?
Does it run on your system and in your DAW's plug-in format (VST3, AU, CLAP, AAX)?
What licence do your samples carry? Commercial use and credit requirements differ.
Do you have the rights to the audio you are separating or sampling?
Is the project still maintained? Check the date of the latest release.

Sources

Each project's own site, repository, release notes or licence, read on 17 September 2026: Vital (vital.audio and the mtytel/vital repository); Surge XT (site, repository and releases); Dexed (site and releases); TAL Software (products, TAL-NoiseMaker); DISTRHO Ports (TAL-NoiseMaker source); Splice (INSTRUMENT, plans, Terms of Use); Valhalla DSP (Supermassive page and 5.0.0 post); Tokyo Dawn Labs (TDR Nova and TDR Nova GE); Airwindows (site and licence); Steinberg (VST 3 developer portal and SDK repository; Dorico blog); CLAP (cleveraudio.org and licence); Apple (Audio Unit v3 documentation); Creative Commons (CC0 1.0, CC BY 4.0, CC BY-NC 4.0 and Sampling+ 1.0 deeds, and the Sampling+ 1.0 legal code); Freesound (FAQ); Looperman (terms); MuseScore (download page and licence); LilyPond (home and download pages); Dorico for iPad (App Store listing); Avid (Sibelius First page; Sibelius Reference Guide 2026.5); Demucs (Meta and author repositories, release notes, licence); Ultimate Vocal Remover (repository); Spleeter (repository, licence, PyPI); Audacity (download page, licence, releases).

Checked September 2026.

Genghis Khan and the Mongol Empire: 10 Popular Claims, Checked

Hacks Vitae — Thu, 17 Sep 2026 00:00:00 GMT

The Mongol Empire built in the 13th century by Genghis Khan and his successors was, in the words of the historian David Morgan, "the most extensive land empire known to history". It is also the subject of a great many confident claims. Here are ten of the most repeated, checked against the medieval sources and the modern studies behind them.

1. "1 in 200 men alive today are descended from Genghis Khan"

DISPUTED. The figure comes from a 2003 study in the American Journal of Human Genetics of a single Y-chromosome lineage, passed from father to son. The authors found that "approximately 8% of the men in this region carry it, and it thus makes up approximately 0.5% of the world total". But the headline number comes with a condition, "if our sample is representative", and the link to Genghis Khan was an inference: "The available evidence suggests that it was carried by Genghis Khan." They also noted that "It is not possible to examine his remains directly".

Later work has pulled both ways. A 2018 whole-sequence study in the European Journal of Human Genetics dated the lineage's common ancestor to about 2,576 years ago and concluded that "direct evidence of an association between C2*-ST and Genghis Khan has yet to be discovered." A 2026 study in PNAS of ancient DNA from Golden Horde elite tombs in Kazakhstan found the lineage in three related men, "confirming the association between the Y-chromosome haplogroup C3* and the Mongol Empire", and its authors think Genghis Khan's father's line most likely belonged to this group, though "conclusive evidence for any specific haplotype ... remains lacking". The lineage is linked to the Mongol ruling class; that it was Genghis Khan's own remains unproven.

Even taken at face value, the claim is about one line of fathers and sons, not about everyone descended from him.

2. "Genghis Khan cooled the planet"

MYTH. A 2011 modelling study in The Holocene looked at whether wars and epidemics, including "the Mongol invasion (~1200 to ~1380)", changed carbon dioxide levels as abandoned farmland grew back into forest. Its title already gives the answer: "minor global effects". The authors found that "Only the Mongol invasion could have lowered global CO2, but by an amount too small to be resolved by ice cores."

Its abstract reports a change in carbon dioxide too small to detect in ice cores, not a change in temperature. The study's press release did not claim cooling either. It said the Mongols "had an impact on the global carbon cycle as big as today's annual demand for gasoline", and added a caveat: "The global impact of forest re-growth in even the long-lasting events was diminished by the continued clearing of forests elsewhere in the world."

3. "Genghis Khan was his name"

MYTH. It was a title. According to The Secret History of the Mongols, the 13th-century Mongol chronicle, he was named Temüjin after a captive his father had just taken. The chronicle says his followers first made him khan, "naming him Cinggis Qa'an", years before 1206; in the Year of the Tiger, 1206, an assembly at the source of the Onan River confirmed him as khan over all the peoples he had united. His birth year is not known: Igor de Rachewiltz, the chronicle's translator, gives "probably in 1162", while Morgan, in the Encyclopaedia Iranica, gives "probably born in 562-63/1167". He died in 1227, during his last campaign, against the Tangut, and no portrait made in his lifetime survives.

4. "The Mongols killed 40 million people"

UNVERIFIED. We could not trace this figure to a count of the dead in any source we could read. The medieval numbers for single cities run into the millions: the historian Juvaini, writing in the 1250s, says the dead at Merv took "thirteen days and nights" to count and came to more than 1.3 million, and other chroniclers give different totals for the same cities. Morgan's judgement is that "Such statements should not be taken too literally" and that "it may be doubted that anyone was counting."

That caution cuts one way only. Morgan adds: "What cannot be doubted, however, is that the death and destruction was on a scale quite beyond contemporaries' previous knowledge or experience."

5. "Genghis Khan wrote a law code, the Yassa"

DISPUTED. Juvaini says the rules were to "be written down on rolls", and that "These rolls are called the Great Book of Yasas and are kept in the treasury of the chief princes." He was describing his own day, the 1250s, and he was born "in all probability, in the year 1226", a year before Genghis Khan died. He is also, in the historian Peter Jackson's words, "virtually our only source on the yāsā". Historians read that evidence differently. Morgan concludes: "It no longer seems possible to ascribe to him the creation of a written legal code", and that Mongol law "appears to have had a much more customary and ad hoc character". Jackson, in the Encyclopaedia Iranica's entry on the Yasa, notes that "there is nothing incongruous in the possibility that edicts were written down", and says they came to be regarded, "if not as a homogeneous and systematically organized code, at least as a recognizable corpus of regulations". The evidence leans against a single written law code drawn up by Genghis Khan himself, because Juvaini is almost the only witness; that his rulings were recorded and kept is quite possible.

6. "The Mongols were religiously tolerant"

PARTLY TRUE. Mongol rulers backed clergy of many faiths. Juvaini says Genghis Khan "honoured and respected the learned and pious of every sect", and the Mongols, according to Columbia University's Mongols in World History teaching resource, "even offered tax benefits to the clerics of Buddhism, Islam, Daoism, and Nestorian Christianity in order to win the support of those religions."

The friar William of Rubruck, who travelled to the Mongol court in 1253–55, heard the Great Khan Möngke (whom he calls Mangu) explain his view in person: "But as God gives us the different fingers of the hand, so he gives to men divers ways."

It had limits. Juvaini records that the Mongol khan Chaghatai enforced rules that no one might slaughter meat in the Muslim way or sit by day in running water, which Juvaini calls "an intolerable imposition" on his Muslim subjects. Tolerance here was a governing policy, not a right.

7. "The Mongols never bathed"

MYTH. The sources describe a specific taboo, not a lack of washing. Juvaini writes that "in the season of spring and summer no one may sit in water by day, nor wash his hands in a stream, nor draw water in gold or silver vessels, nor lay out washed garments upon the plain", because of the belief "that such actions increase the thunder and lightning".

William of Rubruck, writing separately from the Mongol court, reports the same rule and the same reason: "They never wash clothes, for they say that God would be angered thereat, and that it would thunder if they hung them up to dry." Rubruck also describes how they did wash: "When they want to wash their hands or head, they fill their mouths with water, which they let trickle on to their hands, and in this way they also wet their hair and wash their heads."

8. "The Mongols started the Black Death by throwing plague corpses into Caffa"

MYTH. The story of the 1346 siege of Caffa, in Crimea, comes from one account by a notary in Piacenza, Gabriele de' Mussi. In a modern English translation, the besiegers "ordered corpses to be placed in catapults and lobbed into the city". Mark Wheelis of the University of California, Davis, reviewing the account in 2002, found that de' Mussi "does not claim to have witnessed any of the Asian events he describes" and that his "account is probably secondhand and is uncorroborated".

Wheelis thought the attack plausible, and "the best explanation of the entry of plague into the city", but concluded that the siege was "unlikely to have been seriously implicated in the transmission of plague from the Black Sea to Europe". A 2022 study in Nature traced the pandemic's origin to central Eurasia, where "tombstone inscriptions directly dated to 1338-1339" name pestilence as the cause of death, and said that "the exact means through which Y. pestis reached western Eurasia are unknown". The besiegers were the army of the Mongol ruler Janibeg, whose Golden Horde was one of four parts the empire had split into by 1260, and the plague was already in that army: Wheelis records that the siege was lifted because of "an epidemic of plague that devastated his forces". So the corpse story may be true, but the Black Death did not start at Caffa.

9. "A divine wind saved Japan from the Mongols"

PARTLY TRUE. In 1281 a storm did wreck much of Kublai Khan's second invasion fleet. A Korean chronicle, the Goryeo-sa, records that the southern army "met with large wind, and all the Southern Army died". Some historians give the defenders more of the credit: several scholars, Sasaki notes, have suggested that coastal stone walls and "the small fast boats that the Japanese used" had already kept the army from landing in force. The 1281 wrecks at Takashima show a mixed fleet: a large cargo ship from Fujian, "strongly built Korean vessels", and many smaller hulls that Sasaki thinks were mostly built along the Yangtze and "not constructed for long overseas voyages".

The first invasion, in 1274, is less clear. According to the historian Hattori Hideo, Korean records mention a typhoon, and he argues that the army stayed seven days and then withdrew because of it; others read the landing as reconnaissance. In a study of the wrecks, the archaeologist Randall Sasaki writes that "The true reason for this retreat is difficult to identify": the official Yuan history says "the troop was not organized" and "all the arrows had been used", and one reading treats that landing as "more in the nature of a raid than a full-scale invasion".

10. "The Mongols left Europe in 1242 because the Great Khan died"

DISPUTED. The long-standing explanation is that the army turned back so its leaders could take part in choosing a new khan after the Great Khan Ögedei died. The historian Stephen Pow, in a 2012 thesis, argues that "these theories are unconvincing and contradicted by the existing primary source evidence". He points to the historian Rashid al-Din, who says that when one Mongol force turned back, "the news of Khan's death had not yet reached them". Pow suggests instead that "European fortifications produced a strategic problem". A third explanation, from a 2016 study of tree rings and written records, points to cold, wet conditions in early 1242 that turned the Hungarian plain to marsh; a team including Pow disputed it in 2017. Historians have not settled it.

Our reading

The Mongol conquests do not need exaggerating. The destruction was, as Morgan puts it, beyond anything contemporaries had known, and the empire was also a system of relay stations, clerical tax privileges and religious debate. The myths flatten both sides: they turn a conditional DNA estimate into a fact, a carbon model into a cooled planet, one notary's story into the origin of the Black Death, and a thunder taboo into dirt. To Mongols, Genghis Khan is a great national hero; elsewhere he is remembered for the killing. Both are part of the record, and we give no verdict on the man.

Sources

Tatiana Zerjal et al., "The genetic legacy of the Mongols", American Journal of Human Genetics 72(3) (2003).
Lan-Hai Wei et al., "Whole-sequence analysis indicates that the Y chromosome C2*-Star Cluster traces back to ordinary Mongols, rather than Genghis Khan", European Journal of Human Genetics 26(2) (2018).
Julia Pongratz, Ken Caldeira, Christian H. Reick and Martin Claussen, "Coupled climate–carbon simulations indicate minor global effects of wars and epidemics on atmospheric CO2 between AD 800 and 1850", The Holocene 21(5) (2011), abstract; and Carnegie Institution for Science press release, 20 January 2011 (archived).
The Secret History of the Mongols, trans. Igor de Rachewiltz, shorter version ed. John C. Street (2015), CC BY-NC-ND 4.0.
David O. Morgan, "Čengīz Khan", Encyclopaedia Iranica V/2 (1990, updated 2013).
Peter Jackson, "Yāsā", Encyclopaedia Iranica (2013, updated 2017).
ʿAtā-Malik Juvaini, The History of the World-Conqueror, trans. John Andrew Boyle, vol. 1 (Manchester University Press, 1958).
The Journey of William of Rubruck to the Eastern Parts of the World, 1253–55, trans. William Woodville Rockhill (Hakluyt Society, 1900).
Columbia University, Asia for Educators, The Mongols in World History.
Mark Wheelis, "Biological Warfare at the 1346 Siege of Caffa", Emerging Infectious Diseases 8(9) (2002).
Ayken Askapuli et al., "Genomes of the Golden Horde elites and their implications for the rulers of the Mongol Empire", PNAS 123 (2026).
Maria A. Spyrou et al., "The source of the Black Death in fourteenth-century central Eurasia", Nature 606 (2022).
Kawai Atsushi, "Saved by the Wind? The Mongol Invasions of Japan", Nippon.com (2021).
Randall James Sasaki, The Origin of the Lost Fleet of the Mongol Empire, MA thesis, Texas A&M University (2008).
Ulf Büntgen and Nicola Di Cosmo, "Climatic and environmental aspects of the Mongol withdrawal from Hungary in 1242 CE", Scientific Reports 6 (2016); and Zsolt Pinke et al., "Climate of doubt", Scientific Reports 7 (2017).
Lindsey Stephen Pow, Deep Ditches and Well-built Walls: A Reappraisal of the Mongol Withdrawal from Europe in 1242, MA thesis, University of Calgary (2012).

Checked September 2026.

How Large Language Models Work, in Plain Words

Hacks Vitae — Thu, 17 Sep 2026 00:00:00 GMT

AI chatbots such as ChatGPT, Claude and Gemini are built on large language models. You do not need maths to understand the basic idea, and knowing it explains a lot: why answers vary, why a model can sound confident and still be wrong, and why it may not know about recent events. This page explains the main steps in plain words, using the labs' and researchers' own descriptions. Details change quickly; this is a snapshot as of September 2026.

1. It predicts the next piece of text

The models behind these chatbots start from one basic task: given some text, predict what comes next. OpenAI's technical report describes its model this way: "GPT-4 is a Transformer-based model pre-trained to predict the next token in a document."

Anthropic's glossary gives the same idea in plainer words: models "are pretrained to predict the next word, given the previous context of text in the document". To write a long answer, models like these repeat that step: predict a piece, add it to the text, predict the next piece.

2. It works in tokens, not words

The "pieces" are called tokens. OpenAI's help centre explains: "A token can represent a character, part of a word, a whole word, or punctuation." It is direct about the difference: "A token count is not the same as a word count." For English, it gives a rough guide: "1 token is approximately three-quarters of a word. 100 tokens are approximately 75 words."

That ratio is not fixed. Anthropic notes that its Claude 4.7 and later models use a newer tokenizer and that "This tokenizer produces approximately 30% more tokens for the same text." OpenAI adds that "The same text can produce different token counts depending on the model, its encoding, and the language." Tokens matter to you because model limits and costs are counted in them.

3. The Transformer: paying attention to the whole sentence

The GPT-4 report above calls the model "Transformer-based". The Transformer design was introduced in the 2017 paper "Attention Is All You Need", which proposed an architecture "based solely on attention mechanisms, dispensing with recurrence and convolutions entirely". It was tested on translation, including an "English-to-German translation task".

Google's blog post about it explained the key idea, self-attention, which "directly models relationships between all words in a sentence, regardless of their respective position." Its example was the sentence "I arrived at the bank after crossing the river": to know that "bank" means a riverbank, the model has to connect it with "river", several words away.

4. Training: a very large amount of text

A model learns by adjusting a huge set of internal numbers, called parameters, until its predictions improve. Model sizes and, above all, the amount of training text have grown fast:

Model	Parameters	Training text
GPT-3 (2020)	175 billion — "an autoregressive language model with 175 billion parameters"	"All models were trained for a total of 300 billion tokens."
Chinchilla (2022)	70 billion	"We verify this by training a more compute-optimal 70B model, called Chinchilla, on 1.4 trillion tokens."
Llama 3 (2024)	405 billion — "405B trainable parameters on 15.6T text tokens"	15.6 trillion tokens

Researchers found that this kind of growth pays off in a predictable way. A 2020 study reported that "The loss scales as a power-law with model size, dataset size, and the amount of compute used for training". "Loss" here is the paper's measure of prediction error (cross-entropy loss), not a direct score of what a model can do. A 2022 study then concluded that "current large language models are significantly undertrained" and that, "for compute-optimal training", "for every doubling of model size the number of training tokens should also be doubled".

Some of the newest model descriptions give no total: OpenAI says GPT-6 Astra "was trained on diverse datasets", and Anthropic says Claude Opus 5 "was pretrained on large, diverse datasets". Some open-weight developers still publish a figure: DeepSeek's report says "we train DeepSeek-V4-Flash on 32T tokens and DeepSeek-V4-Pro on 33T tokens".

5. From text predictor to assistant

A model trained only to predict text is not yet a chatbot. Anthropic's glossary says: "These pretrained models are not inherently good at answering questions or following instructions". A further round of training shapes it into an assistant.

One method, used for both GPT-4 and Claude, is reinforcement learning from human feedback (RLHF). In OpenAI's InstructGPT work, people first wrote "demonstrations of the desired model behavior". The paper continues: "We then collect a dataset of rankings of model outputs, which we use to further fine-tune this supervised model using reinforcement learning from human feedback." The result was striking but limited in scope: "In human evaluations on our prompt distribution, outputs from the 1.3B parameter InstructGPT model are preferred to outputs from the 175B GPT-3, despite having 100x fewer parameters." The same paper noted that "InstructGPT still makes simple mistakes".

Anthropic's Constitutional AI trained a harmless assistant "without any human labels identifying harmful outputs": "The only human oversight is provided through a list of rules or principles".

6. Reasoning models: thinking before answering

Some models are now trained to work through a problem step by step before giving an answer. OpenAI's system card for its o1 models says: "The o1 model series is trained with large-scale reinforcement learning to reason using chain of thought." In a post dated September 12, 2024, OpenAI also reported that performance improved "with more time spent thinking (test-time compute)".

DeepSeek's R1 paper argued that "the reasoning abilities of LLMs can be incentivized through pure reinforcement learning (RL), obviating the need for human-labeled reasoning trajectories". Thinking also costs tokens. OpenAI's help centre says reasoning tokens "are not visible as answer text, but they count toward output usage and are billed as output tokens."

7. Why the same question can get different answers

When a model picks its next token, there can be an element of chance. Anthropic's glossary describes temperature as "a parameter that controls the randomness of a model's predictions during text generation", and explains: "Lower temperatures result in more conservative and deterministic outputs that stick to the most probable phrasing and answers." Even so: "Even with temperature set to 0, the results will not be fully deterministic and identical inputs may produce different outputs across API calls."

Some newer models do not let developers change it at all. Anthropic's API documentation says: "Models released after Claude Opus 4.6 do not support setting temperature."

8. Why a model may not know recent events

A model's built-in knowledge comes from its training text, which stops at some point. Labs sometimes give two different dates. Anthropic says its Claude Opus 4 and Claude Sonnet 4 models have a training data cutoff of March 2025 but a reliable knowledge cutoff of January 2025, "which means the models' knowledge base is most extensive and reliable on information and events up to January 2025".

To get past that limit, models can be connected to tools such as web search. Meta's Llama 3 paper says the model was trained to use a search engine "to answer questions about recent events that go beyond its knowledge cutoff".

What this explains

Fluent is not the same as checked. The model is trained to produce likely text; predicting the next token does not, by itself, check that text against a source. The GPT-4 report says the model "still is not fully reliable (it “hallucinates” facts and makes reasoning errors)".
Answers vary. Randomness in generation means the same question can produce different wording, and sometimes a different answer.
Recent events are a weak spot unless the model is searching, and then the answer depends on what it found.

Our reading: a large language model is best thought of as an extremely capable text predictor that has been further trained to be helpful. That makes it very useful for drafting, explaining and summarising — and it is exactly why any fact that matters should be checked against a source.

Sources

OpenAI, "GPT-4 Technical Report", arXiv:2303.08774.
Anthropic, Claude documentation: Glossary; Pricing; Messages API reference.
OpenAI Help Center, "Understanding and counting tokens".
Vaswani et al., "Attention Is All You Need", arXiv:1706.03762 (2017).
Google Research blog, "Transformer: A Novel Neural Network Architecture for Language Understanding" (31 August 2017).
Brown et al., "Language Models are Few-Shot Learners" (GPT-3), arXiv:2005.14165.
Hoffmann et al., "Training Compute-Optimal Large Language Models" (Chinchilla), arXiv:2203.15556.
Meta, "The Llama 3 Herd of Models", arXiv:2407.21783.
Kaplan et al., "Scaling Laws for Neural Language Models", arXiv:2001.08361.
Ouyang et al., "Training language models to follow instructions with human feedback", arXiv:2203.02155.
Bai et al., "Constitutional AI: Harmlessness from AI Feedback", arXiv:2212.08073.
OpenAI, "OpenAI o1 System Card", arXiv:2412.16720; "Learning to Reason with LLMs" (12 September 2024).
DeepSeek-AI, "DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning", arXiv:2501.12948; "DeepSeek-V4: Towards Highly Efficient Million-Token Context Intelligence", arXiv:2606.19348.
OpenAI, GPT-6 Astra system card, "Model Data and Training"; Anthropic, Transparency Hub, Claude Opus 5.
Anthropic, Transparency Hub.

Checked September 2026.

History Myths Checked: Nero's Fiddle, the Salem Witch Burnings and the "Dark Ages"

Hacks Vitae — Thu, 17 Sep 2026 00:00:00 GMT

Three stories almost everyone has heard: Nero fiddled while Rome burned, witches were burned at Salem, and Europe fell into a Dark Age when Rome fell. Each has a grain of history in it, and each has grown in the telling. Here is what the sources say.

1. "Nero fiddled while Rome burned"

MYTH. Nero could not have played a fiddle. Writing in The Classical Journal in 1947, Mary Francis Gyles, then a graduate student in history at the University of North Carolina, noted that "musical historians agree that the viol class of instruments did not develop until about the eleventh century", and that "There can be no doubt that the instrument employed by Nero was the cithara". Bowed instruments arrived many centuries after Nero.

The story behind the myth is genuinely ancient, but the three Roman writers who tell it disagree.

Tacitus (Annals XV, trans. John Jackson) says the cause of the fire was uncertain — "whether due to chance or to the malice of the sovereign is uncertain". He writes that "Nero, who at the time was staying in Antium, did not return to the capital until the fire was nearing the house", and gives the singing only as a rumour: "the report had spread that, at the very moment when Rome was aflame, he had mounted his private stage," and "had sung the destruction of Troy."
Suetonius (Nero 38, trans. J. C. Rolfe) states as fact that "he set fire to the city", and that "Viewing the conflagration from the tower of Maecenas" he "sang the whole of the "Sack of Ilium,"" "in his regular stage costume."
Cassius Dio, writing about 150 years after the fire (trans. Earnest Cary), puts Nero on "the roof of the palace", where "assuming the lyre-player's garb, he sang the "Capture of Troy,""

So Tacitus has Nero away at Antium when the fire broke out and back in the city while it was still burning, and he reports the song only as a rumour; the later writers put him watching the fire from a height in Rome and tell the song as fact. Jackson's note on the cause of the fire is blunt: "There is obviously no possibility of deciding the question."

Where the fiddle came from. Gyles traced the instrument through later literature. In Henry VI Part 1, Talbot vows to "Play on the lute, beholding the towns burn". The 1623 First Folio does not print Nero's name in that line; modern editors add it, reading the line as an allusion to him. On that reading, Gyles writes, this is where "for the first time an instrument is placed in Nero's hands during the burning of Rome" — and it is a lute, not a fiddle. A play printed in 1624 calls Nero a "fiddler", and "in 1649 George Daniel is found writing, "Let Nero fiddle out Rome's obsequies."" Her conclusion: "it is impossible to prove that Nero played a musical instrument during the burning of Rome."

Verdict in short: the fiddle is a myth; that Nero sang while the city burned is UNVERIFIED — Tacitus reports it only as a rumour.

2. "Witches were burned at Salem"

MYTH. No one was burned in the Salem witch trials of 1692–93. According to the University of Virginia's Salem Witch Trials Documentary Archive, "nineteen were executed by hanging, one was tortured to death, and at least five died in jail due to harsh conditions."

The court records say how. Bridget Bishop, the first person executed, was sentenced "to be hanged by the neck untill she be de[ad]", and the sheriff's return records that he "Caused the s'd Brigett to be hanged by the neck untill Shee was dead" on 10 June 1692. The hangings continued on 19 July, 19 August and 22 September 1692.

The man who died another way was Giles Corey. The judge Samuel Sewall wrote in his diary: "Monday, Sept. 19, 1692. About noon, at Salem, Giles Corey was press'd to death for standing Mute" — he had refused to enter a plea. (The archive's case index gives 16 September; Sewall wrote at the time, and 19 September 1692 in the calendar then used in the colony did fall on a Monday, by our calculation.)

Witchcraft had been a capital crime in the colony since its 1641 Body of Liberties, which says a witch "shall be put to death". But in 1692 the colony's old charter had been suspended and its new one had only just arrived, and the special court set up that year was told to act "according to the law and custom of England and of this their Majesties' Province", as the Massachusetts Trial Court Law Libraries explain. The colony passed its own witchcraft act only in December 1692, after the last execution. It kept "pains of death as a Felon" for the gravest offences and for a second conviction, and gave lesser ones a year in prison.

Burning witches did happen elsewhere. In Scotland, the University of Edinburgh's Survey of Scottish Witchcraft notes, "Those convicted were almost always strangled at the stake and then their dead body was burned."

3. "The Dark Ages"

MYTH as a description of the Middle Ages as a whole. The image of a dark time after Rome goes back to the Italian poet Petrarch, who in his Latin epic Africa looked forward to a day when "discussis forte tenebris" — the darkness perhaps dispersed — later generations could return to the pure light of old (our translation). He was lamenting lost classical learning, and he never used the term "dark ages" itself. But the idea of such a period starts with him: as the word historian Dave Wilton puts it, "the concept of such a period was first promulgated by Petrarch in the mid fourteenth century."

The English phrase came later, and at first it was religious polemic. Wilton also records that "In early Protestant writing, dark ages was often used to refer to the supremacy of the Roman Catholic Church prior to the Reformation", and that, "As applied to c. 500–1100, the term is a misnomer and has largely been abandoned by professional historians."

Why historians avoid it: the archaeologist Howard Williams summarises the critics' case — the term is seen as "inappropriate, out-dated and pejorative", and "text-orientated", as if centuries with few written sources were beyond study. "Most importantly," he adds, "it is already a rare term in academic contexts." Not every scholar has abandoned it, and Williams himself defends some of its uses.

DISPUTED. Did anything decline? Here historians genuinely disagree about emphasis. One school, which the historian James J. O'Donnell traces to Peter Brown's The World of Late Antiquity (1971), stresses transformation rather than collapse: in O'Donnell's summary, it speaks "of rises rather than falls" and gives more weight to the eastern empire and to religious and cultural history. Another is represented by Bryan Ward-Perkins's The Fall of Rome and the End of Civilization (2005), whose case O'Donnell summarises as "a significant decline in the complexity of organization and the prosperity of society in post-Roman times" in the West. O'Donnell, who counts himself on the other side, conceded that in the post-Roman West "stuff happened, bad stuff, lots of it", while arguing that leaving out the eastern empire means the case "does not go far enough".

So "the Middle Ages were a dark age" is a myth, but so is "nothing was lost when Rome fell in the West". What remains open is how to weigh the losses against what continued and what was new.

Our reading

All three myths turn a disputed or partial story into a clean image: a fiddling emperor, a burning witch, a lights-out millennium. The records underneath are messier and more interesting — a rumour reported by a careful historian, a colony that hanged nineteen people convicted by its special court, and a real argument among scholars about what the end of Rome meant.

Sources

Tacitus, Annals XV, trans. John Jackson, Loeb (1937).
Suetonius, Nero 38, trans. J. C. Rolfe, Loeb (1914).
Cassius Dio, Roman History 62, trans. Earnest Cary, Loeb (1925).
Mary Francis Gyles, "'Nero Fiddled While Rome Burned'", The Classical Journal 42.4 (1947).
Salem Witch Trials Documentary Archive and Transcription Project, University of Virginia: overview; The Salem Witchcraft Papers, ed. Paul Boyer and Stephen Nissenbaum (1977), rev. Benjamin C. Ray and Tara S. Wood; SWP No. 013 (Bridget Bishop); SWP No. 168 (bill of December 1692); diary of Samuel Sewall.
The Massachusetts Body of Liberties (1641), Hanover Historical Texts Project.
Massachusetts Trial Court Law Libraries, "Witchcraft law up to the Salem witchcraft trials of 1692" (archived copy).
The Survey of Scottish Witchcraft, University of Edinburgh.
William Shakespeare, Henry VI Part 1, 1.4: First Folio (1623), Internet Shakespeare Editions; Folger Shakespeare edition.
Francesco Petrarca, Africa IX, ed. N. Festa (1926), Perseus, Tufts University.
Dave Wilton, "dark ages", Wordorigins.org (2024).
Howard M. R. Williams, "Bring Back the Dark Ages!" (2016).
James J. O'Donnell, review of Peter Heather and Bryan Ward-Perkins, Bryn Mawr Classical Review 2005.07.69.
Dennis E. Trout, review of Rita Lizzi Testa (ed.), Late Antiquity in Contemporary Debate, Bryn Mawr Classical Review 2017.07.38.
Encyclopaedia Britannica, "Byzantine Empire".

Checked September 2026.

How to Check Something Online Before You Share It: The SIFT Method

Hacks Vitae — Thu, 17 Sep 2026 00:00:00 GMT

Most of what reaches you online arrives without its context: a screenshot, a quote on a picture, a link from someone you half know. You do not need a research degree to check it. You need a short routine that you will actually use.

A well-known routine is SIFT. It comes from Mike Caulfield — later co-author, with the Stanford researcher Sam Wineburg, of the 2023 book Verified — who set it out in a blog post in June 2019, building on "Four Moves" he had been teaching students since 2017. In his words:

"The four moves: Stop, Investigate the source, find better coverage, trace the original context."

Move 1 — Stop

Before you read or share, ask whether you know the website or source of the information, and the reputation of both the claim and the website. In Caulfield's words: "If you don't have that information, use the other moves".

Caulfield separates two kinds of checking. When you are doing deep research of your own, you may want to chase down individual claims and verify them independently. But, as he puts it, "Quick and shallow investigations will form most of what we do on the web."

Move 2 — Investigate the source

Spend about a minute finding out who is behind what you are reading. Caulfield puts it this way:

"Taking sixty seconds to figure out where media is from before reading will help you decide if it is worth your time"

The trick is not to judge a website by looking at the website. Official-looking logos and domain names are what the Stanford researchers below call "easily manipulated features". Instead, open a new tab, search for the organisation or author's name, and see what other sources say about them.

Researchers call this lateral reading: "leaving a site after a quick scan and opening up new browser tabs in order to judge the credibility of the original site."

Why leaving the page works

The evidence comes from a study by Sam Wineburg and Sarah McGrew of Stanford University, published in Teachers College Record in 2019. They watched three groups evaluate real websites while thinking aloud: 10 PhD historians, 10 professional fact checkers and 25 Stanford undergraduates.

The groups behaved very differently.

Historians and students tended to read vertically — in the study's words, "staying within a website". The researchers found they "often fell victim to easily manipulated features of websites, such as official-looking logos and domain names."
Fact checkers read laterally. Faced with an unfamiliar organisation's site, they left its landing page in about half a minute on average and went to find out who ran it.

The authors summed it up in one line: "Fact checkers, in short, learned most about a site by leaving it."

Two details are worth keeping, because they stop the story being told too neatly:

Fact checkers were not faster at everything. On an open search task, the students found the answer first. The authors note the fact checkers "were the most selective when it came to the sites they visited and took the most time to verify their answers."
Not every historian read vertically. Two of the ten historians behaved more like the fact checkers than like their colleagues.

Move 3 — Find better coverage

If the claim matters more than the source — a statistic, a news event, a scientific finding — then the original post is often the worst place to learn about it. Caulfield's advice:

"your best strategy may be to ignore the source that reached you, and look for trusted reporting or analysis on the claim"

Look for whether well-established news organisations, reference works, official bodies or specialist publications have covered the same claim, and what they say.

This does not mean you must accept whatever the majority says. Caulfield answers that question directly: "Do you have to agree with the consensus once you find it? Absolutely not!" The point is to know what the consensus is before you decide to disagree with it.

The trap: searching the headline itself

"Just search it" sounds like good advice, and it can backfire.

In a set of five experiments published in Nature (online in December 2023, in the January 2024 issue), Aslett and colleagues asked people to use online search to check news articles, including false ones. Searching increased the chance that people rated a false or misleading article as true. In their first study, the authors describe this as "a 19% increase in the probability that a respondent rated a false or misleading article as true."

What people typed mattered. When the search was the false article's own headline or web address, the results were far more likely to include unreliable sources: 77% of those searches returned at least one unreliable news link, against 21% of other searches.

So the SIFT move is not "search the claim". It is find coverage you have reason to trust — search for the topic in your own words, go to sources you already know, and look at who is reporting it.

Move 4 — Trace claims, quotes and media to the original context

A great deal of what spreads online has been cut away from where it came from. As Caulfield puts it, "Much of what we find on the internet has been stripped of context."

Tracing means going upstream until you reach the original:

A quote — find where it was first said or written: the book, the speech, the interview. Our quote checks show how often a famous line turns out to belong to someone else.
A photo — search for earlier copies of the image with a reverse image search tool, and see where and when it first appeared.
A statistic or study — find the report or paper itself, and check the claim matches what it actually found, not what a headline said it found.

What the evidence supports — and what it does not

What it supports. A study of college students in a civics course found that, after instruction in lateral reading, they "were more likely to use lateral reading to fact-check and correctly evaluate the trustworthiness of information than controls."

What it does not yet show.

The foundational study was small. Ten historians and ten fact checkers, all in the United States. The authors themselves note that small samples exaggerate differences, and that more research is needed to know whether the strategies generalise across topics, sites and searches.
The college study is limited. It came from one university and was not a randomised trial, and one of its co-authors is Caulfield himself. It says further research is needed to know whether the improvement is maintained over time.
Whether it lasts has not been established. The studies that looked are few, and they tested quite different things.
- A 2023 study of Canadian school students taught with SIFT-based lessons found that "Gains were maintained from posttest to delayed posttest" — a follow-up about a month later, but with no comparison group at that stage.
- A 2026 randomised controlled trial with 2,666 adults in Germany, published in Psychological Science, tested a short lateral-reading video and a claim-search strategy. Both showed small improvements, not all of them statistically distinguishable from zero. At a two-week follow-up, discernment had improved in every group, including the control group, and "differences between the intervention and control groups were no longer statistically distinguishable." In a report on the study the researchers note that the trained groups' scores did not decay; the control group caught up.
- A 2025 study of 459 upper-secondary students in Swedish classrooms found that a fact-checking workshop did not significantly improve their ability to evaluate credible and misleading news, and students "did not report increased use of digital verification tools three months after the intervention".
- These do not cancel each other out, because they tested different doses on different people: a course of classroom lessons, a five-minute video, a single workshop. What none of them shows is a lasting benefit measured against a comparison group.
The sixty seconds is a rule of thumb, not a finding. It is Caulfield's guidance for investigating a source, not a measured result.

That is a fair place to leave it: a sensible routine, backed by promising but modest evidence, that costs you a minute and can save you from repeating something false.

The routine, on one screen

Stop. Do I know this source? If not, don't share yet.
Investigate. Open a new tab. Who runs this site or account, according to other sources?
Find better coverage. What do sources I already trust say about this claim?
Trace. Where did the quote, photo or number first appear, and does it say what the post claims?
Don't search the headline word for word. Describe the claim in your own words instead.

Sources

Mike Caulfield, "SIFT (The Four Moves)", Hapgood, 19 June 2019.
University of Chicago Press, catalogue page for Mike Caulfield and Sam Wineburg, Verified (2023).
Sam Wineburg and Sarah McGrew, "Lateral Reading and the Nature of Expertise: Reading Less and Learning More When Evaluating Digital Information", Teachers College Record 121(11), 2019 (authors' accepted manuscript, Stanford Digital Repository).
Brodsky, Brooks, Pavlounis and Johnston, "Instruction Increases Canadian Students' Preference for and Use of Lateral Reading Strategies to Fact-Check Online Information", AERA Open 9, 2023.
J. E. Brodsky and colleagues, "Improving college students' fact-checking strategies through lateral reading instruction in a general education civics course", Cognitive Research: Principles and Implications 6, article 23, 2021.
L. Oswald, A. Kozyreva, S. M. Herzog, P. L. Nickl and R. Hertwig, "Boosting Media Literacy Using Lateral Reading and Online Search Interventions", Psychological Science (published online July 2026).
T. Nygren and E. Efimova, "Investigating the long-term impact of misinformation interventions in upper secondary education", PLOS ONE (2025).
K. Aslett and colleagues, "Online searches to evaluate misinformation can increase its perceived veracity", Nature 625 (published online December 2023).

Checked September 2026.

How to Spot a Fake Quote or an Invented Source, Including Ones Made Up by AI

Hacks Vitae — Thu, 17 Sep 2026 00:00:00 GMT

Two kinds of fake travel well online. One is the misattributed quote: a real sentence, credited to someone who never said it. The other is the invented source: a book, study or court case that does not exist at all. AI chatbots can produce the second kind by accident.

Both can be caught with the same habit: go back to the original, and check that it exists and says what it is claimed to say.

Why quotes end up with the wrong person

Garson O'Toole, the researcher behind the website Quote Investigator, traces where quotations come from. In an interview with NPR, he explained why so many witty lines get credited to Mark Twain: "If you preface a quotation by saying it's from Twain, then people are prepared to laugh at it."

The writer Ralph Keyes, author of The Quote Verifier, has a name for the people words attach themselves to. A newspaper review of his work reports that "Keyes calls some of these people 'flypaper figures' because words tend to stick to them." A second review describes another pattern he found: quotations that start out long and clumsy tend to end up "shorter, more graceful and more melodious in the retelling."

Put those together and you get the typical fake quote: a polished line, attached to a famous name, with no date and no source. Our quote checks show the pattern many times over.

How to find where a quote first appeared

The aim is the earliest appearance, with a date — not the most popular one.

Internet Archive offers full-text search inside digitised books. Its help page explains that "This allows you to search inside books and other text items. It uses the OCRed text file derived from the processing of uploaded text formatted files". Because it relies on scanned text, it is worth trying spelling variations.
HathiTrust searches the text of its digitised collection; a University of California help page says "The HathiTrust corpus consists of 17.5 million volumes comprising 6.2 billion pages." Some are marked "Limited (search-only)": the same page notes that "No one may access 'Limited (search-only)' volumes for reading purposes." Our reading: even a search-only hit can show that the words appear in a book with a known date, which may be all you need.
Google Books also searches the full text of books. Its About page says "Book Search works just like web search", and where rights allow, "you'll be able to see a preview of the book, and in some cases the entire text."

What you are looking for: the oldest dated source that contains the words, and whether it credits the person the quote is now pinned on. If the earliest version is decades after the person died, or credits someone else, you have your answer.

Invented sources: what the evidence shows about AI

The researchers Kalai, Nachum, Vempala and Zhang describe the underlying behaviour: "Like students facing hard exam questions, large language models sometimes guess when uncertain, producing plausible yet incorrect statements instead of admitting uncertainty." A guessed reference can look exactly like a real one — author, title, journal, year.

The research.

In a 2023 study in Scientific Reports, Walters and Wilder had two versions of ChatGPT produce 84 short papers on 42 topics and checked all 636 citations. They found that "55% of the GPT-3.5 citations but just 18% of the GPT-4 citations are fabricated." Even among the real citations, "43% of the real (non-fabricated) GPT-3.5 citations but just 24% of the real GPT-4 citations include substantive citation errors."
A 2023 study in Cureus by Bhattacharyya and colleagues checked references in medical articles generated by ChatGPT and found that "47% were fabricated, 46% were authentic but inaccurate, and only 7% were authentic and accurate."

Both studies tested 2023 versions of these tools. Newer models may do better or worse; the point is that a citation from a chatbot is a lead to check, not a source.

Why it happens. Kalai and colleagues' 2025 paper argues that "language models hallucinate because the training and evaluation procedures reward guessing over acknowledging uncertainty." That is their argument rather than a settled consensus.

When invented sources reach a courtroom

United States, 2023. In Mata v. Avianca, a federal judge in New York found that lawyers had submitted "non-existent judicial opinions with fake quotes and citations created by the artificial intelligence tool ChatGPT" and "then continued to stand by the fake opinions after judicial orders called their existence into question." The court was careful about what it was punishing: "Poor and sloppy research would merely have been objectively unreasonable." It found bad faith "based upon acts of conscious avoidance and false and misleading statements to the Court" and imposed "A penalty of $5,000". It also made a point worth keeping: "there is nothing inherently improper about using a reliable artificial intelligence tool for assistance. But existing rules impose a gatekeeping role on attorneys to ensure the accuracy of their filings."

England, 2025. In a judgment of 6 June 2025 covering two cases, the High Court described what these tools can do: "The responses may make confident assertions that are simply untrue. They may cite sources that do not exist." In one of the two cases, a schedule prepared for the court listed forty-five citations that had been put before it, and "In eighteen instances, the case cited does not exist." The court referred lawyers in both cases to their professional regulators.

These are not isolated. A database maintained by the researcher Damien Charlotin, which records court decisions where a court found or implied reliance on hallucinated material (and some where AI use was alleged but not confirmed), listed more than 2,000 cases when we checked it in September 2026.

Five quick checks for any source

Does the DOI work? Many journal articles have a DOI, a permanent identifier. Paste it into doi.org. The DOI system's own documentation says it shows a "DOI Name Not Found" error page when it "cannot find" the identifier. Our advice: a DOI that works but leads to a different paper is just as much a warning sign.
Can Crossref find it? Crossref's Simple Text Query "allows you to retrieve DOI names for journal articles, books, and chapters by cutting and pasting a reference or reference list into the query box". Our reading: no match does not prove a source is fake — not everything has a Crossref DOI — but a confident-looking journal citation with no match deserves a closer look.
Does the journal exist? Periodicals are given an ISSN, which you can look up in the ISSN Portal. Be careful what that proves: the ISSN International Centre notes that an ISSN "does not guarantee the quality or validity of the contents".
Do the details fit? In the Cureus study, fabricated and inaccurate references often had the wrong volume, page numbers or year. Check that the volume and pages exist in that journal, in that year.
Can you find the author? Researchers can register for an ORCID, which the organisation describes as "a unique, persistent identifier free of charge to researchers". Our reading: not every researcher has one, so a missing ORCID proves nothing on its own.

Quotes on pictures

For a quote shared as an image, two tools help.

Google Lens lets you "Use your camera or an image to search", and can copy text out of an image so you can search for the words themselves.
TinEye is a reverse image search engine that "uses image recognition technology rather than keywords, metadata or watermarks"; you submit an image "to find out where it came from, how it is being used, if modified versions of the image exist". One caution from its own help pages: "there might be a long gap between when the image was posted and when our crawlers first found it" — its "first found" date is not necessarily when the image was first posted.

The short routine

Copy the exact words and search for the earliest dated version in book and newspaper archives.
Check that the named person could have said it — and that the earliest source credits them.
For any study or book, check the DOI, the journal and the details.
Treat any citation from an AI chatbot as unverified until you have opened the source yourself.
If you cannot find the original, don't pass it on with a name attached.

Sources

Quote Investigator, About page (Garson O'Toole).
NPR, All Things Considered, interview with Garson O'Toole, 4 April 2017.
Frank Herron, review of Ralph Keyes's The Quote Verifier, Syracuse Post-Standard (reprinted on ralphkeyes.com).
David Murdock, review of Keyes, The Gadsden Times, 22 July 2011 (reprinted on ralphkeyes.com).
Internet Archive Help Center, "Search — A Basic Guide", Full-Text Search.
University of California HathiTrust Help Center, "Full-text, full view, better searching in HathiTrust".
Google Books, "About Google Books".
Walters and Wilder, Scientific Reports 13, 14045 (2023).
Bhattacharyya and colleagues, Cureus 15(5), e39238 (2023).
Kalai, Nachum, Vempala and Zhang, "Why Language Models Hallucinate", arXiv:2509.04664 (2025).
Mata v. Avianca, Inc., No. 22-cv-1461 (PKC), Opinion and Order on Sanctions, S.D.N.Y., 22 June 2023.
Ayinde v London Borough of Haringey and Al-Haroun v Qatar National Bank [2025] EWHC 1383 (Admin), 6 June 2025.
Damien Charlotin, AI Hallucination Cases database.
DOI Foundation, "DOI Resolution Documentation" factsheet.
Crossref, "Simple Text Query" documentation.
ISSN International Centre, "What is an ISSN?".
ORCID, "What is ORCID?".
TinEye Help, "What is TinEye?" and "What does TinEye 'first found on' mean?".
Google Lens.

Checked September 2026.

How to Read a Study Without a Science Degree: 9 Questions to Ask

Hacks Vitae — Thu, 17 Sep 2026 00:00:00 GMT

You do not need to be a scientist to read a study sensibly. You need to know which parts to look at, and which common assumptions to drop.

These nine questions work for most research you will meet in the news or on social media. Where we could, we used sources from outside medicine — this is a guide to reading research, not advice about any health decision.

1. Did you read past the abstract?

A scientific paper usually has the same parts. The University of Michigan Library's guide describes them plainly: abstracts "are always found at the beginning of an article and provide a basic summary or roadmap to the article."

The part that matters most for a careful reader is the methods section. As the same guide puts it: "It is necessary for other researchers to understand the methods used so that they can replicate the study." Our advice follows from that: the methods are where you find out who was studied, how many, and what was actually done.

Then keep the results and the discussion apart. The guide notes that "The discussion section is where you will find the researcher's interpretation of the results." The results are what they found; the discussion is what they think it means.

2. What kind of study is it?

The biggest single question is whether the researchers assigned people to conditions, or simply observed them.

In an experiment with random assignment, chance decides who gets which condition. The open statistics textbook Introductory Statistics from OpenStax explains why that matters: "When subjects are assigned treatments randomly, all of the potential lurking variables are spread equally among the groups." (That is a textbook simplification: in practice random assignment makes the groups similar on average rather than identical.) An observational study cannot do that, so a link it finds may be driven by something else — the trap covered in our guide to numbers that mislead.

Two other labels you will meet:

A systematic review, in a definition quoted by KTDRR from John Last's Dictionary of Epidemiology, is "the application of procedures that limit bias in the assembly, critical appraisal, and synthesis of all relevant studies on a particular topic."
A meta-analysis is "the statistical synthesis of the data from separate but comparable studies, leading to a quantitative summary of the pooled results."

A meta-analysis combines results. It cannot make weak studies strong — which is our reading, and the reason the next question still applies.

3. Was it peer reviewed — and what does that mean?

The organisation Sense about Science describes peer review as a process that "subjects scientific research papers to independent scrutiny by other qualified scientific experts (peers) before they are made public."

Two limits are worth knowing:

It is not a fraud check. The same organisation's guide says: "Peer review is not a fraud detection system." It adds that referees "are likely to detect some wrongdoing, such as copying someone else's research or misrepresenting data", but deliberate fraud can pass.
It is a start, not a finish. In its words: "Publication of a peer-reviewed paper is just the first step".

Some studies appear first as preprints, before any peer review. arXiv, a preprint server covering physics, mathematics, computer science and several other fields, moderates submissions, but is explicit: "Please note that the arXiv moderation process is not a peer-review process." A preprint can be excellent; it simply has not been through that check yet.

4. How big was the sample?

Small studies do more than miss real effects. A 2013 review in Nature Reviews Neuroscience by Button and colleagues warned that "low power also reduces the likelihood that a statistically significant result reflects a true effect". In other words, a dramatic finding from a small study deserves more caution, not less.

5. What does the p-value tell you — and what doesn't it?

In 2016 the American Statistical Association issued a statement on p-values, setting out six principles. They are short enough to quote in full:

"P-values can indicate how incompatible the data are with a specified statistical model."
"P-values do not measure the probability that the studied hypothesis is true, or the probability that the data were produced by random chance alone."
"Scientific conclusions and business or policy decisions should not be based only on whether a p-value passes a specific threshold."
"Proper inference requires full reporting and transparency."
"A p-value, or statistical significance, does not measure the size of an effect or the importance of a result."
"By itself, a p-value does not provide a good measure of evidence regarding a model or hypothesis."

The association's executive director, Ron Wasserstein, summed up the spirit of it: "The p-value was never intended to be a substitute for scientific reasoning".

The practical upshot: "statistically significant" does not mean "true", and it does not mean "big".

6. How big is the effect?

Because significance says nothing about size (principle 5 above), look for the effect size — how large the difference or relationship actually was — and how precisely it was estimated.

Psychology's reporting standards, published by the American Psychological Association, ask authors to report "Effect-size estimates and confidence intervals on those estimates that correspond to each inferential test conducted, when possible". If a study reports a significant result but no effect size, you cannot tell whether the finding matters in practice.

7. Has anyone repeated it?

In 2015 the Open Science Collaboration reported repeating 100 studies from three psychology journals. "Ninety-seven percent of original studies had statistically significant results. Thirty-six percent of replications had statistically significant results". The replication effects "were half the magnitude of original effects, representing a substantial decline." That 36% is one of several measures the project used, and what the results mean is itself DISPUTED. In 2016 Gilbert and colleagues argued that the data "are consistent with the opposite conclusion, namely, that the reproducibility of psychological science is quite high." The original team replied that "both optimistic and pessimistic conclusions about reproducibility are possible, and neither are yet warranted."

The other side matters too. The US National Academies of Sciences, Engineering, and Medicine caution that "nor does a single failed replication conclusively refute the original claims" — and, in the same sentence, that a successful replication does not guarantee the original was right. Our reading: what builds confidence is several good studies pointing the same way. Our cognitive biases check shows what that looks like for fourteen famous findings.

Two newer practices are designed to reduce these problems, and are worth looking for:

Pre-registration. The Center for Open Science describes it as "specifying your research plan in advance of your study and submitting it to a registry". The reason: "the same data cannot be used to generate and test a hypothesis".
Registered Reports. A journal reviews the study plan before any data exist and, if the plan is approved, "the journal virtually guarantees publication if the authors conduct the experiment in accordance with their approved protocol". According to the Center for Open Science, "over 300 journals use the Registered Reports publishing format either as a regular submission option or as part of a single special issue" (as read in September 2026).

8. Who paid for it, and who wrote the headline?

Funding and conflicts of interest. In the wording of the International Committee of Medical Journal Editors, "The potential for conflict of interest and bias exists when professional judgment concerning a primary interest (such as patients' welfare or the validity of research) may be influenced by a secondary interest (such as financial gain)." Its recommendations ask authors to state "Sources of support for the work, including sponsor names along with explanations of the role of those sources". Look for a funding or disclosure statement near the end of the paper. Our advice: funding does not make a study wrong; it is a reason to read the methods more closely.

Press releases. A 2014 study in The BMJ by Sumner and colleagues examined press releases about health and biomedical research from UK universities and found that "Exaggeration in news is strongly associated with exaggeration in press releases." It also found "there was little evidence that exaggeration in press releases increased the uptake of news". That study looked only at health research. Our advice, in any field: go back to the paper, not the press release.

9. Has it been retracted?

Papers are sometimes retracted after publication.

One public record of retractions is the Retraction Watch database. In September 2023 Crossref announced that "the Retraction Watch database has been acquired by Crossref and made a public resource". The database's user guide notes that "The more fields you fill in, the more specific the search" — searching by the paper's DOI or title is the quickest check.

A note on the "evidence pyramid"

You may see research drawn as a pyramid. Murad and colleagues describe the traditional version, with weaker designs such as case series near the bottom and systematic reviews at the top. It is a useful rough guide, but it has critics. A 2016 paper by Murad and colleagues argued that "Study design alone appears to be insufficient on its own as a surrogate for risk of bias", and proposed to "remove systematic reviews from the top of the pyramid and use them as a lens through which other types of studies should be seen". Our reading: a well-run observational study can be more informative than a poorly run experiment.

The nine questions, on one screen

Did I read the methods, not just the abstract?
Experiment or observation?
Peer reviewed, preprint, or press release only?
How many people or cases?
Am I reading more into the p-value than it says?
How big was the effect?
Has it been repeated — and was it pre-registered?
Who paid, and does the headline match the paper?
Has it been retracted?

Sources

University of Michigan Library, "Reading Scholarly Articles".
OpenStax, Introductory Statistics 2e, section 1.4, "Experimental Design and Ethics".
KTDRR / NCDDR, FOCUS Technical Brief 16, "The Campbell Collaboration" (definitions quoted from Last, 2001, via Chalmers, Hedges and Cooper, 2002).
Sense about Science, I don't know what to believe (first edition 2005, updated 2017).
arXiv, "arXiv moderation" help page.
Button, Ioannidis, Mokrysz, Nosek, Flint, Robinson and Munafò, "Power failure: why small sample size undermines the reliability of neuroscience", Nature Reviews Neuroscience 14 (2013).
American Statistical Association, statement on statistical significance and p-values (press release, 7 March 2016).
Appelbaum and colleagues, "Journal Article Reporting Standards for Quantitative Research in Psychology", American Psychologist 73 (2018).
Open Science Collaboration, "Estimating the reproducibility of psychological science", Science 349 (2015); Gilbert, King, Pettigrew and Wilson, comment, Science (2016); Anderson and colleagues, response to comment, Science (2016).
National Academies of Sciences, Engineering, and Medicine, Reproducibility and Replicability in Science (2019).
Center for Open Science, "Preregistration" and "Registered Reports".
International Committee of Medical Journal Editors, Recommendations: author responsibilities and conflicts of interest.
Sumner and colleagues, "The association between exaggeration in health related science news and academic press releases: retrospective observational study", The BMJ 349 (2014).
Crossref blog post on acquiring the Retraction Watch database (12 September 2023); Retraction Watch Database User Guide.
Murad, Asi, Alsawas and Alahdab, "New evidence pyramid", Evidence-Based Medicine 21 (2016).

Checked September 2026.

Logical Fallacies Cheat Sheet: 16 Common Ones, and When They Are Not Fallacies

Hacks Vitae — Thu, 17 Sep 2026 00:00:00 GMT

A logical fallacy is a flaw in the reasoning behind a claim. It is not the same as the claim being false. That distinction is the most useful thing on this page, so it comes first.

A bad argument can still land on a true conclusion. If someone argues badly that the Earth goes round the Sun, the Earth still goes round the Sun. Spotting a fallacy tells you the argument does not support its conclusion. It tells you nothing about whether the conclusion is true.

Each entry below has a plain definition, an invented everyday example, and — where the sources support it — when the same move is not a fallacy at all. The examples are made up and do not describe anyone real.

Where the list comes from

The study of fallacies is old. The Stanford Encyclopedia of Philosophy puts it simply: "The history of the study of fallacies begins with Aristotle's work, On Sophistical Refutations." That short work describes thirteen ways an argument can look like a refutation without being one, split into those that depend on language and those that do not.

At least four entries in this cheat sheet have ancestors on Aristotle's list. In W. A. Pickard-Cambridge's English translation they are "ambiguity" (equivocation), "assuming the original conclusion" (begging the question), "stating as cause what is not the cause" (false cause), and "the making of more than one question into one" (the loaded question).

Attacks on the person

1. Ad hominem

What it is. In the Internet Encyclopedia of Philosophy's definition: "Your reasoning contains this fallacy if you make an irrelevant attack on the person arguing and suggest that this attack undermines the argument itself."

Example. "Why should we listen to his plan for the library opening hours? He can't even keep his own desk tidy."

When it isn't. The encyclopedia draws a useful line: a person's characteristics may affect their credibility as an authority without affecting the quality of their argument. Our reading: facts about a person matter when you are relying on their word, not when you are judging their reasoning.

2. Tu quoque ("you too")

Tu quoque is Latin for "you too". The Internet Encyclopedia of Philosophy classes it as "a kind of Ad Hominem Circumstantial Fallacy": dismissing an argument because the person making it does not follow it themselves.

Example. "You say I should stop checking my phone at dinner, but you check yours at breakfast."

When it isn't. Hypocrisy is a fair reason to look harder at what someone says. The encyclopedia puts the limit precisely: discovering that a speaker is a hypocrite is a reason to be suspicious, "but it is not a sufficient reason to discount it."

A related modern word is whataboutism. Merriam-Webster dates whataboutism to 1978. The Irish Times traces the idea to a letter it published on 30 January 1974 describing "the Whatabouts", and the word Whataboutery to its columnist John Healy three days later.

3. Genetic fallacy

What it is. Judging a claim by where it came from when that origin is beside the point. The encyclopedia's wording: it "attempts to discredit or support a claim or an argument because of its origin (genesis) when such an appeal to origins is irrelevant."

Example. "That recipe tip came from a cartoon, so it can't work."

When it isn't. The definition contains its own exception — "when such an appeal to origins is irrelevant." Our reading: a source's track record is often relevant, so a claim that comes only from an account known for invented stories deserves more checking than one from a careful reference work.

Distorting the question

4. Straw man

What it is. To "attribute an easily refuted position to your opponent, one that the opponent would not endorse, and then proceed to attack the easily refuted position."

Example. "She wants the office to recycle more? So she thinks we should sort every staple into its own bin."

Origin. The image is old. The Stanford Encyclopedia notes that the idea may go back to Isaac Watts's Logick (1724), which describes disputers setting up weak versions of opponents' views and defeating them. Merriam-Webster dates the phrase straw man in this sense to 1840, so the term itself is not Watts's coinage.

5. Red herring

What it is. "A digression that leads the reasoner off the track of considering only relevant information."

Example. Asked why the group project is late, someone replies with a long account of how hard the software update was on everyone's laptops.

Origin — DISPUTED. Merriam-Webster traces the phrase to the practice of drawing a red herring across a trail to confuse hunting dogs. The etymologist Michael Quinion, drawing on the Oxford English Dictionary, credited the figurative use to a story William Cobbett published on 14 February 1807. But the Wordorigins.org entry on the phrase cites figurative uses from the 1680s onwards, and concludes that "both the figurative use and the idea that it is grounded in distracted dogs predates" Cobbett. The same entry says the real hunting practice was dragging a red herring along a trail "to train or exercise hounds and horses", and that stories of using one to distract hounds seem to have been invented. We lean towards the Wordorigins account: it cites dated uses older than Cobbett's, and both it and Quinion reject the idea that throwing hounds off the scent was a real practice.

6. Loaded question

What it is. "Asking a question in a way that unfairly presumes the answer."

Example. "Have you stopped leaving the kitchen in a mess?" — either answer admits the mess.

When it isn't. A question that assumes something both sides already accept is just a question. The problem is presuming something controversial.

7. False dilemma

What it is. A reasoner "who unfairly presents too few choices and then implies that a choice must be made among this short menu of choices."

Example. "Either we repaint the whole house this weekend or we accept that it will look shabby forever."

When it isn't. The key word in the definition is unfairly. Some choices really do have only two options — a light switch is on or off. Our reading: it is a fallacy only when real alternatives are being hidden.

Bad reasoning about evidence

8. Appeal to authority

What it is. Treating a claim as true because of who says it.

When it isn't — which is usually. This entry needs the biggest caveat of all. The Internet Encyclopedia of Philosophy is blunt: "Most reasoning of this kind is not fallacious, and much of our knowledge properly comes from listening to authorities." It becomes a fallacy when the "authority" is speaking outside their field, is not trustworthy, is misquoted, or when genuine experts disagree.

Example of the fallacy. "A famous footballer says this phone camera takes the sharpest photos, so it must."

9. Appeal to popularity (bandwagon)

What it is. Suggesting "too strongly" that a claim is correct simply because many people believe it.

Example. "Everyone in the neighbourhood waters their lawn at noon, so that must be the right time."

When it isn't. The encyclopedia notes that what most people believe is, for that reason, somewhat likely to be true, all things considered. Popularity is weak evidence, not no evidence. The fallacy is treating it as proof.

10. Post hoc ergo propter hoc

The Latin means "after this, therefore because of this": B followed A, so A caused B.

Example. "I wore my blue socks and the bus came on time. Blue socks it is."

When it isn't. The encyclopedia is careful here: "Correlations are often good evidence of causal connection, so the fallacy occurs only when the leap to the causal conclusion is done" too hastily. Something happening first is a reason to investigate, not a conclusion.

11. Hasty generalisation

What it is. In the encyclopedia's words, "the key error is to overestimate the strength of an argument that is based on too small a sample for the implied confidence level or error margin."

Example. "I've been to two cafés in this town and both were slow. The whole town is slow."

When it isn't. Generalising is how we learn anything. Our reading: it is fine when your confidence matches the size and fairness of the sample — "both cafés I tried were slow" is honest; "the whole town" is not.

12. Burden of proof

What it is. "Committing the error of trying to get someone else to prove you are wrong, when it is your responsibility to prove you are correct."

Example. "There's a hidden room behind the school gym. Prove there isn't."

When it isn't. It is reasonable to set aside, at least provisionally, an unlikely claim that comes with no adequate evidence. Our reading: declining to believe it is not the same as claiming to have disproved it.

13. Slippery slope

What it is. In the Stanford Encyclopedia's description, arguing that from a starting point "one can by a series of incremental inferences arrive at an undesirable conclusion", so the first step must be refused.

Example. "If we let the children stay up ten minutes later tonight, next they'll want to stay up all night, and then they'll never go to school again."

When it isn't. Not every chain of consequences is a fallacy. The Internet Encyclopedia puts the test clearly: "Arguments of this form may or may not be fallacious depending on the probabilities involved in each step." A chain of steps that are each very likely can be a sound warning. A chain of shaky steps multiplies the doubt at every link.

Tricks with words

14. Begging the question (circular reasoning)

What it is. "A form of circular reasoning in which a conclusion is derived from premises that presuppose the conclusion." Its Latin name is petitio principii.

Example. "This is the most reliable guidebook, because it says so in the introduction, and a reliable guidebook wouldn't lie."

When it isn't. In everyday English, "that begs the question" is also used to mean "that raises the question", and Merriam-Webster lists both senses. That everyday use is not a fallacy — it is just a different meaning of the phrase.

15. Equivocation

What it is. "The illegitimate switching of the meaning of a term that occurs twice during the reasoning."

Example. "A feather is light. What is light cannot be dark. So a feather cannot be dark."

16. No true Scotsman

What it is. Protecting a generalisation from a counterexample by redefining the group — "no true member would do that." The Internet Encyclopedia describes it as a kind of ad hoc rescue of a generalisation.

Example. "No real gardener uses weedkiller." — "My aunt has won prizes for her roses and uses it." — "Then she isn't a real gardener."

Origin. The name comes from a story told by the philosopher Antony Flew. The Fallacy Files lists his discussions of it in several books, including An Introduction to Western Philosophy (1971).

When it isn't. Our reading: sometimes a person is simply clarifying what they meant all along. The warning sign is a definition that changes only after a counterexample appears.

Two warnings before you use this list

1. The fallacy fallacy. As The Fallacy Files puts it, calling an argument fallacious "says nothing about the truth or falsity of the conclusion." Winning the argument about the reasoning is not winning the argument about the facts.

2. Most fallacies sit next door to good reasoning. In a 2015 paper in the journal Argumentation, Maarten Boudry, Fabio Paglieri and Massimo Pigliucci argued that "almost every known type of fallacy is a close neighbor to sound inferences or acceptable moves in a debate." That is why every entry above has a "when it isn't" line. A list of names is a starting point for thinking, not a replacement for it.

Does learning about manipulation make you better at spotting it?

DISPUTED. It depends on what you measure.

Recognition in tests: yes, on the evidence so far. A 2022 study in Science Advances by Roozenbeek and colleagues found that short videos explaining manipulation techniques — including false dichotomies and ad hominem attacks — improved people's ability to recognise them, across six randomised experiments and a field study run on YouTube. The authors also reported better discernment of trustworthy from untrustworthy content and better-quality sharing decisions.
Behaviour in a realistic feed: not shown. A 2025 study in PNAS Nexus taught people about one technique, emotional manipulation, and measured engagement in a simulated social media feed. The lesson reduced engagement only when the feed held nothing but relevant synthetic posts; adding real posts, or posts using other techniques such as ad hominem attacks, "appeared to nullify the effect." The authors note there is "substantial evidence that inoculation increases technique recognition when directly assessed" — the open question is whether it transfers.
Telling reliable from unreliable news: DISPUTED, leaning positive. A 2023 reanalysis of five studies by Modirrousta-Galian and Higham found two inoculation games "did not improve discrimination" between true and fake news, but made people answer "false" more often to everything. A 2026 reanalysis by Simchon and colleagues, covering 33 experiments with 37,025 participants and using a different statistical method, found that "both gamified and video-based interventions consistently improve discrimination between reliable and unreliable news, without increasing response bias."

Our reading: the larger 2026 reanalysis carries more weight than the five-study critique, because it covers far more experiments — so learning to name manipulation techniques probably does help, at least in study conditions. What has not been shown is that the benefit survives the pace and mix of a real social media feed.

Sources

"Fallacies", Stanford Encyclopedia of Philosophy (substantive revision 2024).
Bradley Dowden, "Fallacies", Internet Encyclopedia of Philosophy (entries for each fallacy named above).
Aristotle, On Sophistical Refutations, chapters 4–5, trans. W. A. Pickard-Cambridge.
Merriam-Webster, entries for "straw man", "red herring", "whataboutism", "tu quoque" and "beg the question".
Michael Quinion, "Red herring", World Wide Words.
"In a Word… Whataboutism", The Irish Times.
The Fallacy Files, "No True Scotsman".
The Fallacy Files, "Fallacy fallacy".
Maarten Boudry, Fabio Paglieri and Massimo Pigliucci, Argumentation (2015).
Roozenbeek and colleagues, "Psychological inoculation improves resilience against misinformation on social media", Science Advances (2022).
S. Y. N. Wang, S. C. Phillips, K. M. Carley, H. Lin and G. Pennycook, "Limited effectiveness of psychological inoculation against misinformation in a social media feed", PNAS Nexus (2025).
A. Modirrousta-Galian and P. A. Higham, "Gamified inoculation interventions do not improve discrimination between true and fake news", Journal of Experimental Psychology: General (2023).
Wordorigins.org, "Red Herring".
Simchon and colleagues, "A Signal Detection Theory Meta-Analysis of Psychological Inoculation Against Misinformation", Current Opinion in Psychology 67 (2026).

Checked September 2026.

Related: How to check something online: the SIFT method · Overcoming survivorship bias · First principles thinking

Mansa Musa's Gold: What the Medieval Sources Say About His Pilgrimage to Mecca

Hacks Vitae — Thu, 17 Sep 2026 00:00:00 GMT

In 1324 Mansa Musa, the ruler of Mali, passed through Cairo on his pilgrimage to Mecca, and people there were still talking about his gold years later. Today he is sometimes said to have been the richest person who ever lived. The medieval writers who describe the journey do record enormous gifts of gold. They also disagree about the numbers, pass most of them on second-hand, and record that he ran out of money.

Who told the story

The main accounts were written by people who questioned others who had met him.

al-Umari wrote his Masalik al-absar in 1337, according to UNESCO's General History of Africa. He had not seen the king himself: "From the beginning of my coming to stay in Egypt I heard talk of the arrival of this sultan Musa on his Pilgrimage", he writes, and he questioned the official who had received him and Ibn Amir Hajib, who was governor of Cairo at the time of the visit.
Ibn Khaldun, the North African historian, heard from two informants: a man who joined the king's party on its way home, and the interpreter of Musa's people in Cairo.
al-Maqrizi, writing later, gives more figures, but his French translator notes that he seems to have drawn on al-Umari and Ibn Khaldun, "mais il a puisé aussi à d'autres sources" (but he also drew on other sources).
Ibn Battuta stayed in Mali's capital in 1352–53 and heard stories about the king on his travels there.
al-Sa'di's Tarikh es-Soudan, a chronicle of the western Sudan, is dated by UNESCO's General History to 1656, more than three centuries after the journey.

How much gold did he bring?

PARTLY TRUE. He brought a great deal of gold. The sources do not agree how much, and none gives a weight we can convert.

Ibn Khaldun's Cairo informant, the interpreter al-Hajj Yunus, said the king arrived "avec quatre-vingts charges de poudre d'or, pesant chacune trois quintaux (kintar)": with eighty loads of gold dust, each weighing three qintars (our translation).
Ibn Amir Hajib told al-Umari a different number: the king left his country "avec cent charges d'or", with a hundred loads of gold, and spent them on the journey.
Ibn Khaldun gives the hundred too, in a second passage introduced with "On rapporte" (it is reported): a hundred loads of gold dust, each of three qintars. al-Umari's informant gives no weight for a load, and no source here says what a qintar weighed, so none of these figures can be turned into kilograms.

In Cairo, the official who received him told al-Umari that Musa "forwarded to the royal treasury many loads of unworked native gold and other valuables", and that "He left no court emir nor holder of a royal office without the gift of a load of gold." al-Maqrizi says the king had "environ cinquante mille dinars", about fifty thousand dinars, presented to the sultan before meeting him.

Did 60,000 people travel with him?

PARTLY TRUE. He travelled with thousands of people. The famous 60,000 comes from the latest source.

al-Maqrizi: "plus de dix mille de ses sujets", more than ten thousand of his subjects.
Ibn Khaldun's informant who travelled with the party: "Douze mille jeunes esclaves", twelve thousand young slaves in brocade and Yemeni silk, carried the king's belongings.
al-Maqrizi again, introducing it with "On dit" (it is said): fourteen thousand young slave women for the king's personal service.
al-Sa'di: "soixante mille", sixty thousand men, with five hundred slaves in front of the king, "chacun d'eux tenant à la main une baguette d'or", each holding a gold staff.

al-Sa'di's chronicle also puts the pilgrimage "dans les premières années du neuvième siècle de l'hégire", the first years of the ninth Islamic century, which began more than seventy Islamic years after 724, the year Ibn Khaldun gives. Its French translator, O. Houdas, corrected it in a note: "elle est de 724 (1324) d'après Ibn-Khaldoun". UNESCO's General History adds its own caution: "the figures given by the Arab writers are probably exaggerated".

Did he crash the price of gold in Egypt?

PARTLY TRUE. The price fell. The only credible figures for how far, and for how long, come from one writer.

The official who received the king told al-Umari that the Cairenes "exchanged gold until they depressed its value in Egypt and caused its price to fall." al-Umari then gives his own figures. Before the visit, "The mithqal did not go below 25 dirhams and was generally above". Afterwards: "The mithqal does not exceed 22 dirhams or less. This has been the state of affairs for about twelve years until this day".

By our arithmetic, a fall from at least 25 to at most 22 is a fall of at least 12 percent, lasting about twelve years on al-Umari's account.

al-Maqrizi gives a far bigger figure: "Le dinär descendit jusqu'à six dirhems", the dinar fell to six dirhams. His French translator, Gaudefroy-Demombynes, calls this "invraisemblable et contraire à l'indication donnée par le masälik": implausible, and contrary to al-Umari. We follow al-Umari, who lived in Egypt within a generation of the visit and gives a price for before and after. On his figures it was a real fall. It was not a collapse.

He ran out of money

The same informant who described the hundred loads told al-Umari what happened to them. Having no money left when he came back to Cairo, "il dut emprunter aux marchands sous sa Caution": he had to borrow from the merchants on his own guarantee (our translation). The merchants did well: "pour trois cents dinars ils eurent sept cents dinars de gain", for three hundred dinars they made seven hundred in profit. And "Il leur en envoya plus tard le montant avec large poids": he later sent them what he owed, and generously.

Ibn Battuta heard a similar story on his travels in Mali, about a merchant from Alexandria, Siraj al-Din: "Mensa Moûça eut besoin d'argent, et il en emprunta à Siràdj eddîn" (Mansa Musa needed money, and borrowed it from Siraj al-Din). The merchant died on his way to collect, but his son went on to Mali, where "il reçut son argent" (he received his money).

Ibn Khaldun tells the story of the same lenders differently. In his Kitab al-'Ibar, the Banu al-Kuwayk lent the king 50,000 dinars, the agents sent to Mali to collect died there, and "Mansa Müsä mourut avant d'avoir payé et ils ne réussirent donc à rien tirer de lui": Mansa Musa died before paying, and they got nothing from him. The two accounts do not agree about how that debt ended. Gaudefroy-Demombynes, who translated both al-Umari and this passage, thought Ibn Khaldun's story probably accurate but incomplete, and that the king had repaid "une partie au moins de ses dettes", at least part of his debts.

Was he the richest person who ever lived?

UNVERIFIED. No surviving record could settle it, and the medieval claim was narrower than the modern one.

The Catalan Atlas, made around 1375 and now in the Bibliothèque nationale de France, shows him holding "a round gold object, perhaps an orb, coin or nugget of gold". Its caption, in the British Library's translation, calls him "the richest and noblest ruler of this whole region because of the abundance of gold that is found in his lands". Of his region, not of the world. The library attributes the atlas to Abraham Cresques; the BnF's catalogue says it was made "sinon par lui, du moins dans son entourage proche", if not by him, then by his close circle.

Against any modern ranking:

Nearly every figure was reported to the writers second-hand, and some are marked "it is said".
The sources disagree, on the gold and on the people.
No source says what a qintar weighed.
He borrowed money on the way home.
al-Umari doubted the king's own account of his kingdom. Musa had said it took about a year to cross; al-Umari preferred another informant's four months, "car Müsa Mansa peut bien avoir grossi l'importance de son royaume", because Musa may well have exaggerated the size of his kingdom.

And the gold a king gives away on a state journey is not the same thing as a private fortune.

Our reading

Mansa Musa's journey was real and extraordinary, and Cairo remembered it for years. But the version often told today — sixty thousand people, a crashed economy, the richest person in history — takes the biggest number from each source, including one written three centuries later, and leaves out the loans.

Sources

al-Umari, Masalik al-absar, excerpt in N. Levtzion and J. F. P. Hopkins (eds.), Corpus of Early Arabic Sources for West African History (Cambridge University Press, 1981), via World History Commons.
al-Umari, Masālik el absār fi mamālik el amsār, I: L'Afrique, moins l'Égypte, trans. Maurice Gaudefroy-Demombynes (Paul Geuthner, 1927), including his appendix translating passages of al-Maqrizi.
Ibn Khaldun, Histoire des Berbères et des dynasties musulmanes de l'Afrique septentrionale, trans. Baron de Slane, vol. 2 (Algiers, 1854).
Ibn Khaldun, Kitab al-'Ibar, Cairo ed., vol. V, p. 434, passage translated in Gaudefroy-Demombynes 1927 (additions to p. 93, note 1).
Voyages d'Ibn Batoutah, trans. C. Defrémery and B. R. Sanguinetti, vol. 4 (Société asiatique, 1858).
al-Sa'di, Tarikh es-Soudan, trans. O. Houdas (Ernest Leroux, 1900).
D. T. Niane (ed.), General History of Africa IV: Africa from the Twelfth to the Sixteenth Century (UNESCO, 1984).
Eleanor Jackson, "African kings on medieval and Renaissance maps", British Library Medieval Manuscripts blog (14 July 2022).
Bibliothèque nationale de France, catalogue record for Espagnol 30 (Atlas catalan).

All translations from the French are ours.

Checked September 2026.

Medieval People Did Not Think the Earth Was Flat: Where the Myth Really Came From

Hacks Vitae — Thu, 17 Sep 2026 00:00:00 GMT

The story goes that people in the Middle Ages thought the Earth was flat, and that Columbus sailed west to prove them wrong. The medieval writers themselves say otherwise. Below is what they wrote, who did disagree, and how the myth was put together centuries later.

What medieval scholars wrote

MYTH. The educated medieval world did not teach a flat Earth. Three examples, from the 8th and 13th centuries:

Bede, the English monk, explained around 723 or 725 that the Earth "is, in fact, a sphere set in the middle of the whole universe." He drew the distinction the myth forgets: "It is not merely circular like a shield [or] spread out like a wheel, but resembles more a ball, being equally round in all directions" (The Reckoning of Time, in Faith Wallis's translation, as quoted by the British Library).
Sacrobosco's De sphaera, a 13th-century astronomy treatise, has a section headed "The Earth a Sphere". It argues: "That the earth, too, is round is shown thus." One of its proofs is that the same eclipse of the Moon is seen at different hours in different places. It even explains why the Earth looks flat: "But it seems flat to human sight because it is so extensive." (Translation by Lynn Thorndike, 1949.)
Thomas Aquinas, needing an example of something two sciences can both prove, reached for this one: "the earth, for instance, is round" (Summa Theologiae I, question 1, article 1). He offered it as a plain example, without arguing for it.

Alison Hudson, writing on the British Library's medieval manuscripts blog, sums up the record: "It is actually rare to find groups in the classical, Late Antique and medieval eras who believed in the flat Earth."

The real exceptions

A few early Christian writers did reject a round Earth, and they should not be explained away.

Lactantius mocked the idea: "Or is there any one so senseless as to believe that there are men whose footsteps are higher than their heads?" (Divine Institutes III.24.)
Cosmas Indicopleustes, a 6th-century Egyptian monk, most probably from Alexandria, described a flat, oblong world modelled on the Tabernacle: "The Deity accordingly having founded the earth, which is oblong, upon its own stability". His editor dates the work to around 535–547. His influence was limited: a historian quoted by that editor says it "is only slightly, and occasionally, traceable."

How many such writers were there? The historian Jeffrey Burton Russell counts "at least two and at most five" early Christian fathers, and adds that "tens of thousands of Christian theologians, poets, artists, and scientists took the spherical view". Not every early writer took a side: Basil of Caesarea listed the shapes proposed — "spherical or cylindrical", a disc, or "a winnowing basket" — and declined to choose (Hexaemeron IX.1).

The antipodes: a different question

Some writers who accepted a round Earth still doubted that people lived on the far side of it. Saint Augustine, in The City of God (XVI.9), rejects "the fable that there are Antipodes, that is to say, men on the opposite side of the earth", while granting that even if it be "supposed or scientifically demonstrated that the world is of a round and spherical form", it does not follow that anyone lives on the other side. Bede too, the British Library notes, "denied that anyone lived in the southern hemisphere at all".

That distinction matters, because the myth was built partly by blurring it.

How the myth was built

Washington Irving, 1828. Irving's biography of Columbus dramatised a hearing at Salamanca in which, he wrote, "To his simplest proposition, the spherical form of the earth, were opposed figurative texts of scripture." Even Irving's own scene is more mixed than the legend: he adds that "Others, more versed in science, admitted the globular form of the earth", and that among their objections was that the circumference "must be so great as to require at least three years to the voyage".

Letronne, 1834. The French scholar Antoine-Jean Letronne opened by describing a time "pas encore bien loin de nous" — not yet far from us — when a map-maker "devait poser en principe que la terre était une surface plane" — had to take it as a principle that the Earth was a flat surface (our translation). He conceded that some theologians let the Earth be round, on condition that there were no antipodes. He listed "Saint Augustin, Lactance, saint Basile, saint Ambroise" among those who rejected a round Earth. Putting Augustine on that list turns doubt about the antipodes into belief in a flat Earth.

Draper, 1870s. John William Draper's History of the Conflict between Religion and Science said the philosophy of the Church Fathers held of the Earth "that it is a flat surface, over which the sky is spread like a dome", and that "the question of the shape of the earth was finally settled by three sailors".

White, 1896. Andrew Dickson White's A History of the Warfare of Science with Theology in Christendom concedes that thinkers such as Aquinas and Dante "felt obliged to accept the doctrine of the earth's sphericity." But it framed the history as a war, claiming that "the great majority of the early fathers of the Church, and especially Lactantius, had sought to crush it" — a claim the count above does not support, and White himself writes a few lines on that Ambrose and Augustine "had tolerated it".

Russell's summary: "No one before the 1830s believed that medieval people thought that the earth was flat." He names Irving (1828) and Letronne as the two who established the idea, which by the 1860s was being "served up in schools and in schoolbooks as the solemn truth". The charge against a few Church Fathers is older, though: in the 17th century Robert Burton's The Anatomy of Melancholy already said Augustine, Basil and Lactantius held the Earth "round as a trencher" — a flat plate — rather than a ball.

What the Columbus argument was about

MYTH. Columbus did not have to persuade experts that the Earth was round. As the historian of science Darin Hayton puts it, "Columbus and all his detractors knew that the earth was round."

Even Irving, whose Salamanca scene helped start the legend, wrote that the success of Columbus's undertaking "depended upon two happy errors, the imaginary extent of Asia to the east;" and the supposed smallness of the Earth. Columbus thought the world was small enough to reach Asia by sailing west. The Americas lay in between.

What we cannot know

UNVERIFIED. Almost every surviving source comes from educated people: monks, scholars, map-makers. Round-Earth ideas did reach books outside the schools — Mandeville's Travels says "that the land and the sea be of round shape and form" — but no record measures what most ordinary, unlettered people believed about the shape of the Earth, if they thought about it at all. What can be said is that the scholarly record is overwhelmingly for a sphere, and the rest is largely unrecorded.

Our reading

The flat-Earth Middle Ages is largely a 19th-century creation, though the charge against a few Church Fathers is older, and it says more about the 19th century than the medieval one: it was useful to writers who wanted a story of science defeating religion. The history has more texture — a few genuine flat-Earth writers, a real and separate debate about the antipodes, and a Columbus who was wrong about the number that mattered.

Sources

Bede, The Reckoning of Time, trans. Faith Wallis (Liverpool University Press, 1999), as quoted in Alison Hudson, "'The Earth is, in fact, round'", British Library Medieval Manuscripts blog (24 May 2018).
The Sphere of Sacrobosco, trans. Lynn Thorndike (1949).
Thomas Aquinas, Summa Theologiae I, q.1, a.1, trans. Fathers of the English Dominican Province (1920).
Augustine, The City of God XVI.9, trans. Marcus Dods.
Lactantius, Divine Institutes III.24, trans. William Fletcher.
Cosmas Indicopleustes, The Christian Topography, trans. and intro. J. W. McCrindle (Hakluyt Society, 1897).
The Travels of Sir John Mandeville (Macmillan, 1900).
Washington Irving, A History of the Life and Voyages of Christopher Columbus, vol. I (John Murray, 1828).
A.-J. Letronne, "Des opinions cosmographiques des Pères de l'Église", Revue des Deux Mondes (1834).
John William Draper, History of the Conflict between Religion and Science (1870s).
Andrew Dickson White, A History of the Warfare of Science with Theology in Christendom (1896).
Darin Hayton, "Washington Irving's Columbus and the Flat Earth" (2 December 2014).
Basil of Caesarea, Hexaemeron IX.1, trans. Blomfield Jackson.
Robert Burton, The Anatomy of Melancholy.
Jeffrey Burton Russell, "The Myth of the Flat Earth", American Scientific Affiliation (1997), summarising Inventing the Flat Earth (1991).

Checked September 2026.

Music Visualizers Compared: Free and Paid Tools for Live Shows, Videos and Players

Hacks Vitae — Thu, 17 Sep 2026 00:00:00 GMT

A music visualizer turns sound into moving images. The name covers very different tools: software a VJ runs live at a gig, a website that renders a video for your new single, a feature inside editing software, or the light show built into a music player. This guide sorts them by what they are for, then gives each one's free limits and prices from the vendor's own pages, read on 17 September 2026.

This is not a ranking. The right tool depends on whether you need live visuals or a finished video, your computer, and whether you will be paid for the work.

How to read this guide

Prices are shown as each page displayed them. Resolume served euros; most others served US dollars. We have not converted anything. Prices vary by country, may or may not include tax, and change often.
"Free" rarely means free for everything. Watch for watermarks, resolution caps, time limits and bans on commercial use. Those limits are quoted from the vendors below.

1. Live and VJ software (real-time)

These run on your own computer and react to audio as it plays, for concerts, clubs and streams.

Tool	Platforms	Free version	Paid (as shown)
Resolume Avenue / Arena	macOS, Windows	Demo with no time limit; watermark and "a robot voice" until registered	Avenue €299, Arena €799 per computer
TouchDesigner (Derivative)	Windows, macOS 13+	Non-Commercial: unpaid use only, 1280×1280 cap	Educational $300, Commercial $600, Pro $2,200 USD
Synesthesia	macOS 10.13+, Windows 10+	Unlimited trial, watermarked output	Standard $199, Pro $399
VDMX6 (VIDVOX)	macOS 13+	Demo cannot save projects	$199 USD, Plus $349 USD
NestDrop	Windows only	Classic edition is free	Midnight $50, Midnight Pro $75 USD

Details worth knowing:

Resolume: the demo lets you "try them as long as you like before you decide to buy", but until you register you get "the Resolume logo watermarked on the video output and a robot voice in the audio". "The serial works on macOS and Windows", and "You receive free updates for 12 months when you buy a license." Students and teachers can "Get 50% discount on all software licenses". Current version: 7.27.1, dated 17 July 2026.
TouchDesigner: "TouchDesigner Non-Commercial is free for non-commercial use ONLY", its "image resolution is limited to 1280x1280", and it "does not include a subset of Pro/Commercial Operators". The rule for paid work is plain: "If you or your organization are compensated for your work with TouchDesigner, you must have a Commercial or Pro license." The Educational licence has the same features as Commercial but is "for educational non-paying/non-commercial use only". Latest build listed: 2025.33230, dated 1 September 2026.
Synesthesia: "A Synesthesia license never expires. License purchase includes 1 year of updates." For now, though: "We are currently not enforcing any restrictions on software updates, so all updates are free for all users." A licence "must be synced to the internet once every 30 days to remain active."
VDMX6: the demo is "Fully functional except for the ability to save projects". Its TouchDesigner plug-in support "Requires a paid TouchDesigner license installed on the same computer."
NestDrop is built on MilkDrop (see the next section): "Since the Milkdrop engine is at the core you can easily bring in your own Milkdrop presets."

2. The free MilkDrop family

MilkDrop is the classic swirling visualizer. It began as a Winamp plug-in; MilkDrop 2's source code was released in 2013, and open-source reimplementations such as projectM and Butterchurn carry it on.

MilkDrop. "MilkDrop is a music visualizer - a "plug-in" to the Winamp music player." It "was written by Ryan Geiss". Its page lists system requirements of "Windows 98, ME, 2000, XP, Vista, or 7". Its author's site records: "May 15, 2013: MilkDrop 2 has officially been open-sourced."
projectM "reimplements the esteemed Winamp Milkdrop by Geiss in a more modern, cross-platform reusable library". It is licensed LGPL-2.1, and the latest library release is 4.1.7 (14 July 2026). Two cautions from its own page: "The available end-user frontends are development previews, which are not feature-complete and may have bugs", and "The projectM library does not ship with any presets." The Google Play apps using it "are not created or supported by the projectM developers".
Butterchurn "is a WebGL implementation of the Milkdrop Visualizer" that runs in a web browser. It is MIT-licensed, and "Butterchurn requires the browser support WebGL 2".
Plane9 is a free Windows visualizer and screensaver: "It features over 250 predefined scenes to choose from." The newest download is "v2.5.1.3", "released December 29, 2016", although its undated FAQ says it "Is still actively being developed". If you post videos made with it, its FAQ says "Videos must include a text credit", which "must appear within the video itself and/or be included in the description", and "You may not sell or license posted videos containing Plane9 content, or place the videos behind a paywall." Plane9 also publishes a separate "Commercial use license" for its scenes.

3. Web apps that render a video from your track

Upload audio, pick a template, download a video. These suit a single release or a social clip rather than a live show.

Tool	Free version	Paid (as shown)
Specterr	Watermark, 720p 30fps, 1 video a day, 5-minute videos, videos expire in 10 days	Pro $16.5/month billed yearly ($198), 60 videos a year; Unlimited $49/month billed yearly ($588)
Kapwing	Watermark, 720p, videos up to 4 minutes; its "Audio Waveforms" feature is not ticked for the Free plan	Pro $16 per member per month, billed annually
Rotor	Previews free; downloads need credits or a plan	Credits from $9.99 USD; artist plans $119.99, $179.99 or $239.99 USD a year
Neural Frames (AI)	No free plan shown on the pricing page	Knight $39/month, or $26/month billed yearly
VEED	Adding a sound wave is free	Paid plans (not compared here)
Renderforest	Some templates are premium (the page has a "Premium Only" filter); free-plan limits not compared here	Paid plans (not compared here)

Details worth knowing:

Specterr says: "You own videos that you export and may use them for personal and commercial purposes." It adds: "You must contact us about a custom plan if you intend to resell videos made with Specterr." Its page does not say whether the ownership terms cover watermarked free exports.
Kapwing has an "Audio Waveforms" feature: "Export projects with a dynamic waveform layer visualizing an audio file". Its pricing table ticks it for Pro, Business and Enterprise, not Free. The Free plan offers "Unlimited exports with a watermark", "Export videos up to 4 minutes long" and "720p video quality".
Rotor: "Any video you create on Rotor is 100% yours to own, distribute, and use commercially. Exported videos are completely watermark-free." Each artist plan is "Valid for one artist only". Rotor describes its clip library as "real-life footage, professional animations and striking visualizers"; it does not say the clips react to your audio.
Neural Frames lists "Audioreactive Effects that pulse with your music". Its FAQ says "You own the videos you create with Neural Frames and may publish, monetize, distribute, or use them for client work", and that "Neural Knight includes 1080p upscaling but does not include 4K upscaling."
VEED says it "animates the waveform based on your audio's actual frequencies", and "there is no charge to add sound waves to your videos". Its pricing page lists "No watermark" as a paid-plan feature, so check the export terms before relying on the free version.
Renderforest listed 128 music visualization templates.

4. Do it yourself in editing software

If you already use animation software, you can drive any property from the audio.

Adobe After Effects: "The Convert Audio To Keyframes keyframe assistant analyzes audio amplitude within the work area and creates keyframes for audio amplitude." It reads amplitude only; Adobe's page mentions no frequency bands. The keyframes go into three sliders: "Left Channel, Right Channel, and Both Channels".
Red Giant Sound Keys (Maxon) goes further: it "is a plugin for After Effects that analyzes your audio and generates keyframes you can easily link to your layer's animation properties", using "amplitude or frequency ranges". Maxon sells it as part of Red Giant.
Blender (free): the Graph Editor's Sound to Samples "Creates a sampled curve based on a sound file." Older guides call it Bake Sound to F-Curves, the name used in the Blender 3.6 manual; the 4.0 manual uses the new name.

5. Built into players and streaming apps

Spotify Canvas is not a visualizer: it is a short loop or still image the artist makes in advance and uploads ("An MP4 or JPG file"). Spotify describes it as "an 8-second visual loop that shows in a vertical format", and says "It appears in the Now Playing View instead of your album artwork." The specification: "3-8 seconds long", "Vertical 9:16 ratio", "Between 720px - 1080px tall". Spotify also advises: "Avoid rapid video cuts or intense flashing graphics."
Apple Music on a Mac has a built-in visualizer: "Use the visualizer to watch colorful light shows." To start it: "Choose Window > Visualizer."
Windows Media Player (legacy): Microsoft's archived SDK documentation, filed under "previous versions", says "Several standard visualizations are included with Windows Media Player, including Spikes, Bars, and DotPlane", and that the Windows Media Player SDK "is a legacy feature". It does not describe the newer Media Player app in Windows 11.

For developers: on 27 November 2024 Spotify stopped "new Web API use cases" from using endpoints including "Audio Features" and "Audio Analysis". The change applied to "New apps that are registered on or after today's date" and to "Existing apps that are still in development mode without a pending extension request"; "Applications with existing extended mode Web API access that were relying on these endpoints remain unaffected by this change." Its reference now marks Audio Analysis as deprecated, and its documentation lists a further "February 2026 Dev Mode Changes" migration.

Questions to ask before you choose

Live or rendered? Live shows need real-time software (section 1 or 2). A video for a release needs a renderer (section 3 or 4).
Will you be paid? Some free licences, like TouchDesigner Non-Commercial, rule out paid work.
Is there a watermark, a resolution cap or a length limit on the free version?
What does the licence say about posting or selling the result? Plane9's credit rule is one example.
Does it run on your computer? VDMX6 lists only macOS requirements; NestDrop says it runs "on Microsoft Windows only"; Plane9 lists Windows 7 to 10.

Sources

Each vendor's own product, pricing, licensing, download and help pages, read on 17 September 2026: Resolume (shop, download); Derivative (TouchDesigner Non-Commercial, Licensing, download); Synesthesia (pricing, FAQ); VIDVOX (VDMX home); N.E.S.T. Immersion (NestDrop); Ryan Geiss (geisswerks.com, MilkDrop page); projectM (GitHub repository and releases); Butterchurn (GitHub repository); Plane9 (download, FAQ, home, Commercial use license); Specterr (pricing); Kapwing (pricing); Rotor (pricing, Canvas page); Neural Frames (pricing); VEED (music visualizer tool page, pricing); Renderforest (music visualisations); Adobe (After Effects user guide); Maxon (Sound Keys); Blender manuals (5.2, 4.0 and 3.6); Spotify (Canvas guidelines, Web API blog of 27 November 2024, Web API reference and February 2026 migration guide); Apple (Music User Guide for Mac); Microsoft (Windows Media Player SDK documentation).

Checked September 2026.

Music Myths Checked: 11 Things People Believe About Music

Hacks Vitae — Thu, 17 Sep 2026 00:00:00 GMT

Some of the most repeated facts about music come from a single study, a marketing story or a rumour that took on a life of its own. Below are eleven common beliefs, each checked against the original research, court records, museum collections or the companies involved.

How to read the verdicts: MYTH means the sources contradict the belief. PARTLY TRUE means there is something to it, but the usual version overstates it. DISPUTED means researchers disagree.

1. "Listening to Mozart makes you smarter"

MYTH. The idea goes back to a 1993 letter in Nature, "Music and spatial task performance". A 1999 meta-analysis in the same journal summarised the claim as "listening to ten minutes of Mozart's music increased the abstract reasoning ability of college students, as measured by IQ scores, by 8 or 9 points". It concluded that "any cognitive enhancement is small and does not reflect any change in IQ or reasoning ability in general".

A larger meta-analysis in 2010 went further. According to the University of Vienna's summary, it covered "about 40 independent studies" with "more than 3000 participants", and found "there remains no support for gains in spatial ability specifically due to listening to Mozart music".

2. "Auto-Tune was invented for Cher (or T-Pain)"

MYTH. Auto-Tune existed before either of them used it. Antares, the company that makes it, says: "AutoTune was first released in 1997 by Dr. Andy Hildebrand." The Recording Academy, which gave him a Technical Grammy in 2023, says the same, and notes that he previously pioneered "3-D seismic-mapping technologies in the oil industry".

The famous audible effect came later: the Recording Academy says that "everything changed in 1998" when Cher's "Believe", with its "deliberately unsubtle pitch-processing effect", brought the sound to the public.

3. "Beethoven was completely deaf when he wrote his music"

MYTH. His hearing declined over decades. In June 1801 he wrote to a friend: "My hearing has been getting weaker and weaker for 3 years". The Beethoven-Haus in Bonn notes that in 1812 he met "the mechanic Johann Nepomuk Maelzel", and that "Maelzel constructed various ear trumpets for Beethoven". It adds that "By around 1814/15, Beethoven was as good as deaf in his right ear."

Even in 1815, a 2021 medical review notes, people could still talk to him "by speaking into the left ear". How much he could hear at the very end is argued. The well-known story from the premiere of the Ninth Symphony in May 1824 — that the contralto soloist had to turn him round to see the applause — comes from her own account, told to the music writer George Grove decades later. Grove's book notes that the anecdote "is given in several forms in the books". The usual view is that by his last decade he was profoundly deaf: from 1818 he relied on "conversation books" in which friends wrote their side, and the same review repeats that at the 1824 premiere he "did not even hear the frenetic jubilation". Theodore Albrecht, who translates the conversation books, has argued that he could still hear, "mostly with his left ear", at that premiere. Most accounts do not go that far.

4. "Humans hear from 20 Hz to 20,000 Hz"

PARTLY TRUE. It is a textbook figure: "Humans can detect sounds in a frequency range from about 20 Hz to 20 kHz." The same textbook adds that "the upper limit in average adults is often closer to 15–17 kHz".

A 2021 study of 162 healthy adults found that among those under 30, "52.2% could respond to 20 kHz", while "no responsiveness was recorded to 20 kHz in the 51 ~ 60 group". At the low end, a review notes that "humans can perceive infrasound, if the level is sufficiently high". So the range is a rough guide for young ears, not a fixed limit for everyone.

5. "432 Hz is the natural tuning, and 440 Hz was imposed"

MYTH. Concert pitch varied for centuries. One history of the standard notes pitches of "A=423 Hz in the case of Handel and A=422 Hz in the case of Mozart", and that "in 1859, a French government commission made A=435 Hz law in that country". The international standard was agreed at a 1939 conference: Nature reported in 1939 on "an International Conference on May 11 and 12 last", attended by delegates from "France, Germany, Great Britain, Holland and Italy". A=440 was later reaffirmed by the International Organization for Standardization, whose standard says the note A "shall be 440 Hz".

We found no source showing that 432 Hz is more "natural". On health claims, the evidence is thin: a 2019 pilot study of "33 volunteers" reported that 432 Hz music "can decrease heart rate more than 440 Hz tuned music", and its authors called for "repeating the experiment with a larger sample pool". A few small trials since then have reported mixed results: a 2025 trial with 43 cancer patients found that, on most heart measures, a sound session tuned to 432 Hz gave effects that "were not significantly different" from the same session tuned to 443 Hz, the concert pitch of German orchestras, and another 2025 study of 28 kickboxers found warming up to their preferred music at 440 Hz "more effective" than at 432 Hz. Studies this small do not establish an effect either way.

6. "The loudness war is over because streaming normalises volume"

PARTLY TRUE. Some streaming services now turn loud tracks down. Spotify says: "We adjust tracks to -14 dB LUFS", and "Negative gain is applied to louder masters". That removes much of the reason to master music as loud as possible.

But it is not universal. Spotify notes that "The web player and 3rd-party devices (like TVs or speakers) don’t use loudness normalization." YouTube's volume-balancing feature "is turned off for YouTube Music and official music videos." And the Audio Engineering Society recommends a different level for track normalisation than Spotify does. We found no measurement showing that commercial releases have actually become quieter, so "over" goes further than the evidence.

7. "Paul McCartney died in 1966 and was replaced"

MYTH. The rumour spread through US student newspapers and radio in autumn 1969. The Columbia Journalism Review describes how the autumn 1969 wave took off after a student newspaper article of 17 September 1969 asked "Is Beatle Paul McCartney Dead?"

McCartney answered it himself in LIFE magazine on 7 November 1969. Explaining one of the supposed "clues", a police badge, he said: "I picked up that O.P.D. badge in Canada."

8. "A Stradivarius sounds better than a modern violin"

MYTH (in blind tests). When players and listeners could not see the instruments, the old violins did not come out ahead. In a 2012 study, "the most-preferred violin was new; the least-preferred was by Stradivari". In a 2014 study, "6 of the 10 soloists chose a new instrument" for a hypothetical tour, and "Soloists failed to distinguish new from old at better than chance levels." A 2017 listener study found "The new violins projected better than the Stradivaris".

Two caveats. All three studies come largely from one research team, and the 2014 study drew a published letter in PNAS arguing that a player's memory of a violin's timbre fades within seconds, too fast to compare instruments while swapping them, so such tests would tend to find no difference unless the differences are striking. The authors replied that they were testing whether a player can tell old from new just by playing a violin, without comparing it with others, so "Memory is therefore not the issue". And the old instruments really are physically different: a separate study found that "maples in Stradivari violins have very different chemical properties compared with their modern counterparts". The blind tests show that experts could not reliably hear the difference they expected.

9. "Music lessons raise children's IQ"

DISPUTED. A 2004 experiment that randomly assigned "144" children found that those given music lessons showed "greater increases in full-scale IQ. The effect was relatively small".

A 2020 meta-analysis of 54 studies concluded that the belief "is incorrect": "Once the quality of study design is controlled for, the overall effect of music training programs is null". A 2022 re-analysis of the same data disagreed: "Reanalyzing their data file resulted in a significant effect size (g = .234)". Neither side finds more than a small effect; they disagree on whether it survives once weaker studies are accounted for.

But a 2024 review in the Annual Review of Psychology, co-written by the author of the 2004 study, concluded that "the evidence that music training causes nonmusical benefits is weak or nonexistent". The weight of evidence leans against a real IQ effect, though the reanalysis keeps the question open.

10. "'Happy Birthday' was under copyright until recently"

PARTLY TRUE. For decades music publishers claimed a copyright in the song and charged for its use. In September 2015 a US federal court held that the defendant publishers, as "Summy Co.'s purported successors-in-interest", did not own a valid copyright in the lyrics: "Because Summy Co. never acquired the rights to the Happy Birthday lyrics". The court noted that copyright in the earlier song "Good Morning", which supplied the melody, had "expired in 1949." The case ended in a settlement, with a final judgment entered on 30 June 2016.

So the song was claimed and licensed as a copyright until 2015, but a court found that claim to the lyrics was not valid. This covers the US case only.

11. "Mozart wrote 'Twinkle, Twinkle, Little Star'"

MYTH. Mozart wrote a set of variations on a tune that already existed. The official catalogue of his works lists them as "Twelve variations in C on "Ah, vous dirai-je, Maman" for clavier", and the catalogue's general note on his keyboard variations says that "The typical set of variations by Mozart is based on a preexisting melody". The French tune was in print decades before his variations.

The English words are a separate poem: the Poetry Foundation lists "The Star" among Jane Taylor's poems, noting that "Twinkle, Twinkle Little Star" was "first published as “The Star”".

What these have in common

A single result grows in the retelling. The Mozart effect started as a small finding on one kind of task.
A product's famous use is not its origin. Auto-Tune existed before the song that made it famous.
A gradual change becomes a dramatic one. Beethoven's hearing loss took decades, not a single moment.

Sources

Rauscher, Shaw and Ky, "Music and spatial task performance", Nature (1993); Chabris, "Prelude or requiem for the 'Mozart effect'?", Nature (1999); University of Vienna summary of Pietschnig, Voracek and Formann, "Mozart effect–Shmozart effect: A meta-analysis", Intelligence (2010), via ScienceDaily (10 May 2010).
Antares, "About" and "Auto-Tune Inventor Dr. Andy Hildebrand Wins 2023 Technical GRAMMY®!"; Recording Academy, 2023 Special Merit Awards and "Dr. Andy Hildebrand" feature; US Patent 5,973,252.
Beethoven-Haus Bonn, "Heard with Beethoven's ear"; George Grove, Beethoven and His Nine Symphonies (3rd ed., 1898); Thomas et al., "Aetiology of Ludwig van Beethoven's hearing impairment", European Archives of Oto-Rhino-Laryngology (2021); Boydell & Brewer, "Beethoven could hear after all" (3 February 2020).
Purves et al. (eds.), Neuroscience (2nd ed.), "The Audible Spectrum"; Wang et al., "Extended high-frequency audiometry in healthy adults with different age groups" (2021); Møller and Pedersen, "Hearing at low and infrasonic frequencies" (2004).
Kaye, "International Standard of Concert Pitch", Nature (1939); Cavanagh, "A brief history of the establishment of international standard pitch a=440 hertz"; Calamassi and Pomponi, Explore (2019); ISO 16:1975, Acoustics — Standard tuning frequency; Hohneck et al., BMC Complementary Medicine and Therapies (2025); a 2025 kickboxing warm-up study, PeerJ.
Spotify for Artists, "Loudness normalization"; YouTube Help, "Control video volume on your device"; Audio Engineering Society, AESTD1008 (2021).
Columbia Journalism Review, "Misinformation and the saga of ‘Paul is Dead’" (20 December 2021); LIFE, 7 November 1969.
Fritz et al., PNAS (2012, 2014, 2017); Tai et al., "Chemical distinctions between Stradivari's maple and modern tonewood", PNAS (2017); Tai, "Role of timbre memory in evaluating Stradivari violins", PNAS (2014), and Fritz, Curtin and Poitevineau's reply.
Schellenberg, "Music lessons enhance IQ", Psychological Science (2004); Sala and Gobet, Memory & Cognition (2020); Bigand and Tillmann, "Near and far transfer: Is music special?", Memory & Cognition (2022); Schellenberg and Lima, "Music Training and Nonmusical Abilities", Annual Review of Psychology (2024).
Marya v. Warner/Chappell Music, Inc., C.D. Cal. (2015) and docket (final judgment 30 June 2016).
International Mozarteum Foundation, Köchel catalogue, KV 265; Poetry Foundation, Jane Taylor.

Checked September 2026.

Numbers That Mislead: 10 Statistical Traps and How to Spot Them

Hacks Vitae — Thu, 17 Sep 2026 00:00:00 GMT

A misleading number is rarely a false one. Usually every figure is accurate and the story built on them is not — because the comparison is missing, the base is tiny, the sample is small, or the data were filtered before anyone looked.

These ten traps cover most of it. Every worked example below uses the published figures, except one marked as invented, and we have redone the arithmetic so you can follow it.

1. Ignoring the base rate

Here is a classic problem from the psychologists Amos Tversky and Daniel Kahneman, who trace it to their own work in 1972.

A city has two cab companies. 85% of the cabs are Green and 15% are Blue. A cab is involved in a hit-and-run at night. A witness says it was Blue, and tests show the witness identifies colours correctly 80% of the time. What is the probability the cab really was Blue?

The typical answer is about 80%. Tversky and Kahneman report that "The median and modal answer is typically .80". The correct answer is about 41%.

The arithmetic. Out of every 100 cabs, 15 are Blue and the witness correctly calls 12 of them Blue (15 × 0.8). But 85 are Green and the witness wrongly calls 17 of them Blue (85 × 0.2). So of the 29 cabs the witness would call Blue, only 12 really are. Their own working: "P(B/W) = 12/(12 + 17) = .41."

The lesson. Evidence has to be weighed against how common something was to begin with. A fairly reliable witness is not enough when, in their words, "the base rate is more extreme than the witness is credible."

2. Correlation that means nothing

Two things rising and falling together does not show that one causes the other.

Tyler Vigen built a whole project on this, called Spurious Correlations. His site pairs unrelated data series that happen to match — for example, the distance between Uranus and the Moon and electricity generation in Japan, which from 1980 to 2021 correlate at about 0.985. The match is real in the numbers and meaningless in the world. Vigen is open about how he finds them. "I have 25,237 variables in my database," his site explains. "I compare all these variables against each other to find ones that randomly match up."

Sometimes the link is real but runs through something else. The textbook Forecasting: Principles and Practice gives the standard illustration: ice-cream sales and drownings rise together, but "They are both caused by a third variable (temperature)."

The check. Ask what else could drive both numbers, and whether the pair was found by searching through many possibilities.

3. Relative change without the base

The EU statistics office, Eurostat, uses a simple example: employment in a country rises from 4.8 million to 5.2 million. That is an increase of 0.4 million people — or, in relative terms, about 8.3% (5.2 ÷ 4.8 = 1.083).

Both descriptions are true. Which one sounds impressive depends on the base. A rise from 2 to 4 is "doubled" and "+100%", and it is still only 2. When you see a percentage change, find the starting number.

4. Percent versus percentage points

These two are routinely confused, and the difference can be large.

Eurostat's example: in the EU in 2022, 85.9% of women and 81.4% of men aged 20 to 24 had completed at least upper-secondary education. In its words, "85.9 % minus 81.4 % = 4.5 percentage points."

But relative to the men's figure, the women's rate is about 5.5% higher (85.9 ÷ 81.4 = 1.055). "4.5 points" and "5.5%" describe the same gap. A percentage point is the difference between two percentages; a percent change is a ratio.

5. Survivorship bias

You only see what made it through the filter.

The famous story is Abraham Wald, a statistician at Columbia University's Statistical Research Group during the Second World War. His work was a series of memoranda written in 1943. When the Center for Naval Analyses reprinted them in 1980, it noted that "this work was never published externally".

The logic is sound: holes on the planes that came back show damage a plane could survive. Hits in places that brought planes down are under-represented, because those planes never returned to be counted.

Two details are usually lost in the retelling:

Wald's bombing-mission example was hypothetical. It begins "Of 400 planes on a bombing mission, 359 return" and states that, "for the observed data of this hypothetical example, the engine area is the most vulnerable". It was a method for estimating vulnerability, not a report of real bombers.
The famous picture is modern. Cameron Moll wrote in 2022 that "sometime around 2005 I hastily plotted fictitious red dots on a poorly-chosen commercial aircraft outline", and that the image widely shared online "is a recreation of my diagram".

Our survivorship bias guide covers how the same trap shows up in success stories.

6. Simpson's paradox — when the total says the opposite

In autumn 1973, the University of California, Berkeley admitted a higher share of the men who applied to graduate school than of the women: about 44% of the men and 35% of the women.

It looked like discrimination. When P. J. Bickel, E. A. Hammel and J. W. O'Connell examined it department by department, in a paper published in Science in 1975, the pattern changed. The aggregate data showed "a clear but misleading pattern of bias against female applicants." Few individual departments showed a significant difference, and once the data were pooled in a way that took each department into account, they found "a small but statistically significant bias in favor of women."

Why the total flipped. In the six largest departments, most women applied to the four that admitted the smallest share of applicants, while about half the men applied to the two that admitted the largest share. Department A, for example, admitted 62% of male applicants and 82% of female ones — but far more men than women applied there.

The authors did not conclude that there was no problem. They put the gap down not to "any pattern of discrimination on the part of admissions committees, which seem quite fair on the whole", but to "prior screening at earlier levels of the educational system" — in their account, women were steered towards fields of study that were more crowded.

The check. When a total looks damning or glowing, ask whether the groups being combined are comparable.

7. Regression to the mean

Extreme results tend to be followed by less extreme ones. The standard name for it is regression to the mean.

Daniel Kahneman described in his Nobel biography how he learned this while teaching flight instructors in the Air Force. A seasoned instructor told him that when he praised cadets for a clean aerobatic manoeuvre they generally did worse next time, and when he screamed at them for a bad one they generally did better. Kahneman's explanation was regression to the mean: praise tends to follow an unusually good attempt and scolding an unusually bad one, and the next attempt tends to be more ordinary either way. His summary: "we are statistically punished for rewarding others and rewarded for punishing them."

The check. When something improves after an unusually bad result — or worsens after an unusually good one — ask whether it would have moved anyway.

8. Small samples swing wildly

Small groups produce more extreme averages, in both directions.

Howard Wainer showed this in American Scientist with school test scores. Looking at the fifth-grade reading scores of 1,662 Pennsylvania schools, he found that "of the top-scoring 50 schools (the top 3 percent) six were among the smallest 3 percent of the schools." But the bottom of the list told the same story: "Nine of these (18 percent) were among the 50 smallest schools."

Small schools were over-represented at both ends. In Wainer's words, "smaller schools are expected to have higher variance and hence should be over-represented at both extremes." Tversky and Kahneman had described a related error in 1971: "People have erroneous intuitions about the laws of chance."

The check. Before trusting a "best" or "worst" ranking, look at how big the groups at the top and bottom are.

9. Bar charts that don't start at zero

A bar's length is read as its size. Cut off the bottom of the axis and small differences look large.

The UK Government Analysis Function is direct about it: "We advise that bars on bar charts should always start at zero." Its reason: "When bars do not start at zero, the distance between bar height or length is exaggerated."

An invented example. Two values of 50 and 55 differ by 10%. Start the axis at 45, and the bars show 5 against 10 — one looks twice the other.

10. "Average" can hide almost everything

The mean is pulled by extreme values; the median is not.

The Australian Bureau of Statistics uses the retirement ages of 11 people, from 54 to 60. The mean is about 56.6 and the median 57. Change one of the two 60s to 81, and the mean becomes "(54+54+54+55+56+57+57+58+58+60+81 = 644), divided by 11 = 58.5 years" — while the median stays at 57. As the bureau puts it, "The mean is more sensitive to the existence of outliers than the median or mode."

The check. When you see "the average", ask whether it is the mean or the median, and how spread out the values are.

The one-minute number check

Compared with what? A number with no comparison says very little.
What was the base? Big percentages often sit on small starting numbers.
Points or percent? They are different figures.
How big was the sample? Extreme results cluster in small groups.
What got filtered out before you saw it?
Does the total hide groups pulling in different directions?
Mean or median — and where does the axis start?

Sources

Amos Tversky and Daniel Kahneman, Evidential Impact of Base Rates, Office of Naval Research Technical Report No. 4, 15 May 1981.
Tyler Vigen, Spurious Correlations (website).
Rob J. Hyndman and George Athanasopoulos, Forecasting: Principles and Practice, 3rd edition, section 7.8.
Eurostat, "Statistical concept — Percentage change and percentage points", Statistics Explained.
Abraham Wald, Statistical Research Group memoranda on aircraft vulnerability (1943), reprinted by the Center for Naval Analyses (CRC 432, July 1980).
Cameron Moll, "Abraham Wald and the airplane diagram with red bullet holes – here's the origin story", personal website, 24 March 2022.
P. J. Bickel, E. A. Hammel and J. W. O'Connell, "Sex Bias in Graduate Admissions: Data from Berkeley", Science 187 (1975).
Daniel Kahneman, Biographical, The Nobel Prize (2002).
Howard Wainer, "The Most Dangerous Equation", American Scientist 95 (2007).
Amos Tversky and Daniel Kahneman, "Belief in the Law of Small Numbers", Psychological Bulletin 76 (1971).
UK Government Analysis Function, "Introduction to data visualisation", module 5: bar charts.
Australian Bureau of Statistics, "Measures of central tendency".

Checked September 2026.

Open-Weight Is Not Open-Source: What "Open" AI Models Actually Let You Do

Hacks Vitae — Thu, 17 Sep 2026 00:00:00 GMT

When an AI company says it has released a model "openly", it often means you can download the trained model and run it yourself. That is useful. It is also not the same thing as open-source — and the difference matters if you want to build on the model, study how it was made, or use it in a business.

This page explains the difference and shows what the licences on several downloadable models say, as read in September 2026. It is an explainer, not legal advice; if you plan to use a model commercially, read its licence in full.

What "weights" are

A trained AI model is, at its core, a very large set of numbers learned during training. The Open Source Initiative (OSI), which published the definition of open-source AI discussed below, explains: "Open Weights refer to the final weights and biases of a trained neural network."

Having those numbers lets you run the model. It does not tell you how the model was built. As the OSI puts it: "However, Open Weights differ significantly from Open Source AI because they do not include:" the training code, the training dataset and information about the data.

That still gives you real options. In a September 2026 post, the OSI wrote that with open weights "you can choose to run a model locally" and "You can also fine-tune the model with your own data". It also set out the limit: "Open-weights allow you to exercise some freedoms, but not all, and only to a certain degree."

What "open-source AI" requires under the OSI definition

The OSI published version 1.0 of its Open Source AI Definition in October 2024; its announcement is datelined "RALEIGH, N.C., Oct. 28, 2024". It starts from four freedoms. An open-source AI system must let you:

"Use the system for any purpose and without having to ask for permission."
"Study how the system works and inspect its components."
"Modify the system for any purpose, including to change its output."
"Share the system for others to use with or without modifications, for any purpose."

To make those freedoms real, the definition lists three things that must be available:

Data information — "Sufficiently detailed information about the data used to train the system so that a skilled person can build a substantially equivalent system." It does not require every piece of training data to be published. It requires a complete description of the data, "including (if used) of unshareable data", plus "a listing of all publicly available training data and where to obtain it" and "a listing of all training data obtainable from third parties and where to obtain it, including for fee".
Code — "The complete source code used to train and run the system."
Parameters — "The model parameters, such as weights or other configuration settings." And: "Parameters shall be made available under OSI-approved terms."

So weights are one of three required parts. Under this definition, a model whose weights are released under a standard licence is still not open-source AI if the training code and data information are missing.

What the OSI has said about Meta's Llama models

The OSI has been explicit about Meta's Llama models. In a February 2025 post it wrote: "Llama 3.x is still not Open Source by any stretch of the imagination." Among its reasons, it said the licence "Fails at the Open Source Definition point 5, discriminates against users".

It said the same about Llama 4 in a LinkedIn post in April 2025: "Llama 4 is still not #opensource". We have not found an OSI statement on the other licences in the table below, so we do not attribute a verdict to the OSI on them.

What the licences say (September 2026)

Each entry below was read from the model's official page or its licence file. The summaries are simplified; the exact conditions are in each licence.

Model	Licence	What stands out
DeepSeek-V4-Pro	MIT	The model page says: "This repository and the model weights are licensed under the MIT License".
OpenAI gpt-oss-120b / 20b	Apache 2.0	The model page describes a "Permissive Apache 2.0 license". Its usage policy says: "By using OpenAI gpt-oss-120b and gpt-oss-20b, you agree to comply with all applicable law."
Google Gemma 4	Apache 2.0	Google's Gemma terms page says "For Gemma 4 terms, see the Gemma 4 license". Its own Gemma Terms of Use apply to earlier Gemma models, and say you must not use them "for the restricted uses set forth in the Gemma Prohibited Use Policy".
Meta Llama 4	Llama 4 Community License Agreement	Measured on the Llama 4 release date: if the products or services of a licensee or its affiliates had more than 700 million monthly active users, "you must request a license from Meta", and you have no rights under the agreement "unless or until Meta otherwise expressly grants you such rights". Its Acceptable Use Policy also restricts uses including "Military, warfare, nuclear industries or applications, espionage".
Qwen3.8-2.4T-A95B (Alibaba)	Qwen3.8-Max License	If the licensee or its affiliates run a "Model as a Service or AI Work Assistant business" and their combined revenue exceeds US$50,000,000 in any 12 consecutive months, "the licensee shall obtain a separate license from Qwen before Using the Software or its derivative works for any commercial purpose". An AI Work Assistant is defined as "an independent AI-powered product primarily designed for AI-assisted coding or office productivity", excluding single-purpose tools and assistants built mainly for other domains or as a feature of another kind of product. Internal use is exempt only where it does not make "the Software, its outputs, or its underlying model capabilities available to any third party".
Mistral Medium 3.5	Modified MIT License	"You are not authorized to exercise any rights under this license if the global consolidated monthly revenue of your company (or that of your employer) exceeds $20 million (or its equivalent in another currency) for the preceding month."
Kimi K3 (Moonshot AI)	Kimi K3 License	If the licensee or its affiliates operate a "Model as a Service business" and their combined revenue exceeds 20 million US dollars over any 12 consecutive months, "the Licensee must enter into a separate agreement with Moonshot AI before using the Software or its derivative works for any commercial purpose". Exempt: internal use, defined as use that does not make the Software, "its outputs, or its underlying capabilities available to third parties", and use through "Moonshot AI's official products or certified inference partners".
GLM-5.3 (Z.ai)	GLM-5.3 License	If the licensee or its affiliates operate a "Model as a Service business" and their combined revenue exceeds 10 billion US dollars over any 12 consecutive months, the licensee "must pass Z.AI's security review before using the Software or its derivative works for any commercial purpose".

A note on scope for Llama 4: its Acceptable Use Policy also contains an exclusion for "an individual domiciled in, or a company with a principal place of business in, the European Union", but it applies to the multimodal Llama 4 models, and the policy exempts end users of products built on them. It is not a blanket ban.

Licences can differ within one company

The same company can publish different models under very different terms. On Hugging Face, Mistral Large 3 and Mistral Small 4 are tagged "apache-2.0", while Mistral Medium 3.5 uses the Modified MIT licence above. Alibaba's smaller Qwen3.8-27B is tagged "apache-2.0", and Z.ai's GLM-5.3-Flash is tagged "mit", although the larger models in the table use custom licences. Google publishes Gemma 4 under Apache 2.0, while its earlier Gemma models remain under its own terms.

That is why the only reliable check is the licence file for the exact model you are using.

How to read an "open" AI model in two minutes

Find the LICENSE file in the model's official repository, not just the label on the page.
Is it a standard licence such as MIT or Apache 2.0, or a custom one with the model's name in it?
Look for thresholds — revenue, monthly users, or types of business that need separate permission.
Look for use restrictions — a policy listing things you may not do with the model.
Check for training code and data information if "open-source" is being claimed. Weights alone are not enough under the OSI definition.

Our reading: "open" in AI marketing usually means "downloadable". That is valuable — you can run the model privately, inspect its behaviour and adapt it. But when the word "open-source" is used, it is reasonable to ask for the licence, the training code and the data information, and to judge the claim by what is published.

Sources

Open Source Initiative, "The Open Source AI Definition – 1.0"; "The Open Source Initiative Announces the Release of the Industry's First Open Source AI Definition" (28 October 2024); "Open Weights"; "Meta's LLaMa license is still not Open Source" (February 2025); "Open Weights Are Good. Open Source Is Better." (16 September 2026).
DeepSeek, DeepSeek-V4-Pro-0813 model page, Hugging Face.
OpenAI, gpt-oss-120b model page, Hugging Face; gpt-oss LICENSE and USAGE_POLICY, GitHub.
Google, Gemma Terms of Use and Gemma 4 licence, ai.google.dev.
Meta, Llama 4 LICENSE and USE_POLICY, meta-llama/llama-models, GitHub.
Alibaba Qwen, Qwen3.8-2.4T-A95B LICENSE and Qwen3.8-27B model page, Hugging Face.
Mistral AI, Mistral-Medium-3.5-128B LICENSE; Mistral-Large-3 and Mistral-Small-4 model pages, Hugging Face.
Moonshot AI, Kimi-K3 LICENSE, Hugging Face.
Z.ai, GLM-5.3 LICENSE and GLM-5.3-Flash model page, Hugging Face.

Checked September 2026.

The Battle of Mactan, 1521: What Pigafetta Saw and What the Other Accounts Say

Hacks Vitae — Thu, 17 Sep 2026 00:00:00 GMT

On 27 April 1521, on the island of Mactan, the forces of the chief Lapulapu defeated a landing party under Ferdinand Magellan, and Magellan was killed. The Philippines now marks the day by law. But the familiar story comes mostly from a single source, written by a man on the losing side, and the other early accounts do not agree with him.

Where the story comes from

The main account is Antonio Pigafetta's, an Italian who sailed with Magellan and was in the fight. The English edition by Emma Helen Blair and James Alexander Robertson (1906) notes that his manuscript was "composed ca. 1525, of events of 1519–1522" — about four years after the battle. Four manuscripts of his account survive, and they differ: "The MSS. of the Bibliothèque Nationale are both shorter than the Italian MS."

Pigafetta was no neutral observer. Straight after describing Magellan's death, he writes his praise: "Had it not been for that unfortunate captain, not a single one of us would have been saved in the boats".

What Pigafetta says happened

A demand for submission. According to Pigafetta, on "Friday, April twenty-six, Zula, a chief of the island of Matan," sent word that he could not send all he had promised "because of the other chief Cilapulapu, who refused to obey the king of Spagnia," and asked for a boatload of men to fight him. Magellan decided to go himself, and before the fighting sent word that the islanders must "obey the king of Spagnia, recognize the Christian king as their sovereign, and pay us our tribute". The "Christian king" was the ruler of Cebu, recently baptised, whom Magellan had named Don Carlo.

The numbers. "At midnight, sixty men of us set out armed with corselets and helmets," Pigafetta writes, with "twenty or thirty balanguais" (boats) of Cebuano allies. "When morning came forty-nine of us leaped into the water up to our thighs," while "The other eleven men remained behind to guard the boats." The defenders "had formed in three divisions to the number of more than one thousand five hundred persons."

The death. Pigafetta names none of the attackers. "They shot the captain through the right leg with a poisoned arrow," then "One of them wounded him on the left leg with a large cutlass," which "caused the captain to fall face downward, when immediately they rushed upon him with iron and bamboo spears and with their cutlasses".

The date and the dead. "That battle was fought on Saturday, April twenty-seven, 1521." Pigafetta counts "Eight of our men" killed with Magellan, and notes that "four Indians, who had become Christians and who had come afterward to aid us were killed by the mortars of the boats" — that is, by the expedition's own guns. "Of the enemy, only fifteen were killed". Afterwards the people of Mactan refused to hand over Magellan's body: "they intended to keep him as a memorial."

How the other early accounts disagree

PARTLY TRUE. The familiar version holds in outline — a landing, a defeat, Magellan killed — but not in its details.

	Pigafetta	Genoese pilot	Maximilianus Transylvanus	Account from Ginés de Mafra's notes
Date	27 April 1521	28 April	—	—
Magellan's force	60 set out, 49 landed	50–60	40	40, of whom 34 landed
Defenders	more than 1,500	3,000–4,000	about 3,000	up to 50 from one house, then a second, uncounted ambush
Magellan's men killed	8	6	7	12, including Magellan

The Genoese pilot, whose account is quoted in the Blair and Robertson notes, "dates the battle April 28", puts the defenders at "as many as three thousand or four thousand men", and says Magellan "was killed there, with six of his men".
Maximilianus Transylvanus was not there. He wrote from interviews with survivors: "I have taken care to have everything related to me most exactly by the captain and by the individual sailors who have returned with him." He gives "forty of his men", "about three thousand" defenders, and "Magellan, and seven of their comrades" lost.
An account based on the notes of Ginés de Mafra, a sailor on the voyage, was written up by someone else and published in 1920. It has 40 men, of whom Magellan "salió … en tierra con treinta y cuatro" (landed with thirty-four). Up to 50 men burst out of a house where they had hidden, and a second group then attacked from behind ("a manera de celada", like an ambush), leaving "doce de los nuestros" (twelve of ours) dead, Magellan among them. In this version he "se desangró tanto que cayó muerto" — bled so badly that he fell dead (our translations). The same text calls him "Sebastian de Magallanes", a clear error.
Even Pigafetta's own manuscripts differ. A shorter French manuscript says "one of those Indians hurled a poisoned bamboo lance into his face which laid him stiff and dead" (Robertson's translation).
Numbers grew in retelling. Richard Eden's early English account, the editors note, gave each of the three divisions "two thousand and fiftie men" — 6,150 in all.

Navarrete's documents, as cited by Blair and Robertson, name seven men killed with Magellan, one of whom "died April 29". The fairest summary is that Magellan died with between six and eleven of his men, depending on the count.

Did Lapulapu kill Magellan?

UNVERIFIED. No early account says so. Pigafetta names "Cilapulapu" only as one of Mactan's two chiefs and as the one who, by his rival's account, "refused to obey the king of Spagnia", and describes the killing by unnamed attackers. On the battle itself, Transylvanus says only that "The King of Mauthan" drew up his men.

Lapulapu's place in memory was built much later. The National Quincentennial Committee, working with the National Historical Commission of the Philippines, traces how writers of the reform and independence movements celebrated him. In 1895 Emilio Jacinto wrote of the blood of Kalipulako, the king of Mactan, "niyaong pinatay niya ang lilong si Magallanes", which the Committee translates as "after he killed the liar Magellan". The 1898 proclamation of independence is more careful. In English translation, Magellan "was killed in battle that took place in said shores to which battle he was provoked by Chief Kalipulako of Mactan".

What is well supported: Magellan was killed by Mactan's fighters. What is less certain is what exactly Mactan refused. Pigafetta, reporting what Lapulapu's rival said, has him refusing to obey the king of Spain; Transylvanus has the king of Mactan refusing "to do homage to one whom he had been accustomed for so long to command" — the ruler of Cebu.

The name

The name appears in several forms. Pigafetta's text as printed by Blair and Robertson has "Cilapulapu". The Committee explains that "the “Çi” in the recorded name of the Mactan leader is most likely the ancient honorific title ‘Si.’" Spanish chroniclers wrote other versions, and the Committee notes that "Mariano Ponce, used Kalipulako as his penname".

The Committee recommended in 2019 writing the chief's name "as Lapulapu (without hyphen)", while the 2018 law and the city keep the hyphen: Lapu-Lapu City.

UNVERIFIED. A popular story says the fish called lapu-lapu was named after the chief. We found no historical source for which name came first.

How Mactan is remembered

Republic Act No. 11040, approved on 29 June 2018, declares 27 April "a special working public holiday throughout the country", "to commemorate the victory of Lapu-Lapu and his men over the Spaniards in the historic Battle of Mactan on April 27, 1521". No image of Lapulapu from his own time is known to us. The profile on the Philippines' 5,000-piso commemorative banknote is, in the Committee's words, "based on the portrait commissioned by the National Historical Commission of the Philippines from artist Carlo Caacbay".

Our reading

Mactan is remembered as a clear victory, and it was one. But the details most often repeated — the numbers and the manner of Magellan's death — rest mostly on one partisan eyewitness writing years later, and the other accounts do not match. Lapulapu's own hand in the killing rests on no early account at all. No written account from Mactan's side is known. That gap is part of the history too.

Sources

Antonio Pigafetta, Primo viaggio intorno al mondo (Ambrosiana manuscript, c. 1525), trans. James Alexander Robertson, in Emma Helen Blair and James Alexander Robertson (eds.), The Philippine Islands, 1493–1898, vol. 33 (Arthur H. Clark, 1906), including the editors' notes on the Genoese pilot, the French manuscript, early English translations and Navarrete.
Maximilianus Transylvanus, letter, trans. James Baynes, in Henry E. J. Stanley (ed.), The First Voyage Round the World, by Magellan (Hakluyt Society, 1874).
Libro que trata del descubrimiento y principio del estrecho que se llama de Magallanes por Ginés de Mafra, ed. Antonio Blázquez y Delgado-Aguilera (Real Sociedad Geográfica, 1920).
National Quincentennial Committee (with the NHCP), "Lapulapu versus Lapu-Lapu" (2021) and "Lapulapu in the Eyes of our Heroes" (2019), archived copies.
Republic Act No. 11040 (2018), via The Lawphil Project.
Philippine Declaration of Independence (12 June 1898), English translation, Wikisource.
"History body has last word on Lapulapu", Philippine Daily Inquirer (2 May 2021), archived copy.

Checked September 2026.

The First Recorded Sound Was Not Edison's: How Sound Recording Really Began

Hacks Vitae — Thu, 17 Sep 2026 00:00:00 GMT

Ask who first recorded sound and the usual answer is Thomas Edison. His own national historical park says otherwise: "That honor goes to Edouard-Léon Scott de Martinville, a French inventor who in 1857 devised his phonautograph". The difference is playback. Scott could record sound, but not hear it again; Edison built the first machine that did both. Here is how recording began, from the patents, the archives and the people who recovered the earliest recordings.

A short timeline

Year	What happened
1857	Scott patents the phonautograph, which traces sound on paper
1860	Scott records "Au clair de la lune"
1877	Charles Cros deposits a sealed note describing playback (April); Edison builds his tinfoil phonograph (December)
1887	Emile Berliner patents the "Gramophone"
1925	Electrical recording goes into regular use at major US record companies
1935	The Magnetophon tape recorder is shown in Berlin
1971	The first commercial digital recording is released in Japan
2008	Scott's 1860 recording is played back for the first time

1. The phonautograph: recording without playback (1857)

Scott's patent papers were deposited in Paris on 24 March 1857. In his own later note, he refers to "mon brevet d’invention délivré le 25 mars 1857 sous le nº 31470" — his patent issued on 25 March 1857, number 31,470.

His machine turned sound into a visible trace. It was meant to be looked at, not listened to. First Sounds, the group that recovered his recordings, is plain about this: "Scott did not intend his phonautograms to be played back". The Edison park agrees, calling it "a record-only device".

It still had a long life. First Sounds notes: "As a laboratory instrument it contributed for decades to the nascent science of acoustics."

2. "Au clair de la lune", 1860 — and the playback that got the speed wrong

"Scott recorded the French folksong "Au Clair de la Lune" on April 9, 1860". According to First Sounds, "It remains the earliest clearly recognizable record of the human voice yet recovered." Older phonautograms survive too, some from as early as 1853–54. First Sounds adds: "Many phonautograms from 1857 also survive, but they lack the tuning-fork timecode" needed to play them at the right speed.

In 2008, First Sounds released sound files made from Scott's paper traces using "the Lawrence Berkeley National Laboratory's virtual stylus technology". The first release sounded like a woman or a child singing. The speed turned out to be wrong. The team later explained: "we chose a playback speed that presumed the singer was a woman or an adolescent", then "we realized that this speed was too fast".

The clue was a recording of speech. Scott had written that his tuning fork made "500 simple vibrations per second". In early 2009 the team played a spoken recording at the speed used for the 2008 release, and it "sounded very much like a tape played back at twice the proper speed". Instead, "adjusting the timecode to 250 Hz gave us a natural-sounding low male voice". Patrick Feaster of First Sounds explains that, "despite Scott’s own statements to the contrary", the phrase "would clearly have meant 250 Hz in the terminology of mid-nineteenth century acoustics". Comparing it with another of Scott's recordings, the researchers concluded that the singer sounds "very much like the speaker" in a recording Scott made of himself. So it is probably Scott singing, though that is a voice-match judgement rather than proof.

Even the words were first misheard: the recording was later found to be "Au clair de la lune, mon ami Pierrot", "rather than “Au clair de la lune, Pierrot répondit,” as originally announced".

3. The playback idea, on paper first: Charles Cros (1877)

The idea of playing back a recording was written down in France before Edison built his machine. The French inventor Charles Cros sent a sealed note to the French Academy of Sciences, "deposited at the French Academy of Sciences on April 30, 1877". It was opened and read on 3 December 1877. In it he described recording the movement of a vibrating membrane and using that trace "pour reproduire le même va-et-vient" — to reproduce the same movement.

But the note was sealed, so Edison could not have seen it: "Cros’ proposal was in a sealed envelope held in confidence by the French Academy of Sciences." And "Cros never built a paléophone, nor was a paléophone ever built to his specifications." The name paléophone came later: "for most of 1877 Cros had no name for his idea."

4. Edison's phonograph (1877)

Edison's idea came out of telephone work. According to the Thomas A. Edison Papers at Rutgers, "In July 1877, while developing his telephone transmitter, Edison conceived the idea of recording and playing back telephone messages." His machinist John Kruesi built the tinfoil phonograph "during the first week of December 1877".

Edison's national park adds a detail the usual story leaves out: "this iconic machine was not the first phonograph – but it was the first to be shown outside the lab". An earlier paper-strip device from July 1877 "remained just a prototype in the lab". On 7 December 1877 Edison and two colleagues took the tinfoil machine to Scientific American, whose editors reported that "the machine inquired as to our health, asked how we liked the phonograph".

His patent application was "filed December 24, 1877", and the patent, No. 200,521, is "dated February 19, 1878". Its purpose: "to record in permanent characters the human voice and other sounds".

What about "Mary had a little lamb"? The story is real but more tangled than "the first words ever recorded". Rutgers records Edison's recollection that he "recorded the nursery rhyme "Mary Had a Little Lamb"" on the tinfoil machine. His colleague Charles Batchelor, writing decades later, remembered Edison reciting it, "one of our favorite stereotyped sentences used in experimenting on the telephone", into the earlier paper-strip prototype, which played back "ary ad elll am". And according to the national park, Kruesi's machine went first to Batchelor, "who tested and adjusted the finicky mechanism before showing it to Edison". Both accounts are later recollections.

Did Edison know about Scott? The park says: "Primary sources are silent on whether Edison even knew of the phonautograph in 1877."

5. Discs: proposed in 1877, made to win by Berliner

Emile Berliner's gramophone is remembered as the start of disc records. The history is not that simple. Edison's park notes that "In 1877 and 1878 plate (later called “disc”) designs appeared in laboratory sketches and patent applications nearly as much as cylinder designs", including a British patent of April 1878.

Berliner's first gramophone patent, "GRAMOPHONE. No. 372,786. Patented Nov. 8, 1887.", describes wrapping the record "tightly about the drum", and says it works "all substantially in the manner of the well-known phonautograph by Leon Scott". A later patent, dated 15 May 1888, describes a surface "which may be a flat disk or a cylinder". The disc format won in the end: its "disc records became synonymous with sound playback" for most of the 20th century.

6. Electrical recording (1925)

Before this, records were made by the acoustic method. Western Electric's electrical process changed that. It was offered to Victor and Columbia "in the fall of 1924", and "Columbia promptly signed with Western Electric, followed a few weeks later by Victor." "Electrical recording at Columbia began on a regular basis in April" 1925 — though "neither Columbia nor Victor billed their new releases as electrically recorded until 1926".

In a 1926 paper, J. P. Maxfield and H. C. Harrison wrote that with the new system "a range of frequencies from 60 cycles to 6000 can be recorded with reasonable success".

7. Magnetic recording and tape

Magnetic recording is older than tape. Valdemar Poulsen's US patent for magnetic recording was filed in 1899 and published in 1900. Practical coated tape came from Germany. (An American, Joseph O'Neill, applied in 1926 for a US patent on a paper strip carrying "a trail or line of magnetic material", granted in 1927, but "did not pursue it commercially".) Fritz Pfleumer patented a "sound paper machine", filed in 1928, and licensed it to AEG. BASF says "The first 50,000 meters of magnetic audio tape are supplied to the electronics corporation AEG in 1934", and according to the Computer History Museum, "The Magnetophon K1 recorder and Type C tape debuted at the Berlin Radio Show in August 1935."

8. The first digital recordings: Japan first

Commercial digital recording did not start in the US. Denon (Nippon Columbia) worked with a recorder from the Japanese broadcaster NHK: "In 1969-1971, Denon leased an NHK stereo PCM recorder and conducted numerous test recordings." According to a history published in the ARSC Journal, "the first commercial digital recording was Nippon Columbia NCB-7003, "Something" by Steve Marcus, released January 1971".

In the US, a 1976 digital test recording at the Santa Fe Opera was not released from the digital tape: the album "was made from a mix-down of Bruck's 16-track analog tape." Telarc's April 1978 recording with the Cleveland Symphonic Winds became "the first classical intended-for-release commercial digital recording in the U.S."

Our reading

"Who invented sound recording?" has at least three fair answers: Scott first recorded sound, Cros first described playing it back, and Edison first built a machine that did both. Most "firsts" in this story survive only with a qualifier attached — first recognisable voice, first commercial digital album, first US classical one. Those qualifiers are where much of the history is.

Sources

Thomas Edison National Historical Park (US National Park Service): "Origins of Sound Recording" — the inventors; Scott de Martinville; Charles Cros; Edison's path to the phonograph; Thomas Edison; frequently asked questions.
First Sounds: Scott's phonautograms; "Earlier playback" (speed reinterpretation); Patrick Feaster, "Édouard-Léon Scott de Martinville: An Annotated Discography", ARSC Journal 41:1 (2010).
Thomas A. Edison Papers, Rutgers University, "Tinfoil Phonograph".
US Patent 200,521 (Edison, 1878); US Patents 372,786 (1887) and 382,790 (1888) (Berliner); US Patent 661,619 (Poulsen, 1900).
Comptes rendus hebdomadaires des séances de l'Académie des sciences, t. 85, séance du 3 décembre 1877 (Cros).
Discography of American Historical Recordings, UC Santa Barbara, "Columbia Corporate History: Electrical Recording and the Late 1920s".
J. P. Maxfield and H. C. Harrison, "Methods of High Quality Recording and Reproducing of Music and Speech Based on Telephone Research", Bell System Technical Journal 5:3 (1926).
Computer History Museum, "1935: Audio recorder uses low-cost magnetic tape"; BASF, corporate history, 1934.
Thomas Fine, "The Dawn of Commercial Digital Recording", ARSC Journal 39:1 (2008).

Checked September 2026.

The Iron Maiden Was Never a Medieval Torture Device: What the Museums Say

Hacks Vitae — Thu, 17 Sep 2026 00:00:00 GMT

The iron maiden — a spiked cabinet that closed on its victim — is one of the most famous "medieval torture devices". The museums that own iron maidens now say it was nothing of the kind. Here is what they say, where the story came from, and the real punishment it grew out of.

The museums' verdict

MYTH. The Medieval Crime Museum in Rothenburg ob der Tauber holds one of the best-known iron maidens (the historian Klaus Graf called it the most famous). The museum describes it like this (our translations from the German):

"Der metallbeschlagene Holzmantel der Eisernen Jungfrau stammt aus dem 15./16. Jahrhundert." — The iron maiden's metal-clad wooden cloak dates from the 15th/16th century.
"Die metallenen Dornen hingegen sind auf die Zeit der Befreiungskriege (1813-1815) zu datieren." — The metal spikes, by contrast, date to the Wars of Liberation (1813–1815). The museum says they are French socket bayonets, added later.
"Die Eiserne Jungfrau diente als Schandmantel für Frauen." — It served as a cloak of shame for women, who were released after a while "körperlich unversehrt": physically unharmed.

So the object itself may be old; what is invented is the spiked killing machine.

A second museum is blunter. Schloss Kyburg in Switzerland, where an iron maiden arrived in 1876, calls it "längst als Fälschung durchschaut" — long since exposed as a fake — and in 2026 is running an exhibition on how it went "von der Legende zum «realen» Objekt": from legend to a "real" object.

The historian Klaus Graf, summarising the specialist study by Wolfgang Schild, put it simply in a 2001 lecture: the execution tool "iron maiden" is "eine fiktion des 19. jahrhunderts" — a fiction of the 19th century. He explains that only in the first half of the 19th century were early-modern shame cloaks — which served as instruments of punishment and torture, and were sometimes called "maidens" — fitted inside with iron spikes, adapting the objects to the gruesome fantasies of literature and legend.

Where the story came from

In 1793 the Altdorf law professor Johann Christian Siebenkees printed a short note in his Materialien zur Nürnbergischen Geschichte, headed only "Aus einer Chronik" (from a chronicle). It says that in 1533 an iron maiden was set up in Nuremberg at the wall of the Fröschturm: an iron figure seven feet high ("7 Schuh hoch") that spread its arms toward the condemned and, when the executioner touched a step, hacked him into small pieces with broad sabres, which fish in hidden waters then swallowed. Graf calls it an alleged ("angebliche") chronicle note, and Siebenkees did not name the chronicle. The volume's title page is dated 1792, which may be why that year is sometimes given; the note appears in its twelfth part, which reports an event of 24 April 1793. Later accounts add a coin forger executed with the device on 14 August 1515; that detail is not in the passage Siebenkees printed.

TRUE. Siebenkees doubted the story in the same breath. After remarking that such secret courts had once existed in several towns ("Solche heimliche Gerichte waren ehehin in mehrern Städten"), he wrote: "Ob von dem hier beschriebenen noch eine Spur vorhanden ist, weiß ich nicht, habe auch nie gelesen, daß davon ein Gebrauch gemacht worden. Sollte vielleicht die ganze Sache eine Legende seyn?" — Whether any trace of what is described here still exists I do not know, and I have never read that any use was made of it. Might the whole thing perhaps be a legend? (our translation)

How a legend became an exhibit

In the 19th century, spiked "iron maidens" went on show: Schloss Kyburg's arrived there in 1876. The San Diego Museum of Man says its own copy "was made in 1828 as decoration for the “Gothic Hall” of a patrician palace in Milan". The legend also reached fiction: Bram Stoker's short story "The Squaw" is set in Nuremberg's torture tower, where "the engine known as the Iron Virgin" stands "near the centre of the room".

What happened to the Nuremberg iron maiden is confused. One museum account says the original was destroyed in the Allied bombing of Nuremberg in 1944; other accounts give 1945, the year Nuremberg's castle torture museum closed; and Graf wrote that the most famous one, once in Nuremberg, is now shown in Rothenburg. More than one object has carried the Nuremberg name (the San Diego museum describes its own as a copy made in 1828 for a palace in Milan), so these accounts may describe different objects. We could not settle which.

The real punishment behind it

TRUE. Cloaks of shame existed, and museums hold them.

Schwäbisch Gmünd, about 1680 (Landesmuseum Württemberg): a painted wooden and iron Schandmantel. The museum explains that the convicted person "sollte also weniger körperlich gezüchtigt, sondern öffentlich auf dem Marktplatz bloßgestellt werden" — the aim was less to punish physically than to expose them publicly in the market square.
Groß Salze (today Bad Salzelmen), 1742 (Salzlandmuseum): a cloak made of barrel staves, used, the museum says, "für Vergehen wie z. B. Beleidigungen oder kleinere Diebstähle" — for offences such as insults or petty theft. The convicted person stood in it for some hours in front of the town hall or was led through the streets. The museum says only 14 other examples survive in other German museums.

These were humiliating and uncomfortable, and meant to be. But they were public shaming, not execution — and the dated examples come from the late medieval and early modern periods, not only the Middle Ages.

Our reading

The iron maiden shows how a myth can make its own evidence. A story printed in the 1790s needed an object; the 19th century built one by adding blades to a real punishment; and the object then seemed to prove the story. The museums that hold these objects are the ones now telling visitors so.

Sources

Johann Christian Siebenkees (ed.), Materialien zur Nürnbergischen Geschichte, vol. 2 (Nuremberg, 1792; twelfth part), "IX. Miscellaneen", no. 1, p. 753; Bayerische Staatsbibliothek copy, Internet Archive.
Mittelalterliches Kriminalmuseum Rothenburg ob der Tauber, "Eiserne Jungfrau – Wahrheit oder Fiktion?" (4 November 2019).
Museum Schloss Kyburg, exhibition "Eiserne Jungfrau «Fantasiemaschine»" (2026).
Klaus Graf, "Mordgeschichten und Hexenerinnerungen – das boshafte Gedächtnis auf dem Dorf", lecture, Freilichtmuseum Neuhausen ob Eck (21 June 2001), archived copy; summarising Wolfgang Schild, Die Eiserne Jungfrau. Dichtung und Wahrheit.
San Diego Museum of Man, "Medieval Imposter: the Iron Maiden" (archived 2015).
Landesmuseum Württemberg, Schandmantel, inv. WLM 5383, via museum-digital.
Salzlandmuseum, Schandmantel, inv. V F4/10, via museum-digital.
Bram Stoker, "The Squaw", in Dracula's Guest and Other Weird Stories (1914).

Checked September 2026.

The Library of Alexandria Was Not Destroyed in One Fire: What the Ancient Sources Say

Hacks Vitae — Thu, 17 Sep 2026 00:00:00 GMT

The Library of Alexandria is often said to have died in one great fire, set by one villain. Depending on who tells the story, the villain is Julius Caesar, a Roman emperor, a Christian mob or a Muslim caliph. The ancient sources do not support any single version. Here is what each one says.

One caution first. The historian Roger Bagnall, in a study of the evidence, writes of "our nearly total ignorance about it", and notes: "There is no ancient account of the foundation of the Library." Everything below should be read in that light.

48 BC: Caesar's fire

PARTLY TRUE. In 48 BC, during the fighting in Alexandria, Julius Caesar set fire to ships in the harbour. Books did burn — but the writers who describe it disagree about how many, where, and even who started it.

Caesar himself (Civil War III.111) writes that he burned the ships: "omnesque eas naves et reliquas, quae erant in navalibus, incendit" — he set fire to all those ships and the others in the dockyards (our translation). He says nothing about books. An author describing his own war has reasons to leave things out, so his silence is a hint, not proof.
Seneca (about a century later): "Forty thousand books were burned at Alexandria" (On Peace of Mind 9.5, trans. Aubrey Stewart). He does not say the whole library burned.
Plutarch (around AD 100): the fire "spread from the dockyards and destroyed the great library" (Life of Caesar 49.6, trans. Bernadotte Perrin). His translator added a note: "The destruction of the Library can have been only partial."
Aulus Gellius (2nd century AD) says the Ptolemies collected "nearly seven hundred thousand volumes", burned "not intentionally or by anyone's order, but accidentally by the auxiliary soldiers" (Attic Nights 7.17.3, trans. John C. Rolfe).
Cassius Dio (early 3rd century AD) lists "the docks and the storehouses of grain among other buildings" as burned, "and also the library", whose volumes, "it is said, were of the greatest number and excellence" (Roman History 42.38.2, trans. Earnest Cary).
Orosius (early 5th century AD) says the fire burned 400,000 books "proximis forte aedibus condita" — that happened to be stored in nearby buildings (6.15.31, our translation). He adds that he had seen book cupboards in temples, which, he says, were emptied when they were plundered "by our own people in our own times" (6.15.32, our translation).

So the accounts pull in different directions. Plutarch names the great library, and Gellius says the Ptolemies' collection was all burned. Dio's Greek speaks of storehouses "of grain and of books" (Cary's translation says "the library") and gives the books' number and quality only as hearsay; Seneca gives only a number; and Orosius places the burned books in buildings near the harbour.

Did the Library survive Caesar?

PARTLY TRUE. Two pieces of evidence show that the Museum and a substantial collection of books continued in Roman Alexandria. Neither names the palace Library itself:

Strabo, the geographer who visited Alexandria about two decades later, describes the Museum as part of the royal palaces, with "a public walk, an Exedra with seats", and an official "who formerly was appointed by the kings, but is now appointed by Caesar" (Geography 17.1.8, trans. H. L. Jones). He does not mention the Library.
The emperor Domitian (ruled AD 81–96) restored libraries that had been destroyed by fire, Suetonius says, partly by "sending scribes to Alexandria to transcribe and correct them" (Suetonius, Domitian 20, trans. J. C. Rolfe).

Bagnall concludes there was "still some substantial library in Roman Alexandria", and points out that belief in a Caesarian destruction "requires the uneconomical assumption that the Library was destroyed in the fire but the Museum was not."

The 270s: the palace quarter destroyed

In the early 270s (272 or 273, depending on the scholar), fighting under the emperor Aurelian wrecked much of the palace district, where the Library had stood. The historian Ammianus Marcellinus says the city's "walls were destroyed and she lost the greater part of the district called Bruchion" (22.16.15, trans. Rolfe).

Bagnall notes signs of a consensus forming that major destruction of the palace library most likely came in 273, but adds: "there is no direct evidence concerning the Library in the ancient sources for this".

AD 391: the Serapeum

The Serapeum, a great temple of the god Serapis, was said to hold a second, "daughter" library. In 391 Christians destroyed the temple. Socrates Scholasticus records that "At the solicitation of Theophilus bishop of Alexandria the emperor issued an order at this time for the demolition of the heathen temples", and "Then he destroyed the Serapeum" (Ecclesiastical History 5.16, trans. A. C. Zenos).

DISPUTED. The three accounts of 391 that we read — Rufinus, Socrates, and the pagan writer Eunapius — do not mention books. Eunapius, who was hostile to the Christians, says "they demolished the temple of Serapis and made war against the temple offerings", but says nothing about a library. Orosius, a Christian writing in the early 5th century, is the exception, though he names neither the Serapeum nor a year: he says the book cupboards he had seen in temples were emptied "by our own people in our own times". Scholars read this differently. Bagnall leaves open whether a library survived until the Serapeum fell. Dirk Rohmann, in the journal Klio (2022), argues that Orosius "is our most important witness that the Christian mob destroyed not only the temple building, but also the library". The evidence is thin enough that neither reading is proven.

The Caliph Omar story

MYTH. A famous story has the caliph Umar ordering the books burned after the Arab conquest of Egypt in 642. According to the historian Bernard Lewis:

The earliest source "is a Baghdadi physician called ‘Abd al-Latif, who was in Egypt in 1203", who mentions it only in passing.
The full story appears in Ibn al-Qifti, who "wrote a history of learned men in about 1227", and was later repeated by Barhebraeus, "the principal source used by Western historians".
It "first became known to Western scholarship in 1663".

Even the earliest of these was written more than 550 years after the event. Historians have long doubted it. Edward Gibbon wrote: "For my own part, I am strongly tempted to deny both the fact and the consequences." Alfred Butler, in 1902, noted that the story "makes its first appearance more than five hundred years after the event to which it relates", and that a scholar who appears in it, "Philoponus, was dead long before the Saracens invaded Egypt". Lewis also reports an objection others have raised: the story has the books heating the city's four thousand bathhouses for almost six months, and to do that "a library of at least 14 million books would have been required." Bagnall's verdict on such stories: they "originate centuries after the fact and are surely fiction."

How big was it? Nobody knows

UNVERIFIED. The famous figures do not agree, and some come in different versions in different manuscripts:

Figure	Source	What it counts
40,000	Seneca	books burned in Caesar's war
400,000 (some manuscripts: 40,000)	Orosius	books burned in buildings near the harbour
nearly 700,000 (some manuscripts: 70,000)	Aulus Gellius	books collected by the Ptolemies

Bagnall's conclusion is blunt: "the ancient figures for the size of the Library or the number of volumes lost in the Alexandrine War do not deserve any credence." The actual numbers, he estimates, were "probably lower, perhaps by as much as one order of" magnitude.

What probably happened

Bagnall argues against looking for one culprit: "It is hard to give up villains, but it looks as if we must abandon the search for some individual or small group to blame." Papyrus rolls decay, and without constant recopying a collection wears out; in his words, "Even without hostile action, then, the Library, or Libraries, of Alexandria would not have survived antiquity."

How much each event contributed — Caesar's fire, the destruction of the 270s, the loss of the Serapeum, and centuries of neglect — is still argued. What the scholars we read agree on is narrower: no single event destroyed all of Alexandria's books.

Our reading

A library that died in one fire makes a better story, and each version comes with a villain that suits the teller. The evidence points somewhere less dramatic and more useful: collections survive only while someone keeps paying to copy, repair and use them.

Sources

Roger S. Bagnall, "Alexandria: Library of Dreams", Proceedings of the American Philosophical Society 146:4 (2002).
Caesar, De Bello Civili III.111 (Latin text, The Latin Library).
Seneca, De Tranquillitate Animi 9.5, trans. Aubrey Stewart (1900).
Plutarch, Life of Caesar 49.6, trans. Bernadotte Perrin, Loeb (1919).
Aulus Gellius, Attic Nights 7.17.3, trans. John C. Rolfe, Loeb (1927).
Cassius Dio, Roman History 42.38.2, trans. Earnest Cary, Loeb (1916).
Orosius, Historiae adversus paganos 6.15.31–32 (Latin text, The Latin Library).
Strabo, Geography 17.1.8, trans. H. L. Jones, Loeb (1932).
Suetonius, Domitian 20, trans. J. C. Rolfe, Loeb (1914).
Ammianus Marcellinus 22.16, trans. John C. Rolfe, Loeb (1940).
Socrates Scholasticus, Ecclesiastical History 5.16, trans. A. C. Zenos.
Eunapius, Lives of the Philosophers and Sophists, trans. Wilmer Cave Wright, Loeb (1921).
Rufinus, Church History 11.23, trans. Philip R. Amidon (1997).
Bernard Lewis, "The Vanished Library", letter, The New York Review of Books (27 September 1990).
Dirk Rohmann, "The Destruction of the Serapeum of Alexandria, Its Library, and the Immediate Reactions", Klio 104.1 (2022).
Edward Gibbon, The History of the Decline and Fall of the Roman Empire, ch. 51.
Alfred J. Butler, The Arab Conquest of Egypt (Clarendon Press, 1902), ch. XXV.

Checked September 2026.

The Machines Behind Modern Music: How MIDI and the Roland TR-808 Came About

Hacks Vitae — Thu, 17 Sep 2026 00:00:00 GMT

Two machines from the early 1980s still shape how music is made. MIDI is the standard that lets keyboards, computers and music software talk to each other. The Roland TR-808 is a drum machine that was not a commercial success at first and became famous later. This page explains how each came about, using the records of the MIDI Association, Roland and the people involved — including the places where those records disagree.

Part 1: MIDI

How it started (1981–1982)

According to the MIDI Association's history, the effort began when Roland's president, Ikutaro Kakehashi, "first approached Tom Oberheim at the June 1981 Chicago NAMM show". Oberheim recommended that the Japanese companies also contact Dave Smith of Sequential Circuits, who previewed his own proposal to Oberheim that summer, and in October 1981 Smith and Chet Wood presented a paper at the Audio Engineering Society convention proposing a "Universal Synthesizer Interface": "a specification designed to enable inter-connecting synthesizers, sequencers and home computers with an industry-wide standard interface."

That first proposal was different from the MIDI we know. A 1982 Sequential document quoted by the MIDI Association says: "The USI was proposed to be serial, operating at 19.2 kBaud, with TTL levels, and connected through phone jacks." A meeting at the winter NAMM show in early 1982 decided to "add optoisolation to prevent audio ground loops, and to increase the speed to 31.25 kBaud".

Even the name was negotiated. "The Japanese suggested UMI Universal Music Interface (You-Me) and Dave Smith countered with the name we all know today- MIDI".

Not a two-man invention

Smith and Kakehashi are the names usually given. The MIDI Association's own history puts it differently: "MIDI was not developed by two people or two companies". It credits Roland engineer Tadao Kikumoto with work that "introduces the concepts of Running Status, Status Bytes and Data Bytes", and records that when Dave Smith suggested 16 channels, a July 1982 fax from Roland, acting for the Japanese companies, agreed to "Increase the number of channels to 16 channels".

Even MIDI's round five-pin plug has a link to the second half of this story. Roland proposed "using a 5 PIN DIN which Roland was already using for DIN Sync in their TR-808 drum Machine".

The first public demonstration (January 1983)

"1983 Winter NAMM Anaheim, California Jan 21-23 1983 The stage was finally set for the first public demonstration of MIDI." The Recording Academy describes what happened: "They connected two competing manufacturers' electronic keyboards, the Roland JP-6 synthesizer and Sequential Circuits Prophet 600".

It was not an instant hit. The MIDI Association says the idea "that instantly MIDI became an overnight success was very far from the reality".

The written standard came later that year: "The IMA MIDI Spec document dated August 5, 1983 is the one that was registered with the United States Copyright Office". It was short: "The early MIDI 1.0 specification is a very short 14 page document."

Recognition

In 2013 the Recording Academy honoured the two leads: "Ikutaro Kakehashi and Dave Smith receive Technical GRAMMY Award from the Recording Academy®". The ceremony "was held during GRAMMY Week on Saturday, Feb. 9, 2013."

MIDI 2.0

MIDI 2.0 arrived 37 years after the first demonstration. The core MIDI 2.0 specification overview, published by Japan's AMEI and the MIDI Manufacturers Association, is dated 20 February 2020. The MIDI Association's own timeline gives two adoption dates on the same page: "January of 2020" and "February 20, 2020".

It does not replace the original: "MIDI 2.0 is an extension of MIDI 1.0. It does not replace MIDI 1.0". The MIDI Association names two main advances: higher resolution and "Profiles" that let devices agree how to behave. Original MIDI messages "were primarily limited to 7-bit data (0–127)". MIDI 2.0 adds "16-bit Note Velocity (0–65,535)" and "32-bit Controller Data" — so "32-bit" applies to controllers, not to velocity.

Part 2: The Roland TR-808

What it was (1980)

Roland's company history lists it simply: "1980 TR-808 Rhythm Machine This programmable rhythm machine allowed musicians to program an entire song". Users could build their own patterns "Through a row of 16 distinctively colored buttons".

Who designed it

Roland's own account names a team: "Working under Tadao Kikumoto (seen here with an early prototype), Hiro Nakamura was tasked with developing the analog voice circuits". Kikumoto has described his own part: "I proposed creating a Drum Synthesizer, instead of an expensive PCM", and "the expandable Decay knob of the TR-808 was my intention".

Why it sounds the way it does

The 808's sounds are made by analog circuits, not recordings of real drums. Roland explains why: "Memory chips (needed to play back sampled recordings of “real” drum sounds) were too expensive at the time, so the engineers used analog synthesis".

Roland also tells a story about its parts: "Transistors that had been rejected as “out of specification” were purchased by Roland and used as part of the TR-808’s sound-generating capability." According to Roland, "One tiny component, the 2SC828-R, was so important to the final sound that, once supplies were exhausted, the TR-808 was discontinued." We did not find an independent engineering source for this; it is Roland's account.

A slow start

The 808 is often called a commercial failure. Roland's own words are softer: "it wasn’t considered a commercial success", and it was "only modestly successful when first released". Digital drum machines were arriving: "In 1982, the competitor LinnDrum was launched, helping shift the trend away from analog and towards digital sample-based drum machines."

Kikumoto says the team knew nothing of its popularity in the US and Europe "until 1992, when a well-informed guy from an overseas department told us about it".

How long it was made

Roland's own pages do not settle it. One says: "The Roland TR-808 was officially in production for just two years—from 1980 to 1982. Around 12,000 units were manufactured". Another says only that it was "out of production by 1983", which fits either year. Wikipedia's article on the machine says production ended in 1983. We could not settle whether production ended in 1982 or 1983.

Early records

Roland's editorial site names several early uses. Kikumoto recalls that Roland's Japanese sales department "seemed to have rented one" to Yellow Magic Orchestra just before launch, and Roland's editorial site says: "The year of its release, Roland’s seminal drum machine appeared live for the first time in a performance of YMO’s “1000 Knives.”" It became "a crucial component of the band’s 1981 album, BGM." Roland also notes that "Afrika Bambaataa famously used the 808 on “Planet Rock.”" (1982), and that the machine reached "mainstream success that same year with Marvin Gaye’s hit" — "Sexual Healing" (1982). These examples come from Roland's own site.

Phil Collins's "In the Air Tonight" used the 808's predecessor, the CR-78. Roland says the CR-78 "became popular with established musicians like Phil Collins, who famously used it for the core groove on “In the Air Tonight.”"

The re-creations

Roland has since made new versions. The TR-8 (announced in 2014) is digital; it uses "Analog Circuit Behavior (ACB) technology", which Roland describes as "part-by-part analyses and faithful recreation of tonality and behavior". The TR-08, announced on "August 8, 2017", likewise uses ACB "to carefully recreate all the details and quirks of the original hardware by modeling each analog circuit". The TR-1000, from 2025, is different: Roland calls it "the first Roland drum machine with a true analog engine in over 40 years", with circuits from the TR-808 and TR-909 "faithfully recreated from the original designs".

Our reading

Both stories are usually told as the work of one or two people and a single breakthrough moment. The records say otherwise: MIDI was negotiated between competing companies over several years, and the 808 was a team's compromise with the price of memory chips that was not considered a commercial success and became famous only after later musicians took it up. Where the records disagree on a date, we have given both rather than pick one.

Sources

The MIDI Association: "MIDI History Chapter 6 — MIDI Begins 1981–1983"; "Chapter 7 — MIDI Associations 1983–1985"; "55th GRAMMY Award for MIDI"; "Detailed timeline of MIDI 2.0 developments since January 2020"; "The State of MIDI 2.0" (February 2026 update).
Audio Engineering Society e-library: Smith and Wood, "The 'USI', or Universal Synthesizer Interface", Convention 70, paper 1845 (1981).
Recording Academy: "Technical GRAMMY Award: Ikutaro Kakehashi and Dave Smith".
AMEI and MMA, M2-100-U MIDI 2.0 Specification Overview, version 1.0 (20 February 2020).
Roland: company history; "The Future Is Still 808"; "Tadao Kikumoto: An Exclusive Conversation" (2020); "Yellow Magic Orchestra and the 808 revolution"; "Nothing sounds quite like an 808"; TR-8, TR-08 and TR-1000 product pages; TR-08 press release (8 August 2017).
Wikipedia, "Roland TR-808"; Sonic State, TR-8 review (14 February 2014).

Checked September 2026.

Website Visualizers Compared: Sitemaps, Crawl Maps, Device Previews, Heatmaps and AI Site Builders

Hacks Vitae — Thu, 17 Sep 2026 00:00:00 GMT

"Website visualizer" is not an established product category, and none of the vendors below uses the term on the pages we read. In this guide it means a tool that turns a website — its structure, its look on different screens, or how visitors move through it — into something you can see. That is our own working definition, and it covers five kinds of tool, one per section.

Everything was read from each company's own pages on 17 September 2026. This is not a ranking, and nothing here is legal advice.

How to read this guide

Prices are shown as each page displayed them. Some pages were served in euros or pounds, others in US dollars, depending on where they were read. We have not converted anything. Prices vary by country and change often.
"Could not read" means the price only appears after the page runs JavaScript, or its billing period was not stated in the page text. We did not guess.

1. Visual sitemap builders

These draw a site's pages as a tree, either from scratch or by crawling an existing site.

Tool	Builds from a live site?	Free version	Paid (as shown, USD)
Octopus.do	Yes: 300 pages free, 3,000 paid	1 user, 1 project, "no time limit"	Pro $16/month, or $14/month billed yearly
Slickplan	Yes	"There will always be a free version of Slickplan" (one sitemap, one user); "14 day free trial, no credit card required"	Basic $10.79/month billed annually
FlowMapp	Yes: "Generate sitemap from URL"	1 project, 20 pages	Pro $15/month, or $8.25/month billed annually
VisualSitemaps	Yes	25 pages a month	Mini $29/month; Pro $99/month
WriteMaps	Yes: "Crawl websites"	2 projects, 10 pages per project	Unlimited $16/month
GlooMaps	No, drawn by hand	Free, "No user account needed"	—
PowerMapper	Yes, desktop crawler	Free trial	Could not read

Details worth knowing:

Octopus.do lets you "Start off a project with crawling, using a template, or AI-generate it from a prompt".
GlooMaps: "The sitemap is available for 14 days unless re-visited." Its footer reads "©2018".
PowerMapper runs on "Macintosh and Windows" and "builds a visual site map of web site structure automatically."
Several of these also add AI generation: Slickplan mentions its "AI assistant", FlowMapp an "AI sitemap generator", and WriteMaps says "Use AI to generate sitemaps".

2. SEO crawl maps

SEO crawlers can draw the pages they find as a diagram of the site's internal links or folders.

Tool	What it draws	Free version	Paid (as shown)
Screaming Frog SEO Spider	Force-directed diagrams and tree graphs	Up to 500 URLs; site visualisations included	£199 a year on its product page
Sitebulb	Interactive crawl maps	14-day trial	Desktop prices could not read; Cloud "From £95/mo"
JetOctopus	"Crawl budget visualization"	7-day trial, 10k URLs, no card	Could not read the billing period
Ahrefs Free (formerly Ahrefs Webmaster Tools)	Site Audit charts and graphs	Free for verified sites	Paid Ahrefs plans from €27/month as shown

Details worth knowing:

Screaming Frog: "The Screaming Frog SEO Spider is free, however you can purchase a licence to crawl more than 500 URLs and have access to advanced features." In its pricing table, "Site Visualisations" is ticked for the free version. It runs on "Windows, macOS and Linux", and its crawl visualisations show the site "by shortest path to a page" and, by default, not every link; right-clicking a node and choosing "Show Inlinks" updates a force-directed diagram "to show all incoming links". "Licences are per user" and "Licences last 1 year."
Sitebulb runs on Windows or Mac, and its pricing page lists Crawl Maps under "All plans get".
Ahrefs says "Ahrefs Webmaster Tools is now Ahrefs Free", that it can be used "on an unlimited number of verified websites", and describes its Site Audit output as "Visualize data with charts and graphs" rather than a link map.
Semrush is now owned by Adobe, which "today announced the completion of its acquisition of Semrush Holdings, Inc." on 28 April 2026.

3. Device previews and mockups

These show how a page looks on different screen sizes, or place a screenshot in a device frame.

Tool	What it does	Free version	Paid (as shown)
Responsively App	Many screen sizes side by side	Free and open source	—
Polypane	Browser with synced panes	"Use for free for 14 days"	Could not read
Sizzy	Browser with device previews	14-day free trial	$15/month, or $12/month billed $144 a year
Chrome DevTools device mode	Built-in screen simulation	Free, in Chrome	—
BrowserStack	Real browsers and devices in the cloud	Free trial	Freelancer $12.50/month billed annually, or $19 month-to-month
Mockuuups Studio	Device mockups, including from a live URL	Free mockups	Professional $15 billed per month, or billed $120 per year
Screely (screely.app)	Screenshot frames in the browser	Free	—

Details worth knowing:

Responsively App: "Any click, scroll or navigation that you perform in one device will be replicated to all devices in real-time." It is licensed under the AGPL-3.0, and its latest release is 1.18.0 (17 February 2026).
Polypane works on "Windows, Mac, Linux". Sizzy lists "macOS, Windows and Linux".
Chrome's own documentation is candid about device mode: "Think of device mode as a first-order approximation of how your page looks and feels on a mobile device." It adds: "With device mode you don't actually run your code on a mobile device."
BrowserStack says it covers "3500+ real desktop & mobile browser combinations".
LambdaTest is now TestMu AI. Its announcement says: "LambdaTest became TestMu AI on January 12, 2026".
Screely: the screely.app tool says images are "Processed locally on your device. No uploads, no account, no watermark." The older address, screely.com, no longer hosts a screenshot tool: on 17 September 2026 it served Thai-language pages about online football betting. We could not confirm whether screely.app is by the same maker.

4. Heatmaps and session replay

Heatmaps show where visitors click and scroll in aggregate; session replay records what individual visitors did.

Tool	Free version	Paid (as shown)	What the vendor says about masking
Microsoft Clarity	"No limits on traffic"	—	"Content in the input boxes is masked in all modes"
Hotjar (priced through Contentsquare)	200k monthly sessions, 1 project	Growth from $49, from 7k monthly sessions; billing period could not read	"By default, Hotjar suppresses user input."
Mouseflow	500 sessions a month	Essential 25 a month billed annually, or 39 month to month, for 5,000 sessions; the page's structured data gives the currency as EUR	"we do not record keystrokes by default"
Fullstory	30,000 monthly sessions	Could not read	Offers a "Private by Default" mode
Crazy Egg	30-day free trial	Starter $29/month; "All plans are billed annually."	Could not find
PostHog	5K recordings a month	Usage-based beyond that	Inputs: "we mask these by default"
Lucky Orange	7-day free trial	Build $32/month paid annually, or $39/month	Could not find
Smartlook	Existing accounts only	No new plans or renewals sold	—

Details worth knowing:

Smartlook is shutting down. It says "Smartlook has joined Cisco!", that from "31 May 2026 – End of Sale (EoS): No new Smartlook subscriptions will be sold", and that on "30 September 2027 – Platform Decommissioned: The Smartlook platform will be deactivated".
Microsoft Clarity says it "masks sensitive content on your website by default". Since 31 October 2025, for visitors from the EEA, UK and Switzerland, "If cookie consent is not provided to Clarity by a website owner, Clarity will not place a cookie", and "session recordings, heatmaps, and funnel tracking, may be limited, incomplete, or unavailable."
Hotjar also notes that "Hotjar disables most content suppression by default", and that changing the settings "will not apply the updated settings retroactively".
PostHog: "Our privacy controls run in the browser or mobile app. So, masked data is never sent over the network to PostHog."
Fullstory offers settings to "Capture data only with user consent".

What regulators say

These are quoted, not interpreted. Which rules apply to a particular site depends on where it and its visitors are.

UK — the ICO finalised its guidance on storage and access technologies on 29 April 2026. Under "You must obtain consent for:", it lists "logs or recordings of individual visitors to your website and the actions they took (if not obtained for the purposes of security)". In a table of activities "when using aggregate statistical information" that are "likely to meet the statistical purposes exception", it lists "User interactions with pages on your website (eg average scroll depth or the total number of hits on sections of a page)". It adds: "The statistical purposes exception does not allow you to monitor or track individual visitors to your service." The guidance does not use the words "session replay" or "heatmap".
France — the CNIL has published a draft recommendation on session replay; its public consultation closed on 22 April 2026, and we found no final version on its site. The CNIL describes these tools as recording "mouse movements, screen touches, clicks, page scrolling and, in some cases, forms inputs". The draft says (in the CNIL's English courtesy translation): "The purposes pursued by the deployment of session replay tools are subject to the prior consent of users."
EU — the EDPB's Guidelines 2/2023 (version 2.0, adopted 7 October 2024) set out what counts as storing or accessing information on a user's device under the ePrivacy consent rule (Article 5(3)), including JavaScript that makes the browser send information. They do not name session replay, and they "do not address the circumstances under which a processing operation may fall within the exemptions from the consent requirement".

5. AI tools that draft a site's structure

Many AI builders turn a prompt into pages. The ones below say they produce a sitemap, page structure or screens.

Tool	Structure output	Free version	Paid (as shown, USD)
Relume	"Brief, sitemap, wireframe, designs"	"Free forever · No card needed"	Pro from $14/month
Framer AI	Editable pages and sections	$0, "500 AI credits to try"	Basic $10/month, Pro $30/month on yearly billing
Uizard	"Generate projects, screens, and themes"	3 AI generations a month, 2 projects	Pro $12/month billed annually
Google Stitch	UI designs and front-end code	Could not read	Could not read

Details worth knowing:

Relume says it can "Generate a full site structure". The old relume.io address now redirects to relume.ai.
Framer: "Framer's AI agent creates editable pages, sections, copy, and visuals directly in your project." Its old /wireframer/ page now redirects to /ai/.
Google Stitch began in May 2025 as "a new experiment from Google Labs", and on 18 March 2026 Google said: "Today, we are evolving Stitch into an AI-native software design canvas." The old Galileo AI address, usegalileo.ai, now redirects to Stitch.

Questions to ask before you choose

Do you need to see structure, screens or behaviour? Those are sections 1, 2 and 5 (structure), 3 (screens) and 4 (behaviour).
Does the free version cover your site's size? Check page, URL or session limits.
Is the price monthly or billed yearly, and in which currency?
Is the tool still being sold? Smartlook shows that a working product can be wound down.
If you record visitors, what do your visitors' local rules require? Read the regulator's own guidance.

Sources

Each company's own product, pricing, documentation, help and blog pages, and each regulator's own guidance, read on 17 September 2026: Octopus.do (pricing, home); Slickplan (home, pricing); FlowMapp (pricing); VisualSitemaps (pricing); WriteMaps (pricing); GlooMaps (home); PowerMapper (product and site-structure pages); Screaming Frog (SEO Spider page, pricing, visualisations tutorial); Sitebulb (crawl maps, pricing, Cloud); JetOctopus (pricing); Ahrefs (Ahrefs Free, pricing); Adobe newsroom (Semrush acquisition, 28 April 2026); Responsively App (home, GitHub repository); Polypane (pricing); Sizzy (home); Chrome for Developers (device mode); BrowserStack (pricing); TestMu AI (rebrand announcement); Mockuuups Studio (pricing); Screely (screely.app, and screely.com as observed); Microsoft Clarity (home, consent update, masking and Consent API documentation); Contentsquare (pricing) and Hotjar (help centre); Mouseflow (pricing, help centre); Fullstory (plans, help centre); Crazy Egg (pricing); PostHog (pricing, session replay privacy); Lucky Orange (pricing); Smartlook ("The Future of Smartlook"); UK Information Commissioner's Office (guidance on the use of storage and access technologies); CNIL (session replay consultation and draft recommendation); European Data Protection Board (Guidelines 2/2023, version 2.0); Relume (pricing); Framer (AI and pricing pages); Uizard (pricing); Google for Developers blog (Stitch, 20 May 2025) and Google blog (18 March 2026).

Checked September 2026.

What Is a Frontier AI Model? 8 Definitions That Don't Agree

Hacks Vitae — Thu, 17 Sep 2026 00:00:00 GMT

"Frontier model" is used as if everyone agrees what it means: the most advanced AI systems of the day. In practice, the organisations that define it draw the line in very different places. Some ask how capable a model is, some how dangerous it could be, some how much computing power was used to train it, and one leaves the answer to a classified test.

This page sets out eight definitions in their own words, so you can see what someone means when they use the term. It is about the meaning of a phrase — not about what any law requires, and not legal advice.

First, what "training compute" means

Several definitions below count computing power. The AI research group Epoch AI explains the unit: "Compute usage is commonly measured as the number of floating point operations (FLOP) required to train the final version of the system". The thresholds below are written as powers of ten — 10^25 is a 1 followed by 25 zeros.

1. The industry group: a capability definition that changed

The Frontier Model Forum was announced on 26 July 2023 as "a joint announcement and effort between OpenAI, Anthropic, Google, and Microsoft." Its launch post described frontier models as "large-scale machine-learning models that exceed the capabilities currently present in the most advanced existing models, and can perform a wide variety of tasks".

Its membership definition has since changed. Its About page says that "'Frontier AI' refers broadly to those general purpose AI models that constitute the state of the art, a collection which will shift over time as the field progresses." Then, "For purposes related to its membership", it defines a frontier AI model as "a general-purpose model that outperforms, based on a range of conventional performance benchmarks or high-risk capability assessments, all other models" "that have been widely deployed for at least 12 months". It adds that this definition is "intentionally more expansive than only 'current state of the art' models". As read in September 2026, its members are "Amazon, Anthropic, Google, Meta, Microsoft, and OpenAI."

2. The UK government, 2023: "match or exceed" today's best

For the international AI Safety Summit, the UK government published a discussion paper, Frontier AI: capabilities and risks, on 25 October 2023. For the purposes of the summit it described frontier AI as "highly capable general-purpose AI models that can perform a wide variety of tasks and match or exceed the capabilities present in" the most advanced models, adding: "Today, this primarily includes large language models".

3. The research paper: defined by danger

A 2023 paper, Frontier AI Regulation: Managing Emerging Risks to Public Safety, by Markus Anderljung and 23 co-authors, drew the line around risk rather than capability alone. It described frontier AI models as "highly capable foundation models that could possess dangerous capabilities sufficient to pose severe risks to public safety".

4. The European Union: a different name and a compute presumption

The EU AI Act uses a different label altogether. Its Article 51 is headed "Classification of general-purpose AI models as general-purpose AI models with systemic risk".

Under Article 51(2), a general-purpose model "shall be presumed to have high impact capabilities" when "the cumulative amount of computation used for its training" "measured in floating point operations is greater than" 10^25. The Act also describes high-impact capabilities by comparison, as "capabilities that match or exceed the capabilities recorded in the most advanced general-purpose AI models".

5. California: more than 10^26 operations

California's SB 53, approved by the Governor on 29 September 2025, defines a frontier model as "a foundation model that was trained using a quantity of computing power greater than 10^26 integer or floating-point operations". The count includes more than the first training run: it "shall include computing for the original training run and for any subsequent fine-tuning, reinforcement learning, or other material modifications".

6. New York: a definition that changed within months

New York's RAISE Act shows how quickly these definitions move.

As first signed (chapter 699, December 2025), a frontier model meant "either of the following": a model trained with more than 10^26 operations, "the compute cost of which exceeds one hundred million dollars"; or a model made by "applying knowledge distillation" to such a model, where that process's compute cost "exceeds five million dollars".
As amended (chapter 96, signed March 2026), the definition follows California's wording: a frontier model "means a foundation model that was trained" using computing power greater than 10^26 integer or floating-point operations.

7. The US federal government: a classified test

Executive Order 14409, "Promoting Advanced Artificial Intelligence Innovation and Security", dated 2 June 2026, uses the term "covered frontier model" but sets no public number. It directs the Secretary of the Treasury, the Secretary of War (through the Director of the NSA) and the Secretary of Homeland Security (through the Director of CISA) to "develop and maintain a classified benchmarking process to assess the advanced cyber capabilities of AI models and determine the threshold at which an AI model should be designated a 'covered frontier model' for the purposes of this order". It adds: "Such a determination shall be made by the Director of NSA".

8. The research tracker: top ten at release

Epoch AI, which maintains a public database of more than 3,600 AI models, uses a moving definition for its own data: "Frontier models are models that were in the top 10 by training compute at the time of their release".

It uses a separate, fixed line for scale: "Large-scale models are models that were trained with over 10^23 FLOP of compute, which is a static threshold".

How many models cross these lines?

Counts depend on estimates of training compute, because not every company publishes it. Epoch AI reported: "As of June 2025, we have identified over 30 publicly announced AI models from 12 different AI developers" above 10^25 FLOP, and "The first model trained at this scale was GPT-4, released in March 2023." For the higher line, a May 2025 Epoch report said "the first model in our dataset estimated to use over" 10^26 FLOP "was Grok-3 from xAI, released in February 2025".

What to take from this

Our reading: when someone calls a model "frontier", it is worth asking which definition they mean.

Capability definitions (the industry group, the UK) are relative, so the frontier moves as models improve — a model that was frontier in 2023 may not be now.
Danger definitions (the 2023 paper) depend on judgements about risk that people disagree about.
Compute definitions (the EU, California, New York) are measurable in principle, but outsiders often have to estimate compute, and a fixed number can be overtaken quickly.
A classified test (the US order) is not published, though the order provides for "sharing such assessments with AI developers and researchers as appropriate".

None of these is the "right" one. They were written by different kinds of organisation, for different purposes.

Sources

Frontier Model Forum, "Introducing the Frontier Model Forum" (26 July 2023); "About us"; "Membership".
UK Department for Science, Innovation and Technology, Frontier AI: capabilities and risks discussion paper (25 October 2023).
Markus Anderljung and colleagues, "Frontier AI Regulation: Managing Emerging Risks to Public Safety", arXiv:2307.03718 (2023).
Regulation (EU) 2024/1689 (AI Act), Articles 3(64) and 51, EUR-Lex.
California SB 53 (2025), Business and Professions Code §22757.11, California Legislative Information.
New York Senate bills S6953-B (chapter 699 of 2025) and S8828 (chapter 96 of 2026), New York State Assembly bill pages.
Executive Order 14409, "Promoting Advanced Artificial Intelligence Innovation and Security" (2 June 2026), whitehouse.gov and Federal Register document 2026-11415.
Epoch AI, "Estimating training compute of deep learning models" (2022); "Data on AI Models"; "Over 30 AI models have been trained at the scale of GPT-4" (updated June 2025); "How many AI models will exceed compute thresholds?" (May 2025).

Checked September 2026.

Who Makes Today's Leading AI Models? The Main Labs and Their Current Models

Hacks Vitae — Thu, 17 Sep 2026 00:00:00 GMT

The companies that build large language models are often called AI labs. Some sell access to closed models through apps and an API. Others also publish the model's weights so anyone can download and run it.

This page lists the main labs and each one's current top model, as read from the labs' own documentation on 25 September 2026. Model line-ups change every few weeks, so treat this as a dated snapshot and check the lab's model page for anything newer.

The table

Labs are listed alphabetically. "Context window" is the figure each lab publishes for how many tokens a model can handle at once. Labs do not all count it the same way: for some it includes the model's reply as well as your input.

Lab	Current top model	Released	Weights	Context window
Alibaba (Qwen)	Qwen3.8-Max; weights published as Qwen3.8-2.4T-A95B	API: 3 August 2026	Downloadable, custom licence	262,144 natively, extensible to 1,010,000 (weights)
Anthropic	Claude Fable 5.1; Claude Opus 5.5	1 September 2026; 22 September 2026	Closed	1M tokens
DeepSeek	DeepSeek-V4-Pro; DeepSeek-V4.1-Flash	13 August 2026 (V4-Pro general release); 10 September 2026	Downloadable, MIT	1M tokens
Google DeepMind	Gemini 3.1 Pro (preview); Gemini 3.8 Flash	19 February 2026; 2 September 2026 (general availability)	Closed	1,048,576
Meta	Muse Spark 1.3	2 September 2026 (Muse Spark first announced 8 April 2026)	Closed	1M tokens
Mistral AI	Mistral Medium 3.5	April 2026	Downloadable, custom licence	256k
Moonshot AI	Kimi K3	July 2026	Downloadable, custom licence	1,048,576
OpenAI	GPT-6 Astra	3 September 2026	Closed	1,050,000
xAI (SpaceXAI)	Grok 4.7	API: 21 September 2026	Closed	500k
Z.ai (Zhipu)	GLM-5.3	18 August 2026	Downloadable, custom licence	1M tokens

What the labs say about their own models

These are the labs' own descriptions, not independent assessments.

OpenAI: its model index says "If you're not sure where to start, use GPT-6 Astra". Its system card is headed "Published September 3, 2026 GPT-6 Astra System Card".
Anthropic: its documentation says to "start with Claude Opus 5.5 for most workloads. Use Claude Fable 5.1 for demanding reasoning and long-horizon agentic work, or when your evals on Claude Opus 5.5 at higher effort still fall short".
Google: it describes Gemini 3.8 Flash as "our most intelligent Flash model, engineered for long-horizon software engineering, autonomous agents, and complex enterprise workflows". Gemini 3.1 Pro is still listed as a preview, with "No shutdown date announced".
xAI: SpaceX announced on 2 February 2026 that it had acquired xAI, and xAI's own pages now use the name SpaceXAI. Its models page calls Grok 4.7 "Our flagship model for code and everything else".
Meta: "Today we are announcing Muse Spark, the first in a new series of large language models built by Meta Superintelligence Labs." At launch it said: "It will be available in private preview via API to select partners, and we hope to open-source future versions of the model." Its developer page now describes the API as "Public preview. Now available with expanded global access." Announcing version 1.3, Meta listed "the Muse Spark open weights release" on its roadmap.
DeepSeek: its model card says the 0813 release is the official V4-Pro, "superseding the preview version, with greatly enhanced agentic capabilities".
Alibaba: its model card says "For the first time, Qwen3.8 brings a Qwen-Max-class model to open release", with "2.4T in total and 95B activated" parameters.
Mistral AI: its documentation says the model was "Released as open weights under a Modified MIT license."
Moonshot AI: "Kimi K3 is an open-weight, native multimodal agentic model and our most capable model to date."
Z.ai: "GLM-5.3 currently supports text-only inputs, with a 1M-token context window and a maximum output length of 128K tokens."

Closed or downloadable?

A closed model's weights are not published: you use it through an app or an API, and some closed models are also sold through other companies' cloud platforms, such as Amazon Bedrock or Microsoft Foundry. A model with downloadable weights can be run on hardware you control — but "downloadable" is not the same as "open-source", and the licence decides what you may do with it. Several of the downloadable models above use custom licences with extra conditions for large companies. See Open-weight is not open-source for what those licences say. Some labs publish both kinds: Meta's Muse Glimmer model (30B parameters), for example, is released with open weights under Apache 2.0.

Models get retired

Labs publish schedules for switching old models off, and those dates matter if you build on a model:

Anthropic's deprecation page says of Claude Opus 4 and Claude Sonnet 4: "These models were retired June 15, 2026."
OpenAI says it notified developers of older GPT-5 and o3 snapshots of "their deprecation and removal from the API on December 11, 2026".
Google's deprecation page notes: "The shutdown dates listed in the table indicate the earliest possible dates".
Plans can also change. DeepSeek's change log says: "we have decided to continue providing API services for DeepSeek V4 Pro after September 14, 2026".

How to check what is current

Each lab keeps an official model page. The ones used for this table were:

OpenAI: developers.openai.com/api/docs/models
Anthropic: platform.claude.com/docs/en/about-claude/models/overview
Google: ai.google.dev/gemini-api/docs/models
xAI: docs.x.ai/docs/models
Meta: developer.meta.com/ai/models/muse-spark
DeepSeek: api-docs.deepseek.com
Alibaba Qwen: huggingface.co/Qwen and docs.qwencloud.com
Mistral AI: docs.mistral.ai/models
Moonshot AI: huggingface.co/moonshotai
Z.ai: docs.z.ai

Our reading: the table tells you who makes what, not which model is best. Labs' descriptions of their own models are not independent assessments, and published scores depend heavily on how models are tested — see AI benchmarks explained. The practical test is how a model does on your own tasks.

Sources

OpenAI: API models index; API changelog; GPT-6 Astra model page; GPT-6 Astra System Card; deprecations.
Anthropic: models overview; release notes; model deprecations.
Google: Gemini API models, changelog and deprecations; Gemini 3.8 Flash and Gemini 3.1 Pro model pages.
xAI: API models and release notes; "xAI joins SpaceX" (2 February 2026), x.ai.
Meta: "Introducing Muse Spark" (8 April 2026), about.fb.com; "Introducing Muse Spark 1.3" (2 September 2026), research.meta.ai; Meta Model API and Muse Glimmer pages, developer.meta.com.
DeepSeek: API news, change log and pricing pages; DeepSeek-V4-Pro-0813 model card, Hugging Face.
Alibaba Qwen: Qwen Cloud model changelog; Qwen3.8-2.4T-A95B model card, Hugging Face.
Mistral AI: Mistral Medium 3.5 documentation; Mistral-Medium-3.5-128B model card, Hugging Face.
Moonshot AI: Kimi-K3 model card, Hugging Face.
Z.ai: release notes; GLM-5.3 guide.

Checked September 2026.

Why AI Makes Things Up: What "Hallucination" Means, Why It Happens and How to Catch It

Hacks Vitae — Thu, 17 Sep 2026 00:00:00 GMT

AI chatbots can write a fluent, confident answer that is simply wrong: a court case that never existed, a quotation nobody said, a reference to a paper that was never published. This is usually called hallucination. This page explains what the word means, why it happens, why nobody can give you a single error rate, and the checks that catch it. It is a snapshot as of September 2026.

What "hallucination" means

OpenAI describes it simply: "By this we mean instances where a model confidently generates an answer that isn’t true."

Researchers often define it differently. A 2022 survey describes hallucination as "the generated content that is nonsensical or unfaithful to the provided source content". It separates two kinds:

Intrinsic: "The generated output that contradicts the source content."
Extrinsic: "The generated output that cannot be verified from the source content". The survey adds: "Notably, the extrinsic hallucination is not always erroneous because it could be from factually correct external information".

A later survey of large language models draws a similar line between being wrong about the world and being wrong about what you gave the model. "Factuality hallucination emphasizes the discrepancy between generated content and verifiable real-world facts", while "faithfulness hallucination captures the divergence of generated content from user input or the lack of self-consistency within the generated content".

In plain words: the model may be wrong about the world, wrong about your document, or contradict itself.

Not everyone likes the word

"Hallucination" is a metaphor, and several researchers think it is a misleading one.

An opinion article in PLOS Digital Health (2023) argued: "The model is not “seeing” something that is not there, but it is making things up." It proposed a term from psychiatry instead: "More accurate terminology is found in the psychiatric concept of confabulation".
A philosophy paper in Ethics and Information Technology (2024) argued that "the models are in an important way indifferent to the truth of their outputs". It used the term "bullshit" in the technical sense given to it by the philosopher Harry Frankfurt: speech produced without regard to whether it is true.
A research team writing in Nature (2024) used "confabulations" for a narrower group of errors — "arbitrary and incorrect generations" — and wrote: "We believe that combining these distinct mechanisms in the broad category hallucination is unhelpful."
A review of how the term is used found "a lack of consistency in how the term is used".

Even OpenAI's researchers note that the word, borrowed from human experience, is imperfect: hallucination in language models "differs fundamentally from the human perceptual experience".

Why it happens

Rare facts cannot be predicted from patterns

Language models learn from huge amounts of text. OpenAI's explanation starts there: pretraining is "a process of predicting the next word in huge amounts of text". Some things follow patterns, such as spelling. Others do not: "But arbitrary low-frequency facts, like a pet’s birthday, cannot be predicted from patterns alone and hence lead to hallucinations."

A 2023 survey makes the same point about knowledge in general: models struggle to "memorize all factual knowledge encountered during pre-training, especially the less frequent long-tail knowledge", and training data "does not include rapidly evolving world knowledge or content restricted by copyright laws".

Tests reward guessing

A paper by researchers at OpenAI and Georgia Tech makes a sharper argument. It was first posted in September 2025 as "Why Language Models Hallucinate" and published in Nature in April 2026 as "Evaluating large language models for accuracy incentivizes hallucinations". The published version says that "dominant headline metrics such as accuracy systematically reward guessing over admitting uncertainty". The 2025 preprint put it more bluntly: "language models are optimized to be good test-takers, and guessing when uncertain improves test performance."

The comparison is a multiple-choice exam with no penalty for wrong answers. If a model does not know someone's birthday and says "I don't know", it scores zero. OpenAI's summary: "If it guesses “September 10,” it has a 1-in-365 chance of being right." The 2025 preprint states it formally: "Under binary grading, abstaining is strictly sub-optimal."

One detail is easy to miss: rare facts set a floor. As an illustration, the published paper says that if 20% of birthday facts appear only once in the training data, then "pretrained models should hallucinate on at least 20% of birthday facts". Pretrained models are models before the extra training that turns them into assistants.

This is the authors' argument, from researchers mostly employed by one AI company, not a settled consensus.

DISPUTED. Whether hallucination can be avoided at all is argued both ways. OpenAI's researchers say it can in principle: responding to the claim that hallucinations are inevitable, they write "They are not, because language models can abstain when uncertain." Their preprint acknowledges that "Many have argued that hallucinations are inevitable". A 2024 paper, "Hallucination is Inevitable", argues that under its formal definition "it is impossible to eliminate hallucination in LLMs". The two sides partly define the problem differently: one counts saying "I don't know" as a way out.

How often does it happen? It depends on the test

There is no single "hallucination rate". Every published figure is a rate on a particular task, graded in a particular way — and figures from different tests cannot be compared. Four examples:

Test	What it measures	Result
SimpleQA (November 2024)	Short fact questions; "Each answer in SimpleQA is graded as either correct, incorrect, or not attempted."	GPT-4o answered almost every question and was wrong 60.8% of the time; Claude 3.5 Sonnet declined 35.0% and was wrong 36.1%. The questions were "adversarially collected against GPT-4 responses".
SimpleQA, newer models (OpenAI, 2025)	The same test, as reported for two OpenAI models	gpt-5-thinking-mini declined 52% and was wrong 26%; OpenAI o4-mini declined 1% and was wrong 75% — with accuracy of 22% and 24%.
PersonQA (OpenAI o3 system card, April 2025)	OpenAI's questions about "publicly available facts about people"	Hallucination rate of 0.33 for o3 against 0.16 for the older o1, while accuracy was 0.59 against 0.47. The card said: "More research is needed to understand the cause of these results."
Vectara leaderboard (updated May 2026)	"This evaluates how often an LLM introduces hallucinations when summarizing a document."	From 1.8% to 24.2% across the models listed, as judged by Vectara's own evaluation model.

The second SimpleQA row shows the pattern the "reward guessing" argument predicts: the model that declined far more often made about a third as many errors at almost the same accuracy (22% against 24%). In the first row, the model that declined more often also got fewer answers right (28.9% against 38.2%), though these are two different models, so the table alone does not show why. The PersonQA row shows that a newer model can hallucinate more: o3 answered more questions correctly than o1 but also made more false claims, which the card linked to o3 making "more claims overall".

When it has mattered

A court filing with invented cases

In Mata v. Avianca, a US federal court in New York found in June 2023 that lawyers "submitted non-existent judicial opinions with fake quotes and citations created by the artificial intelligence tool ChatGPT, then continued to stand by the fake opinions after judicial orders called their existence into question". When one lawyer asked the chatbot whether the cases were real, "ChatGPT responded that it had supplied "real" authorities that could be found through Westlaw, LexisNexis and the Federal Reporter."

The court was clear that the tool itself was not the problem: "Technological advances are commonplace and there is nothing inherently improper about using a reliable artificial intelligence tool for assistance. But existing rules impose a gatekeeping role on attorneys to ensure the accuracy of their filings." Finding bad faith based on "acts of conscious avoidance and false and misleading statements to the Court", it imposed "A penalty of $5,000". Our article on spotting fake quotes and invented sources covers the case in more detail.

It was not a one-off. A public database of decisions by courts and tribunals involving AI-generated false material listed "2041 cases identified so far" when it was updated on 14 September 2026 — by one researcher's count, and it notes that it "does not track the (necessarily wider) universe of all fake citations or use of AI in court filings."

References to papers that do not exist

A 2023 study in Scientific Reports checked citations produced by ChatGPT and found that "55% of the GPT-3.5 citations but just 18% of the GPT-4 citations are fabricated". It added: "Even with GPT-4, however, 70% of the cited book chapters are fabricated." The most useful finding for readers is how convincing the fakes were: "most of the fabricated article, book, and website citations include the names of real journals, publishers, and organizations". A real journal name is not evidence that a paper exists.

A 2024 study on references for systematic reviews found similar problems: "Hallucination rates stood at 39.6% (55/139) for GPT-3.5, 28.6% (34/119) for GPT-4, and 91.4% (95/104) for Bard", using its own rule for what counted as a hallucinated paper.

A customer-service chatbot

In Moffatt v. Air Canada (British Columbia Civil Resolution Tribunal, February 2024), a chatbot on the airline's website told a customer they could apply for a bereavement fare after travelling. The airline's own policy page said otherwise. The tribunal held the airline responsible: "It makes no difference whether the information comes from a static page or a chatbot." It found that "Air Canada did not take reasonable care to ensure its chatbot was accurate" and awarded "$650.88 in damages". The decision does not say what technology the chatbot used, so it is an example of an automated answer being wrong, not proof of a language-model hallucination.

What reduces it — and the limits

Looking things up first. The technique usually called retrieval-augmented generation (RAG) has the model search a collection of documents before answering. Its original 2020 paper reported: "Qualitatively, we find that RAG models hallucinate less and generate factually correct text more often than BART", the same kind of model without retrieval. But retrieval "can be easily impacted by irrelevant retrievals", and the authors of the "Why Language Models Hallucinate" preprint note that scoring "still rewards guessing whenever search fails to yield a confident answer".

Specialist tools still err. A study (2024 preprint; Journal of Empirical Legal Studies, 2025) tested commercial legal research tools from providers that had described their methods as "eliminating" or "avoid[ing]" hallucinations, or had promised "hallucination-free" citations. It found they "each hallucinate between 17% and 33% of the time" — fewer errors than a general chatbot, "While hallucinations are reduced relative to general-purpose chatbots (GPT-4)", but not none: "AI tools for legal research have not eliminated hallucinations."

Letting the model say "I don't know". Anthropic's guidance for developers recommends: "Explicitly give Claude permission to admit uncertainty." It also suggests requiring supporting quotes — "If it can't find a quote, it must retract the claim" — and adds its own caveat: the techniques reduce hallucinations but "they don't eliminate them entirely".

Asking more than once. Answers that change each time can be a warning sign. But the Nature team noted that its detection method "does not guarantee factuality because it does not help when LLM outputs are systematically bad" — a model can repeat the same wrong answer every time.

Changing how tests are scored. OpenAI's researchers propose to "Penalize confident errors more than you penalize uncertainty, and give partial credit for appropriate expressions of uncertainty." That is a proposal, not yet a standard.

Seven checks that catch it

Open every citation yourself. Search for the paper, case or book in a library catalogue, a court database or the publisher's site. Do not ask the chatbot whether it is real — in Mata v. Avianca, it said yes.
Check quotations against the source. If you cannot find the words in the original, do not use them.
Take extra care with rare, specific facts: birthdays, obscure names, one-off details — the kind of rare, specific detail the research points to.
Be careful with anything recent. A model's built-in knowledge stops at its training cutoff unless it searches.
Ask again, but do not treat agreement as proof. Changing answers are a red flag; consistent answers can still be consistently wrong.
Prefer "I don't know" to a confident guess — and tell the tool it is allowed to say it.
If an organisation's chatbot tells you something important, check the organisation's written policy and keep a copy of what the chatbot said.

Our reading: hallucination is not a rare glitch that is likely to vanish with the next model. It follows from how these systems are trained and tested, and even the tools that search and cite still get things wrong. The practical answer is the one we try to apply everywhere: a claim is only as good as the source you can open and read yourself.

Sources

OpenAI, "Why language models hallucinate" (5 September 2025); Kalai, Nachum, Vempala and Zhang, "Evaluating large language models for accuracy incentivizes hallucinations", Nature (22 April 2026); preprint "Why Language Models Hallucinate", arXiv:2509.04664.
"Hallucination is Inevitable: An Innate Limitation of Large Language Models", arXiv:2401.11817.
Ji et al., "Survey of Hallucination in Natural Language Generation", ACM Computing Surveys; arXiv:2202.03629.
Huang et al., "A Survey on Hallucination in Large Language Models", arXiv:2311.05232.
Smith, Greaves and Panch, "Hallucination or Confabulation? Neuroanatomy as metaphor in Large Language Models", PLOS Digital Health (2023).
Hicks, Humphries and Slater, "ChatGPT is bullshit", Ethics and Information Technology (2024).
Farquhar, Kossen, Kuhn and Gal, "Detecting hallucinations in large language models using semantic entropy", Nature (2024).
Maleki, Padmanabhan and Dutta, "AI Hallucinations: A Misnomer Worth Clarifying", arXiv:2401.06796.
Wei et al., "Measuring short-form factuality in large language models" (SimpleQA), arXiv:2411.04368.
OpenAI, "OpenAI o3 and o4-mini System Card" (April 2025).
Vectara, Hallucination Leaderboard, GitHub (updated 11 May 2026).
Mata v. Avianca, Inc., S.D.N.Y., Opinion and Order on Sanctions (22 June 2023).
AI Hallucination Cases Database, Damien Charlotin (updated 14 September 2026).
Walters and Wilder, "Fabrication and errors in the bibliographic citations generated by ChatGPT", Scientific Reports (2023).
Chelli et al., "Hallucination Rates and Reference Accuracy of ChatGPT and Bard for Systematic Reviews", Journal of Medical Internet Research (2024).
Moffatt v. Air Canada, 2024 BCCRT 149.
Lewis et al., "Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks", arXiv:2005.11401.
Magesh et al., "Hallucination-Free? Assessing the Reliability of Leading AI Legal Research Tools", arXiv:2405.20362; Journal of Empirical Legal Studies 22(2) (2025).
Anthropic, Claude Platform Docs, "Reduce hallucinations".

Checked September 2026.

Comeback Stories from History, Checked Against the Record

Hacks Vitae — Wed, 16 Sep 2026 00:00:00 GMT

Comeback stories are a staple of self-improvement writing: a figure everyone respects, a low point, a recovery, a lesson. The shape is satisfying enough that the low point rarely gets checked.

So we took six well-known ones and looked at what the record says. Three hold up, and the details are stronger than the usual summaries. Two are partly true. One does not hold up. After them comes a question about the lesson these stories are usually asked to carry.

The one that was assembled later

Lincoln's list is familiar. One common version runs: "Failed in business—1831. Defeated for legislature—1832. Sweetheart died—1835. Nervous breakdown—1836", and so on, one setback after another, until "Elected president—1860".

The list does not come from Lincoln's lifetime. In a 2014 essay for Slate, Scott A. Sandage traced its spread to 1920, when William Randolph Hearst's International News Service first wired it out as a filler story that was reprinted in hundreds of newspapers, magazines and books. Hearst credited it to "the Sing Sing Bulletin", a prison newspaper, but the same short text had already appeared in several newspapers in 1918 and 1919, and a longer version had been published in 1915.

The defeats themselves were real. In 1856 Lincoln wrote privately that "with me, the race of ambition has been a failure—a flat failure." But the list leaves out the wins. He was elected to the Illinois legislature in 1834, built a law practice, and was elected to the Thirtieth Congress in August 1846. His former private secretary, John G. Nicolay, listed the defeats himself in A Short Life of Abraham Lincoln (published October 1902), and then wrote: "Failures? Not so. Every seeming defeat was a slow success."

The list works because it has a rhythm. It is not a record.

The one most worth telling

Grant. It is told less often, perhaps because it is grimmer than the format usually allows.

Grant was a partner in Grant & Ward, a Wall Street firm his son had set up with Ferdinand Ward. On paper Grant was a millionaire. In fact Ward was running a Ponzi scheme, paying out dividends from new investments, and when the firm collapsed in May 1884 Grant lost his entire fortune. In History.com's account he had $80 to his name, and his wife Julia had another $130.

He had turned down earlier suggestions, including Mark Twain's, that he write his memoirs. Now, needing to pay back a loan and support his family, he turned to writing, first articles for Century Magazine. In October 1884 he learned he had incurable throat cancer. On 27 February 1885 he signed with Charles L. Webster and Company, the new publishing firm Twain had set up with his nephew, for a seventy percent royalty on the profits.

He worked through early 1885, at times writing as much as 10,000 words in a single day, and finished three days before he died on 23 July 1885. The National Park Service says the book was an instant best-seller and that the first payment to his widow was $200,000. History.com puts the sales at more than 300,000 copies of the two-volume set and Julia's royalties at $450,000.

What the record does not contain is worth noticing: no epiphany, no mindset shift, no moment of deciding not to quit. A man who could write did the work available to him, against a clock, and it was enough.

What the true ones have in common

Put the three that hold up side by side (Grant, Shackleton and Churchill) and one thing repeats. In each, the turn came from a skill the person already had.

Grant could write. His Personal Memoirs are still in print.
Shackleton's men got home partly because Frank Worsley could navigate. Shackleton chose him for the boat journey because of "his accuracy and quickness as a navigator", and Worsley brought the James Caird to South Georgia with only four sightings of the sun on the way.
Churchill kept doing the work he was known for. After losing the 1945 election he wrote his memoirs of the Second World War, which ran to six volumes (1948–1953/54), and he was prime minister again after the Conservative victory of 1951.

This is our reading of the cases, not a finding. Keeping going shows up in every one of these stories, and in many that ended badly, so it cannot be the whole explanation on its own. Persistence looks like the price of entry; the second skill looks like what turned the corner. It makes a worse poster, but it may be the more useful idea.

A different lesson

If you are in the bad part of one of these stories, the question the cases raise is not how do I keep going. You are probably already doing that.

It is what can I do that I have been treating as secondary, and can it be pointed at the problem in front of me.

Sometimes the answer is nothing, and the story does not end well. Those versions exist in far greater numbers than the ones that get printed, which is one reason these stories are worth checking. The same problem is the whole subject of survivorship bias.

Sources

Scott A. Sandage, "'I Have Never Been Up': The persistent, powerful meme of Abraham Lincoln as a 'failure at 50'", Slate, 28 May 2014 (the failures list, the 1920 International News Service story and its earlier versions; Lincoln's 1856 note).
Abraham Lincoln, fragment on Stephen A. Douglas (1856), in The Collected Works of Abraham Lincoln, vol. 2, ed. Roy P. Basler (1953).
John G. Nicolay, A Short Life of Abraham Lincoln (Century, 1902), Project Gutenberg.
US National Park Service, "The Personal Memoirs of Ulysses S. Grant".
Christopher Klein, "Bankrupt and Dying from Cancer, Ulysses S. Grant Waged His Greatest Battle", History.com.
Benjamin Shapell and Sara Willen, "Mark Twain Eulogizes General Grant, Whose Memoirs He Would Publish", Shapell Manuscript Foundation, 11 August 2015 (the contract date).
Ernest Shackleton, South: The Story of Shackleton's Last Expedition 1914–1917, Project Gutenberg.
James Caird Society, "The James Caird".
Frank Lewis Dyer and Thomas Commerford Martin, Edison: His Life and Inventions (1910), Project Gutenberg, for Walter S. Mallory's recollection.
Quote Investigator, "I Have Gotten a Lot of Results! I Know Several Thousand Things That Won't Work" (2012, updated 2025).
Vincent van Gogh, letter 855 to Anna van Gogh-Carbentus, 19 February 1890, Van Gogh Letters (Van Gogh Museum and Huygens ING); Van Gogh Museum, "How many paintings did Vincent sell during his lifetime?".
Martin Bailey, "Why did Van Gogh fail to sell his work?", The Art Newspaper, 3 June 2022.
GOV.UK, "Past Prime Ministers: Sir Winston Churchill"; the Nobel Prize, "Winston Churchill: Biographical"; Churchill Archives Centre, "Sir Winston Churchill: a biography".

Checked September 2026. Where an account is uncertain (how many paintings Van Gogh sold is the clearest case), this page says so rather than choosing the tidier number.

Comeback Stories, Part 2: Sanders, Jordan, Disney, Beethoven and Jobs

Hacks Vitae — Wed, 16 Sep 2026 00:00:00 GMT

Part one checked six comebacks. In the three that held up, the turn came from a skill the person already had, not from keeping going alone. This part looks at seven more.

Three hold up. Four are real stories carrying a detail that was added or changed along the way.

What got added to the partly true stories

Look at what changed in the four that are partly true:

Jordan is "cut" from basketball, when the cut was from the varsity and he spent that season on the junior varsity.
Sanders gets an exact number: 1,009 rejections.
Disney gets an exact insult ("lacked imagination"), and a newspaper in place of the art studio that actually laid him off.
Astaire gets a longer, funnier screen-test note.

The additions rarely change the outline of what happened. Mostly they make it quotable. Our reading is that a number and a put-down are the two things a motivational post needs, and the record rarely supplies either.

The Jordan story is the most instructive

It has been told for decades, including in an NBA Entertainment film, and the part that changed is the part that makes it a story.

Laney High School in Wilmington, North Carolina, started with the 10th grade. In 1978 Jordan arrived for his sophomore year at "a skinny 5 feet, 9 inches", according to a 1998 newspaper account by Kevin Sherrington of The Dallas Morning News. The varsity coach, Clifton "Pop" Herring, almost never carried sophomores, and that year he had one opening for a sophomore. His team lacked height, so the place went to Leroy Smith, who was 6ft 7in.

Jordan felt it. "It was embarrassing, not making that team," he said in 1991. "They posted the roster and it was there for a long, long time without my name on it." He became the centerpiece of the junior varsity, scoring more than 40 points in a couple of games, grew from 5ft 9in to 6ft 3in before his junior season, and spent his last two years on the varsity.

For a sophomore at that school, missing the varsity was the normal outcome, not a talent being written off. The anger was real, and he used it. That is still a useful story.

The pattern from part one, again

Jobs is the clearest case of what part one found. He founded NeXT in 1985 after he resigned from Apple, "ousted as chairman by a boardroom coup", in the Computer History Museum's words. By 1996 Apple needed a modern operating system and had failed to produce one of its own, so it bought NeXT "in order to make NeXTSTEP the basis for what eventually became Mac OS X". Jobs came back with it, and was firmly in charge by August 1997. He returned carrying the software Apple needed.

Beethoven fits in a different way. His hearing had been failing since his late twenties, and from 1818 friends wrote their side of conversations in books. He went on composing through all of it, including the Ninth Symphony. Our reading is that what carried him was decades of training that the hearing loss could not take away.

This is our reading of the cases, not a law: keeping going kept them in the game, and something they had already built won it.

Sources

Kevin Sherrington, "The Unkindest Cut Of All", The Dallas Morning News, reprinted in The Seattle Times, 11 January 1998.
Victoria Dawson, "How Colonel Sanders Made Kentucky Fried Chicken an American Success Story", Smithsonian Magazine; Rosenberg International Franchise Center, University of New Hampshire, "Harland 'Colonel' Sanders".
D23, "Ub Iwerks"; The Walt Disney Family Museum, "Ub Iwerks: Master of Animation and Technology" and "100 Years Ago: Walt Goes West" (15 August 2023).
Robert DeMarco, "Walt Disney Never Filed Bankruptcy", American Bankruptcy Institute (archived copy); National Archives at Kansas City, "Walt Disney Bankruptcy".
Quote Investigator, "Can't Act; Slightly Bald; Can Dance a Little" (2014, updated 2025).
Beethoven-Haus Bonn, on his hearing; George Grove, Beethoven and His Nine Symphonies (3rd ed., 1898); Thomas et al., "Aetiology of Ludwig van Beethoven's hearing impairment", European Archives of Oto-Rhino-Laryngology (2021); uDiscover Music on Theodore Albrecht's research.
Computer History Museum, "NeXT: Steve Jobs' Dot Com IPO That Never Happened" and "The Deep History of Your Apps"; Pixar, "Our Story".
Jack Loftus, "Praising Failure: James Dyson Talks Vacuum's 5,127 Prototypes", Gizmodo, 10 April 2011; London Business School, "If at first you don't succeed…".

Checked September 2026.

Design Styles Explained: 12 Movements and Where Their Names Come From

Hacks Vitae — Wed, 16 Sep 2026 00:00:00 GMT

Design labels get used loosely, and most of the time it does not matter. It starts to matter when you search for references, brief someone or buy a typeface, because the label is what you type.

So here are twelve labels that often get mixed up, with the dates and the one or two tell-tale signs that help tell them apart, and where each name comes from.

The one mix-up worth fixing first

Victorian is a period, not a style. It covers everything made during Queen Victoria's reign, from 1837 to 1901, which takes in several very different looks. When a poster with swirling flowers, flowing hair and whiplash curves is labelled "Victorian", it is usually Art Nouveau: a style of its own, which emerged in the 1890s and so overlaps only with the last years of Victoria's reign.

A two-second test for the famous pair

Art Nouveau and Art Deco are easy to confuse: the names sound alike and one came after the other. Tate calls Art Deco both a successor to Art Nouveau and a reaction against it. A quick test:

Does it curve like a plant? Art Nouveau, which emerged in the 1890s and lasted until the First World War.
Is it geometric and angular? Art Deco, from the 1920s and 1930s.

Neither test is perfect. Tate notes that plant forms and sinuous curves remained in some Art Deco design.

Names that came later

Several of these labels were given, or caught on, after the fact, often through writers and critics looking back:

Art Deco appeared as a name in 1966 and spread with Bevis Hillier's 1968 book.
Mid-century modern is first recorded in 1953, but it caught on after Cara Greenberg's 1984 book.
Brutalism: a Swedish architect said he coined "new brutalism" in 1950; Alison Smithson used it in 1953; the critic Reyner Banham made it famous in 1955.
Glassmorphism became a popular label in 2020, for a glass look Windows Vista had in 2007.

In our reading, that is normal: the people making the work are rarely the ones who name the movement.

Why the label matters if you design

When you brief someone, search for references or buy a font, the label is the search term. "Victorian poster" and "Art Nouveau poster" bring back very different piles of images, and only one of them is likely to be what you had in mind. Three checks that help:

The date first. A date rules a label in or out faster than a description does: a design from the 1920s cannot be Art Nouveau at its height, and one from before 1837 cannot be Victorian.
Periods versus styles. "Victorian" names a reign; the style is the thing inside it, such as the Gothic Revival or Arts and Crafts.
Where the name comes from. Names that sound like a place or a feeling, such as Memphis or Brutalism, often mean something else.

Sources

Royal.uk, "Victoria (r. 1837-1901)".
Victoria and Albert Museum: "Stained glass: the Gothic Revival and beyond"; "Arts and Crafts: design for the home"; "Art Nouveau – an international style"; object record for a Siegfried Bing chair (O60313).
Tate art terms: "Art Nouveau", "Art Deco", "Bauhaus", "Brutalism", "Minimalism".
New York City Landmarks Preservation Commission, Chrysler Building designation report, LP-0992 (1978).
The Renaissance Society at the University of Chicago, "Art Deco" (exhibition essay).
Art Deco Society of Los Angeles, on Bevis Hillier and Art Deco of the 20s and 30s (1968).
SOS Brutalism, "Bengt Edman / Lennart Holm: Villa Göth, 1950".
Fondation Le Corbusier, "Unité d'habitation, Marseille, France, 1945-1952".
Irénée Scalbert, "Architecture as a Way of Life: The New Brutalism 1953-1956" (Team 10 Online, TU Delft).
The Architectural Review, Reyner Banham, "The New Brutalism" (December 1955), archive record.
Merriam-Webster, "mid-century modern" (first known use).
Dezeen, "Mid-century modern is never going away, says Cara Greenberg" (9 October 2024).
Design Museum, London, "Memphis"; Memphis Milano, "History".
Print magazine, "Swiss Style: The Principles, the Typefaces & the Designers" (2020).
The Brooklyn Rail, "50 years of Helvetica" (May 2007).
Henry Colley March, "The Meaning of Ornament; or its Archaeology and its Psychology", Transactions of the Lancashire and Cheshire Antiquarian Society, vol. VII, 1889 (printed 1890).
TIME, Lev Grossman, "Flatland" (3 June 2013); Apple Newsroom, "Apple Unveils iOS 7" (10 June 2013).
Interaction Design Foundation, "Glassmorphism"; Microsoft Learn archive, "Aero Glass inside a WPF window" (2006); Microsoft Source, Windows Vista consumer launch (29 January 2007).
Robert Browning, Andrea del Sarto, in Men and Women (1855), Project Gutenberg.

Checked September 2026.

Did They Really Say That? 22 Famous Quotes, Checked Against the Source

Hacks Vitae — Wed, 16 Sep 2026 00:00:00 GMT

A quotation is easy to get wrong and hard to take back. It travels as a screenshot, it rarely gets checked, and after enough repetitions the name and the sentence feel welded together.

So we checked twenty-two of them, not against quote-aggregator websites but, wherever we could, against the text each line is supposed to come from: the Meditations, the collected works, the film, the telegram, the novel.

Two hold up as quoted. Six are real but have drifted: wrong speaker, wrong wording, or a condition dropped. Three were written for the stage or the screen. Eleven belong to somebody else. Here is every one, with a verdict.

How we checked

Each quotation was traced to a primary source or a dated record: a passage number in the work, a dated speech or telegram, a named interview, or a published study of where a saying came from. Where nothing like that exists, the entry says so rather than guessing.

Quote-aggregator websites were not used as evidence. They often repeat a line with a famous name attached and no citation.

The verdicts:

TRUE: they said it, and we can point at where.
PARTLY TRUE: real, but the wording, the speaker or the meaning has drifted.
FILM LINE: a screenwriter wrote it. It is often good. It is not history.
FICTION: a playwright or novelist wrote it for a character.
MYTH: the credited person did not say it, as far as the record shows.

The ancients

"You have power over your mind — not outside events. Realize this, and you will find strength."

Credited to: Marcus Aurelius

MYTH. We could not find it in the Meditations. It is not in George Long's translation or in the older Casaubon translation on Project Gutenberg, and we found no book-and-section number attached to it anywhere we looked. It reads as a modern paraphrase.

What Marcus did write, in Meditations 8.47 (Long's translation), is the thought the paraphrase is squeezing: "If thou art pained by any external thing, it is not this thing that disturbs thee, but thy own judgment about it. And it is in thy power to wipe out this judgment now." A similar split opens Epictetus's Enchiridion: "There are things which are within our power, and there are things which are beyond our power."

"Live every day as if it were your last."

Credited to: Marcus Aurelius, usually as an argument for urgency

PARTLY TRUE. The idea is his, Meditations 7.69, but the sentence has conditions attached that rarely survive the screenshot. In Long's translation: "The perfection of moral character consists in this, in passing every day as the last, and in being neither violently excited nor torpid nor playing the hypocrite."

That is less a call to cram more in than a call for composure. The motivational version keeps the first half of a sentence whose second half pulls the other way.

"The good life is one inspired by love and guided by knowledge."

Credited to: Marcus Aurelius

MYTH. This is Bertrand Russell, in his 1925 book What I Believe, chapter "The Good Life".

"Fire tests gold; misfortune, brave men."

Credited to: Seneca

TRUE. De Providentia 5.9: ignis aurum probat, miseria fortes viros. It is a case where the short English version is close to the Latin, because the Latin was already that short.

"The greatest wealth is health."

Credited to: Virgil

MYTH. We found no Latin original. We searched the Eclogues, the Georgics and the Aeneid for the Latin words for health (valetudo, sanitas) and found none, and we found no ancient source that gives the line. It looks like a modern aphorism that picked up a Roman name.

"If the problem can be solved, why worry? If it cannot be solved, worrying will do no good."

Credited to: the Buddha

PARTLY TRUE. There is no evidence that the Buddha said it, but it is Buddhist and old. The argument is Shantideva's, Bodhicaryāvatāra 6.10, written by a Buddhist monk who lived between the late 7th and mid-8th century. The Sanskrit reads yadyastyeva pratīkāro daurmanasyena tatra kim / atha nāsti pratīkāro daurmanasyena tatra kim, which in our literal rendering is: "If there is a remedy, what is the use of dejection? And if there is no remedy, what is the use of dejection?"

Epictetus had divided the world in a similar way centuries earlier, in Enchiridion 1.

"I will either find a way or make one."

Credited to: Hannibal, at the Alps

MYTH. Neither Polybius nor Livy, who both describe the crossing of the Alps, has Hannibal say anything like it, in the English translations we searched.

The oldest version we found is in Seneca's tragedy Hercules Furens, lines 276–77, in the third person, spoken by Amphitryon about Hercules: inveniet viam aut faciet, "he will find a way, or make one." It is about Hercules returning from the underworld.

"With your shield, or on it."

Credited to: a Spartan mother sending her son to war

PARTLY TRUE. It is genuinely ancient, but it is not contemporary reporting. It comes from Plutarch's Sayings of Spartan Women, where one mother, handing her son his shield, says "Either this or upon this." Plutarch, from Chaeronea in Boeotia, lived around AD 45 to 120, centuries after the Sparta the sayings describe.

That does not make it false. It makes it a later collection of Spartan sayings, which is a different kind of source from an eyewitness, and worth knowing when the line is used as evidence of how Spartans behaved.

History

"Beware the Ides of March."

Credited to: Julius Caesar

FICTION. It is Shakespeare's, from Julius Caesar, Act I Scene 2, a play the Folger Shakespeare Library dates to 1599, and the Soothsayer says it to Caesar. Caesar's reply is "He is a dreamer; let us leave him. Pass."

There is a historical kernel. Plutarch writes that a seer warned Caesar of great peril on the Ides of March, and Suetonius names the soothsayer Spurinna. The four famous words are Shakespeare's.

"History is a set of lies agreed upon."

Credited to: Napoleon Bonaparte

PARTLY TRUE. Napoleon said something close, and did not claim it as his own. In the journal Las Cases kept on Saint Helena, for November 1816, Napoleon asks "What then is, generally speaking, the truth of history? A fable agreed upon," adding "as it has been very ingeniously remarked."

Quote Investigator traces the idea to Fontenelle's 1724 essay L'Origine des fables, which was about ancient history and uses fables, not lies. The version with "lies agreed upon" appears later; the orator Wendell Phillips used "series of lies agreed upon" in a speech printed in 1881.

"The unleashed power of the atom has changed everything save our modes of thinking."

Credited to: J. Robert Oppenheimer

MYTH. It is Albert Einstein's, from a telegram appeal for the Emergency Committee of Atomic Scientists, reported by the New York Times in May 1946. The full line ends "…and we thus drift toward unparalleled catastrophe."

In our reading, the misattribution is easy to understand: it is a line about the bomb, so it drifts to the name most people connect with the bomb.

"Now I am become Death, the destroyer of worlds."

Credited to: J. Robert Oppenheimer

TRUE. He said it on camera in NBC's 1965 documentary The Decision to Drop the Bomb, recalling the first nuclear test on 16 July 1945 and quoting the Bhagavad Gita: "I remembered the line from the Hindu scripture the Bhagavad Gita."

What is on record is the 1965 recollection. Whether the line went through his mind at the test itself rests on his word; others present did not recall him saying it there.

"Now we are all sons of bitches."

Credited to: Oppenheimer, at Trinity

MYTH. Kenneth Bainbridge, the physicist who directed the Trinity test, said it to Oppenheimer right after the detonation. In his own 1975 account, Bainbridge wrote that Oppenheimer told his younger daughter in 1966 "that it was the best thing anyone said after the test."

"Be the change you wish to see in the world."

Credited to: Mahatma Gandhi

MYTH. Writing in the New York Times in August 2011, Brian Morton found "no reliable documentary evidence" for the quotation. Quote Investigator's earliest close match is from 1974, by the educator Arleen Lorrance; Gandhi died in 1948. Gandhi's family has said he often said it, but no one has found it in his writings.

The closest real passage is in Gandhi's Indian Opinion article of 9 August 1913, printed in volume XII of his Collected Works: "If we could change ourselves, the tendencies in the world would also change. As a man changes his own nature, so does the attitude of the world change towards him… We need not wait to see what others do."

The short version loses something. Gandhi is describing how the world responds to a changed person. The slogan turns it into an instruction.

"Success is not final, failure is not fatal: it is the courage to continue that counts."

Credited to: Winston Churchill

MYTH. The International Churchill Society lists it among quotes falsely attributed to him: "we can find no attribution" for it.

What Churchill did say, at Harrow School on 29 October 1941, was: "Never give in, never give in, never, never, never — in nothing, great or small, large or petty — never give in except to convictions of honour and good sense."

The screen

"It is a kingdom of conscience, or nothing."

Credited to: a crusader king

FILM LINE. Kingdom of Heaven (2005), written by William Monahan. Balian says it, when a knight tells him Jerusalem has no need of a perfect knight. Earlier in the film, Godfrey of Ibelin introduces the phrase: "A kingdom of conscience. A kingdom of heaven." It is not Baldwin IV's; his separate speech, "a king may move a man", was also written for the film.

The sentence is not in William of Tyre's history of the kingdom.

"It's not personal, it's just business."

Credited to: Don Vito Corleone

FILM LINE, and misquoted. The line is Michael's, not Vito's, and it runs: "It's not personal, Sonny. It's strictly business." He says it while arguing that the family can kill a crooked police captain, which is the edge the paraphrase loses: a young man persuading his family that murder is a business decision.

"Revenge is a dish best served cold."

Credited to: The Godfather

PARTLY TRUE. It is in Mario Puzo's 1969 novel, where Don Corleone says: "Revenge is a dish that tastes best when it is cold." We did not find it in the dialogue of the 1972 film.

The saying is much older than either. The earliest example Quote Investigator found is Eugène Sue's 1841 French novel Mathilde, where the speaker calls it something people commonly say; in English it appears by 1846, in a translation of the same novel: "revenge is very good eaten cold, as the vulgar say."

"Arise, arise, Riders of Théoden!"

Credited to: Tolkien

PARTLY TRUE. The opening cry is Tolkien's, but the speech people usually quote after it is Peter Jackson's film version, adapted from Tolkien's: "Spears shall be shaken, shields shall be splintered! A sword-day! A red day, ere the sun rises!", followed by "Ride now, ride now, ride! Ride for ruin and the world's ending!"

In The Return of the King (Book V, ch. 5), Tolkien has "Fell deeds awake: fire and slaughter!", then "spear shall be shaken, shield be splintered" in the singular, the same "a sword-day, a red day, ere the sun rises!", and the charge line "Ride now, ride now! Ride to Gondor!" The film's "Ride for ruin" adapts Éomer's cry later in the battle, "ride to ruin and the world's ending". Our page on the Ride of the Rohirrim sets the two side by side.

Modern wisdom

"The best way to predict the future is to invent it."

Credited to: Abraham Lincoln

MYTH. Quote Investigator found no substantive evidence that Lincoln used it. Alan Kay has said he coined it in 1971, at an early meeting between Xerox PARC researchers and Xerox planners: "In a fit of passion I uttered the quote!", as he put it in a 1998 email. By 1982 it was being reported as the maxim of the researchers at Xerox PARC.

The idea is older than Kay's wording: in 1963 the physicist Dennis Gabor wrote "The future cannot be predicted, but futures can be invented."

"Be so good they can't ignore you."

Credited to: Warren Buffett

MYTH. It is Steve Martin's answer to people who ask him how to make it in entertainment, as Cal Newport described it in 2008. Newport later used it as the title of his 2012 book So Good They Can't Ignore You.

It is advice about becoming excellent at something. Attaching it to an investor quietly changes what it is about.

"The only limit to our realization of tomorrow will be our doubts of today."

Credited to: Napoleon Hill, among others

MYTH as Napoleon Hill. It is Franklin D. Roosevelt's, word for word, from the address he prepared for Jefferson Day dinners on 13 April 1945. The next line is "Let us move forward with strong and active faith." He worked on the speech on 11 April and died on 12 April 1945, so it was never delivered.

What the misses have in common

Twenty-two quotations, and in our reading the misses fall into four shapes. Once you know them, they are easier to spot.

The name gets upgraded. A less famous author is replaced by a famous one in the same territory. Einstein's line about the atom drifts to Oppenheimer. Steve Martin's line drifts to a billionaire. Alan Kay's line drifts to Lincoln. The sentence survives; the credit moves up.
The paraphrase eats the original. Someone summarises a passage well, the summary is shorter and more quotable, and in time the summary is shared as the quotation. That is the story of the Marcus Aurelius line at the top of this page.
Fiction gets promoted to history. A writer gives a historical character the perfect line, and it is soon quoted as something the character said. Balian, Théoden, the Soothsayer, Michael Corleone.
The condition gets dropped. The quotation is real and the qualifier is gone. Marcus says to pass every day as the last "neither violently excited nor torpid nor playing the hypocrite"; the popular version keeps the urgency and drops the rest.

How to check one yourself, in about two minutes

Ask for the passage. A real quotation from a book has a chapter, a section or a letter number. If nobody can produce one, that tells you something.
Search the text, not the web. Many of the works quoted here are out of copyright and searchable on Project Gutenberg, Wikisource or the Latin Library. Searching the book itself takes seconds.
Treat quote sites as leads, not evidence. Ten of them agreeing can be one unsourced claim copied ten times.
Check the lines you like best first. A quotation that fits your argument perfectly may fit because somebody polished it.
Check whether the person could have said it. Dates, language and subject matter all help.

Sources

Marcus Aurelius, Meditations, 7.69 and 8.47, trans. George Long (Project Gutenberg #15877); Casaubon translation (Project Gutenberg #2680), searched.
Epictetus, Enchiridion, 1, trans. T. W. Higginson (Project Gutenberg #45109).
Bertrand Russell, What I Believe (1925), Project Gutenberg #73782.
Seneca, De Providentia 5.9, Latin text (The Latin Library); Hercules Furens 276–77, Dickinson College Commentaries.
Virgil, Eclogues, Georgics and Aeneid, Latin text (The Latin Library), searched.
Śāntideva, Bodhicaryāvatāra 6.10, Sanskrit text (GRETIL); Monier-Williams Sanskrit-English Dictionary (Cologne Digital Sanskrit Dictionaries); Stanford Encyclopedia of Philosophy, "Śāntideva".
Polybius, Histories, trans. E. S. Shuckburgh (Project Gutenberg #44125); Livy, History of Rome, trans. Spillan and Edmonds (Project Gutenberg #10907), searched.
Plutarch, Sayings of Spartan Women, trans. F. C. Babbitt (LacusCurtius); Plutarch, Caesar 63 (LacusCurtius); Suetonius, Julius 81 (LacusCurtius); Stanford Encyclopedia of Philosophy, "Plutarch".
William Shakespeare, Julius Caesar, I.ii (Project Gutenberg #1522); Folger Shakespeare Library.
Quote Investigator: "What Is History But a Fable Agreed Upon?"; "Be the Change You Wish To See in the World"; "Revenge Is a Dish Best Eaten Cold"; "We Cannot Predict the Future, But We Can Invent It".
Bulletin of the Atomic Scientists: Robert R. Holt, "Meeting Einstein's challenge" (2015); "Trinity, now and then" (2015), quoting Kenneth Bainbridge's 1975 account; National Park Service, "Kenneth Bainbridge".
Contingent Magazine, "The Destroyer" (2023), transcribing NBC, The Decision to Drop the Bomb (1965).
Gandhi, "General Knowledge About Health", Indian Opinion, 9 August 1913, Collected Works vol. XII.
International Churchill Society, "Quotes Falsely Attributed" and "Never Give In" (Harrow, 1941).
Kingdom of Heaven (2005) and The Godfather (1972), scene transcripts on Clip.Cafe; The Godfather subtitle transcript (Subs like Script).
J. R. R. Tolkien, The Return of the King, Book V, ch. 5 (Internet Archive, search-inside).
The Return of the King (2003), film transcript (tk421.net); film page on Clip.Cafe.
Bodhipaksa, "If the problem can be solved why worry?", Fake Buddha Quotes (2018).
Cal Newport, "The Steve Martin Method" (2008) and author page.
Franklin D. Roosevelt, "Undelivered Address Prepared for Jefferson Day" (American Presidency Project); FDR Library, "FDR: Day by Day" (April 1945) and FDR biography.

Checked September 2026. Where a line could not be traced to a primary source, this page says that rather than offering a guess. If you can produce a citation for anything marked here as a myth (a book and page, a dated speech, a letter), we want to see it: hacksvitae@gmail.com.

Did They Really Say That? Part 2: Einstein, Roosevelt, Ford and the Films

Hacks Vitae — Wed, 16 Sep 2026 00:00:00 GMT

The first part of this series checked twenty-two quotations, from ancient writers to modern films. This one is mostly modern: scientists, presidents, businessmen and films, plus two lines of scripture.

Six of these fifteen lines belong to somebody else. Two are genuine. Six are real but have been trimmed, reordered or misquoted, including three well-known lines from film and fiction. And one is an origin story that has no source. Each line has its own verdict and source in the table; the sections below are about the patterns.

The name gets upgraded

The pattern from part one shows up again. A line starts somewhere less famous (a recovery-group meeting, a newspaper column, an online post, a journal article) and after years of repetition it picks up a famous name.

On this page, lines about how to think drift to Einstein, lines about courage and about women making history drift to Eleanor Roosevelt, a witty line about being yourself drifts to Oscar Wilde, and a line about innovation drifts to Henry Ford. An article on the Henry Ford museum's website says he "has become a patron saint of the entrepreneur, a Paul Bunyan of the business world." In two of these cases, the person who really wrote the line can still be named.

The Mary Schmich column is a clear example. Do one thing every day that scares you appeared in her Chicago Tribune column of 1 June 1997. By August 1997 the column was circulating without her name, retitled "Wear Sunscreen" and wrongly described as a graduation speech by Kurt Vonnegut. In 1999 it became the spoken-word song "Everybody's Free (To Wear Sunscreen)", which credited her. The line was later attributed to Eleanor Roosevelt, and in 2012 Schmich wrote that she had seen no evidence Roosevelt said it.

The film lines

The three screen and fiction lines follow a different pattern: the misquote carries its own context.

No, I am your father only makes sense after Luke's line, and the shared version puts Luke's name where the "No" was. Play it, Sam needs the song named. Elementary and my dear Watson are both Holmes's words in Conan Doyle, just never in the same breath. In our reading, each misquote works on its own, which makes it easier to share, and in time it is the version people remember.

The quiet edits

The scriptural and political lines were not misattributed. They were trimmed, and the trim changed them.

The love of money is the root of all evil becomes money is the root of all evil, which is a different claim. Blood, toil, tears and sweat loses the toil. Pride goeth before destruction, and an haughty spirit before a fall is squeezed into one clause. In each case the missing words are the ones that made the sentence longer, and in the verse from Timothy they are the words that carried the meaning.

How to check one yourself

If it sounds exactly like the famous person, it is worth a second look. Several lines on this page ended up with a famous name that seemed to fit.
Look for the earliest version, and its date. A line first found in print in 2004 is hard to pin on someone who died in 1955.
For a film or book line, check the script or the text. Many are easy to search, and they usually settle it in a minute.
For scripture, read the whole verse. The shared version is often missing part of it.
Check the story about the saying, too. An origin story can need a source as much as the quotation does.

Sources

Quote Investigator: "Insanity Is Doing the Same Thing Over and Over Again and Expecting Different Results"; "Everybody is a Genius…"; "Do One Thing Every Day That Scares You"; "Be Yourself. Everyone Else Is Already Taken"; "My Customers Would Have Asked For a Faster Horse"; "Genius Is One Percent Inspiration, Ninety-Nine Percent Perspiration"; "Elementary, My Dear Watson".
The Henry Ford, "Henry Ford's Quotations" (Benson Ford Research Center, 2013); Snopes, "No proof Henry Ford said…" (2025); Henry Ford, My Life and Work (Project Gutenberg #7213), searched.
Laurel Thatcher Ulrich, "Why Well-Behaved Women Seldom Make History", BYU Studies 59:3, quoting her 1976 American Quarterly article; Penguin Random House, Well-Behaved Women Seldom Make History.
Franklin D. Roosevelt, Inaugural Address, 4 March 1933 (American Presidency Project).
International Churchill Society, "Blood, Toil, Tears and Sweat" (13 May 1940); Richard M. Langworth, "Blood, Toil, Tears and Sweat: Origins of a Famous Phrase", Hillsdale College Churchill Project (2018).
Arthur Conan Doyle's four Sherlock Holmes novels and five story collections (Project Gutenberg #244, #2097, #1661, #834, #2852, #108, #3289, #2350, #69700), searched.
Casablanca (1942): AFI's 100 Years…100 Movie Quotes; Wikiquote, Casablanca; The Phrase Finder, "Play it again, Sam"; Wikipedia, Play It Again, Sam (play).
The Empire Strikes Back (1980): ACMI, "Star Wars: Episode V The Empire Strikes Back – 'No, I am your father.'" (2020); a script copy online (IMSDb).
1 Timothy 6:10 and Proverbs 16:18, King James Version (Project Gutenberg #10).
Wikipedia, "Blood is thicker than water", for William Jenkyn's sermon, Reinhart Fuchs and the 1990s–2000s origin claim; Walter Scott, Guy Mannering (Project Gutenberg #5999).

Checked September 2026. The verdicts on the six misattributed lines rest on Quote Investigator, The Henry Ford and Laurel Thatcher Ulrich's own account; the dated speeches on the American Presidency Project and the International Churchill Society; the Holmes verdict on our search of the Gutenberg texts and Quote Investigator; the film lines on AFI, ACMI, Wikiquote, The Phrase Finder and a script copy online; the verses on the King James text; and the proverb on Wikipedia's sourced history and Scott's novel. If you can date any of these earlier, with a source: hacksvitae@gmail.com

Older Than You Think: 12 Modern Life Rules and Where They Actually Come From

Hacks Vitae — Wed, 16 Sep 2026 00:00:00 GMT

Scroll long enough and the same advice comes round again in a new font. Sit with discomfort. Cut the noise. Remember you are going to die. It reads as though somebody worked it out last year.

Often the advice is a rerun, and the original is sharper than the copy. The reverse happens too, and it is the more interesting case: a modern idea gets dressed in an ancient costume, and the label travels with it.

Twelve rules, traced to where and when they were written. Nine are older than they look. Three are a great deal newer.

The verdicts here are about age and origin rather than wording:

TRUE — the idea is old, and the source says roughly what the modern version says.
PARTLY TRUE — the idea is old, but the famous sentence is not, or has lost part of what it said.
MYTH — sold as ancient, or credited to the wrong person, and the source does not bear it out.

Part one: older than you think

"The obstacle is the way."

PARTLY TRUE. The thought is Marcus Aurelius, Meditations 5.20. The sentence is modern: its punch comes from a 2002 translation, and the three-word slogan is shorter still.

Gregory Hays's translation, published by the Modern Library in 2002, gives the end of the passage as "The impediment to action advances action. What stands in the way becomes the way." That is Hays's English. George Long's translation of the same passage, first published in 1862, ends: "and that which is an obstacle on the road helps us on this road." Same idea, far less quotable.

Marcus wrote in Greek. When a line from him sounds like a slogan, the slogan is often the translator's.

Journaling

TRUE. The Meditations is, in the words of the Internet Encyclopedia of Philosophy, "a personal notebook, written by Marcus to himself and for his own use". The same encyclopedia says it was probably written while he was on campaign in central Europe, around AD 171–175.

The Greek title it goes by, ta eis heauton, means "to himself". The encyclopedia adds that this title may be a later addition; it is first recorded around 900, by the bishop Arethas.

So a book people now read as practical philosophy began as a man writing notes to himself. That is a reassuring thing to know if your own journal repeats itself.

"Curate your feed. Your environment shapes you."

TRUE. Seneca, Letters to Lucilius 7, "On crowds", in Richard Mott Gummere's translation:

"To consort with the crowd is harmful; there is no person who does not make some vice attractive to us, or stamp it upon us, or taint us unconsciously therewith. Certainly, the greater the mob with which we mingle, the greater the danger. But nothing is so damaging to good character as the habit of lounging at the games; for then it is that vice steals subtly upon one through the avenue of pleasure."

He goes on to say that he comes home "more greedy, more ambitious, more voluptuous, and even more cruel and inhuman", and describes a midday show at the arena. The mechanism he names, that the harm arrives through pleasure, which is why you do not notice it, is the same argument people make today for a digital detox.

"You don't have a time problem, you have a priorities problem."

TRUE. Seneca, On the Shortness of Life, chapter 1, in John W. Basore's translation: "It is not that we have a short space of time, but that we waste much of it. Life is long enough".

The essay is addressed to Paulinus, who was in charge of Rome's grain supply, and much of it is about people who are busy with everything except their own lives: "There is nothing the busy man is less busied with than living".

"Rock bottom built me."

PARTLY TRUE. Seneca, De Providentia 5.9: ignis aurum probat, miseria fortes viros, "fire tests gold; misfortune, brave men."

One thing the modern version drops. Our reading: in the Latin, the fire tests the gold; it does not make it. Seneca is talking about what hardship shows about a character, and he says the good man must steer his own course through it. He does not promise that any disaster will improve you.

"If you can fix it, don't worry. If you can't, worrying won't help."

TRUE. It is Shantideva, Bodhicaryāvatāra 6.10, by a Buddhist monk who lived between the late 7th and mid-8th century. In the Padmakara Translation Group's English (2006):

"If there's a remedy when trouble strikes, / What reason is there for dejection? / And if there is no help for it, / What use is there in being glum?"

The argument in the original is the one people repeat. We follow the names this line gets shared under in Did They Really Say That?

"Memento mori."

PARTLY TRUE. A few ancient writers describe the reminder, but the famous two-word tag is not in the passage usually cited for it, and the historian Mary Beard, in The Roman Triumph (2007), questions how good the evidence for the custom is.

The passage usually cited is Tertullian's Apologeticus 33. He is arguing that the emperor is a man, not a god, and says that even in a triumph, riding in the chariot, the emperor is prompted from behind: "Respice post te! Hominem te memento!" In S. Thelwall's translation: "Look behind thee; remember thou art but a man." Tertullian does not say who speaks.

He is not the only writer to mention the reminder. Epictetus, in the Discourses recorded by his pupil Arrian, refers to "those who stand behind men in their triumphs and remind them that they are mortal". And Zonaras, in a passage printed among the fragments of Cassius Dio's Roman History, says a public slave rode in the chariot holding the crown and "kept saying to him, 'Look behind!'"

The two-word tag memento mori is not in Tertullian's Latin: his words are hominem te memento, "remember you are a man".

The Stoic habit of keeping death in view is well documented in its own right. Epictetus, Enchiridion 21: "Let death and exile, and all other things which appear terrible, be daily before your eyes, but death chiefly".

"Comparison is the thief of joy."

MYTH as Theodore Roosevelt, but the point is ancient. Quote Investigator "has located no substantive evidence supporting ascriptions to Theodore Roosevelt, Mark Twain, and C. S. Lewis". The earliest match it found using "comparison" and "thief of joy" is from 2003, and in 2004 the familiar wording was in print credited to nobody in particular ("Somebody once said…"). The names came afterwards: Dwight Edwards in 2010, C. S. Lewis in 2011, Roosevelt in 2012, and a variant pinned on Mark Twain in 2013. Roosevelt died in 1919.

The underlying thought is old. In Letters 2, Seneca quotes Epicurus, "Contented poverty is an honourable estate", and then adds his own gloss: "It is not the man who has too little, but the man who craves more, that is poor." A different angle on the same problem: the trouble lies in the wanting, not in what you have.

"Live every day as if it were your last."

PARTLY TRUE. Marcus Aurelius, Meditations 7.69, and the sentence has a condition attached that the motivational version leaves out. In Long's translation: "The perfection of moral character consists in this, in passing every day as the last, and in being neither violently excited nor torpid nor playing the hypocrite."

The condition points toward calm rather than urgency. The half that gets quoted can read as a case for frantic living; the half that gets left out rules that out.

Part two: newer than you think

"We are what we repeatedly do. Excellence, then, is not an act, but a habit."

MYTH as Aristotle. It is Will Durant, The Story of Philosophy, 1926: a popular history of philosophy, summarising Aristotle.

It is a good summary. In the Nicomachean Ethics, Book II, Aristotle argues that "by doing just actions we come to be just; by doing the actions of self-mastery we come to be perfected in self-mastery; and by doing brave actions brave" (Everyman translation, on Project Gutenberg).

So the thought is Aristotle's and the famous sentence is Durant's.

The ikigai Venn diagram

MYTH. The four-circle diagram (what you love, what you are good at, what the world needs, what you can be paid for) is not Japanese and not ancient. It was published on 14 May 2014 by Marc Winn, a blogger in Guernsey.

In a note added to that post he writes: "The diagram is not mine. It began as a purpose venn diagram by the Spanish astrologer Andrés Zuzunaga". Zuzunaga, commenting on the same post, dated his Spanish graphic to 2012. Winn says he had come across the word ikigai in Dan Buettner's work on the Blue Zones.

Winn has since written about the gap himself: "The Japanese idea of ikigai is not a career-planning tool. It doesn't require your work, your gift and the world's needs to line up in one paid role." In Japan, he adds, it is often "something small and domestic, a morning walk, a grandchild, a garden". A Japanese dictionary, the Digital Daijisen, defines the word as 生きるに値するもの, "something worth living for".

"Amor fati" as a Stoic phrase

MYTH. The phrase amor fati is Nietzsche's. The phrase appears in The Gay Science, section 276 (1882). In Thomas Common's translation: "I want more and more to perceive the necessary characters in things as the beautiful:—I shall thus be one of those who beautify things. Amor fati: let that henceforth be my love!" He returns to it in Ecce Homo, written in late 1888 (Anthony M. Ludovici's translation): "My formula for greatness in man is amor fati: the fact that a man wishes nothing to be different, either in front of him or behind him, or for all eternity."

The Stoics did teach accepting what you cannot control, and Marcus Aurelius went further: "Love that only which happens to thee and is spun with the thread of thy destiny" (Meditations 7.57, Long). So the mix-up is understandable. Epictetus, Enchiridion 8: "Demand not that events should happen as you wish; but wish them to happen as they do happen, and you will go on well."

Nietzsche made the phrase a formula of his own. In the same Ecce Homo passage: "Not only must the necessary be borne, and on no account concealed,—all idealism is falsehood in the face of necessity,—but it must also be loved." And he criticised the Stoics directly, in Beyond Good and Evil §9: "You desire to LIVE 'according to Nature'? Oh, you noble Stoics, what fraud of words!"

Why any of this matters

There is a practical reason to know how old a piece of advice is.

An idea that is two thousand years old has had a long testing. Generation after generation could have dropped it and did not. That is not proof it is true, but it is evidence of a kind, the same kind you get from a building that is still standing.

An idea from 2014 has had less time. It may be excellent, and the ikigai diagram is a reasonable career exercise. It is one person's adaptation, though, and it helps to weigh it as that rather than as the accumulated wisdom of Okinawa.

The advice is rarely the problem. The costume is.

How to date one yourself

Search the exact phrase and sort by oldest. Google Books is the quickest tool here. If the earliest hit is recent, the Roman label is doubtful.
Search the ancient work itself. The Meditations, Seneca's letters and essays, the Nicomachean Ethics and most of the other old texts on this page are free to read on Project Gutenberg and Wikisource, in older translations. If the line is there, it takes seconds to find. If it is not, that tells you something too.
Check which translation. A punchy English sentence from an ancient text is often the translator's punch, not the author's. Compare two translations and see what survives both.
Be careful with diagrams. A neat chart with an ancient name on it is worth tracing like any quotation.

And the last word belongs to Ecclesiastes 1:9: "there is no new thing under the sun." The observation that nothing is new is not new either.

Sources

Marcus Aurelius, Meditations 5.20, 7.57 and 7.69, trans. George Long (first edition 1862; Project Gutenberg and Wikisource); Gregory Hays's translation (Modern Library, 2002), as quoted and identified in Donald Robertson's review of The Obstacle Is the Way (2017).
Internet Encyclopedia of Philosophy, "Marcus Aurelius".
Seneca, Letters to Lucilius 2 and 7, trans. Richard Mott Gummere (Wikisource); On the Shortness of Life 1, 7 and 18, trans. John W. Basore (Wikisource), with the Latin from The Latin Library; De Providentia 5.9 (The Latin Library).
Śāntideva, Bodhicaryāvatāra 6.10, in the Padmakara translation (2006), via the Tsadra Foundation's Bodhicitta verse pages; dates from the Stanford Encyclopedia of Philosophy, "Śāntideva".
Tertullian, Apologeticus 33 (Latin, The Latin Library; trans. S. Thelwall, tertullian.org); Epictetus, Discourses 3.24, trans. George Long (Wikisource); Cassius Dio, Roman History, Book 6 fragments with Zonaras's summary, Loeb translation (LacusCurtius); Epictetus, Enchiridion 8 and 21, trans. T. W. Higginson (Project Gutenberg).
Quote Investigator, "Comparison Is the Thief of Joy" (2021).
Will Durant, The Story of Philosophy (1926), Aristotle chapter (Internet Archive); Aristotle, Nicomachean Ethics II.1 (Everyman translation, Project Gutenberg).
Marc Winn, "What is your Ikigai?" (14 May 2014, with a later note and reader comments) and his Ikigai pillar page, The View Inside Me; Digital Daijisen, 生き甲斐, via Kotobank.
Mary Beard, The Roman Triumph (2007), as summarised in Wolfgang Blösel's review, sehepunkte 8 (2008), no. 10.
Friedrich Nietzsche, The Gay Science §276, trans. Thomas Common; Ecce Homo, "Why I Am So Clever" §10, trans. Anthony M. Ludovici; Beyond Good and Evil §9, trans. Helen Zimmern (all Project Gutenberg); Stanford Encyclopedia of Philosophy, "Friedrich Nietzsche's Life and Works", for dates.
Ecclesiastes 1:9, King James Version (Project Gutenberg).

Checked September 2026. If you can date any of these earlier than we have, with a source, we would like to see it: hacksvitae@gmail.com

Older Than You Think, Part 2: Productivity Rules and Where They Come From

Hacks Vitae — Wed, 16 Sep 2026 00:00:00 GMT

Part one traced life advice back to where it was written. This one does the same for work: the rules, razors and laws that fill productivity posts and management slides.

The pattern is different here. Life advice tends to be older than it looks. Productivity advice tends to be older in its idea and newer in its name, and the name often carries a claim the original never made.

When a rule gets promoted to a law

Three of these are worth reading together: the 80/20 rule, Parkinson's law and Hanlon's razor.

Pareto studied how unequally income was spread and fitted a curve to it. The quality-control engineer Joseph Juran later gave Pareto's name to a general principle, the "vital few and trivial many", and in 1974 wrote: "I was forced to confess that I had mistakenly applied the wrong name to the principle."
C. Northcote Parkinson, a historian and satirist, opened a 1955 essay about the growth of government staff with the line people now quote. He called it "a commonplace observation"; his law was about the number of officials.
Robert Heinlein gave a character in a 1941 story a line about villainy and stupidity. In 1980 a similar line was printed as "Hanlon's razor", named after the man who sent it in.

In each case a shrewd observation picked up the authority of a law, largely by being given a formal-sounding name. Our reading: they work well as prompts to look again, and less well as laws. A split you meet at work may not be anywhere near 80/20, and sometimes a problem really is somebody's bad faith.

The chart that was never in the study

The Dunning-Kruger curve is a case of a picture standing in for a finding.

The 1999 study, by Justin Kruger and David Dunning, tested people on humour, grammar and logic. Those who scored in the bottom quarter "grossly overestimated their test performance and ability": their scores put them in the 12th percentile, and they placed themselves in the 62nd. The paper's figures plot each quarter's perceived ability against its actual test score. It has no peak of overconfidence, no valley of despair and no slope of enlightenment.

How big the effect is remains argued over. Kruger and Dunning wrote that they believed "the overestimation we observed was more psychological than artifactual". A 2020 study by Gignac and Zajenkowski, using different statistical tests on 929 people's self-rated intelligence, concluded that "the magnitude of the effect may be much smaller than reported previously". A 2021 study by Jansen, Rafferty and Griffiths, a large-scale replication with about 4,000 people in each of two studies, found support for low performers being less able to judge whether their answers were right, in grammar and logical reasoning. The studies disagree about method, so we leave the size of the effect open.

The oldest one is the most practical

The delegation passage in Exodus 18 is a few verses long. A leader settles every dispute himself "from the morning unto the evening". His father-in-law, Jethro, tells him: "Thou wilt surely wear away, both thou, and this people that is with thee". The fix is a structure: "rulers of thousands, and rulers of hundreds, rulers of fifties, and rulers of tens", with "the hard causes" brought to Moses and the small matters judged by the others. It reads like a consultant's report on an overloaded manager.

Using these rules without leaning on them too hard

A few ways to get the useful part of each rule while keeping its origin in mind:

80/20: treat it as a question, not a fact. Which few things account for most of the result here, and what is the split really?
Parkinson's law: the sentence is a joke with a point. A deadline can stop a task spreading, and it is worth remembering that Parkinson's own subject was offices adding staff.
Hanlon's razor: before assuming someone meant harm, it is worth asking whether a misunderstanding or a mistake would explain it, which is Goethe's point too. It is a first question, not a verdict.
Delegation: Jethro's version is concrete. Decide which kinds of question the leader keeps, and who handles the rest.
Impostor feelings: the 1978 paper describes a pattern, not an illness. A 2020 review found the feelings often come alongside depression and anxiety, and those are matters for a doctor or counsellor rather than for a productivity rule.

Sources

Robert Browning, "Andrea del Sarto", Men and Women (1855), Project Gutenberg; Molly Driscoll, "Ludwig Mies van der Rohe: Father of 'less is more' architecture", Christian Science Monitor (27 March 2012).
Exodus 18, King James Version (Project Gutenberg).
Ben S. Graham, "Allan Mogensen and his Legacy" (2008), on the Work Simplification Conferences.
W. Carew Hazlitt, English Proverbs and Proverbial Phrases (1907), with its key to John Clarke's Paroemiologia Anglo-Latina (1639); Benjamin Franklin, The Way to Wealth (Project Gutenberg).
Quote Investigator, "Time is Money" (2010), citing Franklin's Advice to a Young Tradesman (1748) and The Free-Thinker (18 May 1719); Plutarch, Life of Antony 28, Loeb translation (LacusCurtius).
"Parkinson's Law", The Economist, 19 November 1955 (Internet Archive copy); "Parkinson, C. Northcote", Contemporary Authors entry via Encyclopedia.com.
Stanford Encyclopedia of Philosophy, "William of Ockham"; W. M. Thorburn, "The Myth of Occam's Razor", Mind 27 (1918), Wikisource.
Vilfredo Pareto, Cours d'économie politique (1896–97), Internet Archive scans; Joseph M. Juran, "The Non-Pareto Principle; Mea Culpa" (1974), Juran Institute archive.
Quote Investigator, "Never Attribute to Malice That Which Is Adequately Explained by Stupidity" (2016); Robert Heinlein, "Logic of Empire", Astounding Science-Fiction, March 1941 (Internet Archive); Goethe, The Sorrows of Young Werther, trans. R. D. Boylan (Project Gutenberg).
Pauline Rose Clance and Suzanne Imes, "The Imposter Phenomenon in High Achieving Women: Dynamics and Therapeutic Intervention" (1978); Bravata and others, "Prevalence, Predictors, and Treatment of Impostor Syndrome: a Systematic Review", Journal of General Internal Medicine (2020).
Justin Kruger and David Dunning, "Unskilled and Unaware of It", Journal of Personality and Social Psychology 77(6) (1999); Gignac and Zajenkowski, "The Dunning-Kruger effect is (mostly) a statistical artefact", Intelligence (2020); Rachel A. Jansen, Anna N. Rafferty and Thomas L. Griffiths, "A rational model of the Dunning-Kruger effect supports insensitivity to evidence in low performers", Nature Human Behaviour 5 (2021).

Checked September 2026. If you can date any of these earlier, with a source: hacksvitae@gmail.com

Real or Myth? 14 Popular Claims, Checked

Hacks Vitae — Wed, 16 Sep 2026 00:00:00 GMT

Some facts get repeated so often that they feel checked. They arrive in a numbered carousel, they sound like the sort of thing that would be true, and they can go round for years before anyone asks where they came from.

We took fourteen of them and looked for the earliest and most reliable source we could find for each.

Nine are myths. Three are partly true. One cannot be settled either way. One holds up. The partly-true ones are the most interesting, because a half-right claim lasts much longer than a wrong one: there is always something real underneath to point at.

How to read the sheet

Every row opens. Inside is what the sources say and where they come from.

True: it holds up.
Partly true: something real is in there, but the popular version has added or changed something that matters.
Unverified: no record we could find settles it either way.
Myth: it does not hold up.

The one that is really about all the others

The spider claim is worth singling out, because it is the whole problem in miniature.

The claim itself has nothing behind it. Rod Crawford, the arachnid curator at the Burke Museum in Seattle, told Scientific American that "a sleeping person is not something a spider would willingly approach". Swallowing one is possible, but, as the same article puts it, "it would be a strictly random event".

Then there is the correction you may have heard: that the "eight spiders a year" statistic was invented in 1993 by a columnist named Lisa Birgit Holst, writing in a magazine called PC Professional, to show that people will repeat anything they read.

That story is also made up. The fact-checking site that first told it now says that "PC Professional" and the columnist "do not exist", that her name "is an anagram of the phrase 'this is a big troll'", and that the story was planted in April 2001 as a joke for its long-time readers. The spider claim was already going round before that: the same page quotes a 1999 trivia list that has a version of it.

So a made-up fact picked up a made-up origin story about made-up facts. The second story is satisfying to pass on, because telling it casts the teller as the one who knew better.

It is easy to relax once a claim seems to have been seen through; that is when the second story slips in.

What the partly-true ones have in common

Look at the three together and the same shape appears each time.

Some sharks do die if they stop swimming. What was added is the word sharks, with no qualifier.
Bison do face into storms. What was added is the reason: that they know it gets them through faster.
Milk is usually at the back of the shop. The intent behind that is argued both ways, even by people who work in the trade.

In each case the core is a fact, and the addition is the part nobody has shown. The addition is what makes it shareable, which may be why it is the part that gets added.

The Napoleon line has a related shape. The idea has been credited to him since at least 1827, but the earliest book we found that does so was written in his voice by someone else, and the snappy wording came decades later.

How to check one yourself

Ask what the original source would even look like. "Goldfish remember for three seconds" would need someone to have tested goldfish memory and found three seconds. The tests that have been done found memories lasting months.
Search for the earliest version, not the most popular one. Sort by date. If the oldest version you can find is only a few years old, it is not ancient wisdom.
Take a second look at claims that come with a moral. A fact that also happens to be good advice has had a reason to be improved.
Check the correction as carefully as the claim. The spider story is the reason.

Sources

Each row above names its own source. In full:

Joseph Castro, "Must Sharks Keep Swimming to Stay Alive?", Live Science, 29 May 2013.
Annie Sneed, "Fact or Fiction? People Swallow 8 Spiders a Year While They Sleep", Scientific American, 15 April 2014 (quoting Rod Crawford, Burke Museum); Burke Museum, "Myth: You swallow spiders in your sleep".
Snopes, "Do People Swallow Eight Spiders Per Year?" and "Lisa Birgit Holst", on the invented origin story.
China Manned Space, "Yang Liwei" (official profile).
Snopes, "Did Napoleon say, 'Never interrupt your enemy while he's making a mistake'?" (31 March 2026), which rates the attribution correct.
People's Daily, "'I did not see Great Wall from space': Yang Liwei", 17 October 2003; NASA, "China's Wall Less Great in View from Space", 9 May 2005; Phil Plait, "No, You Can't See the Great Wall of China from Space", Scientific American, 9 August 2024.
Walter Isaacson, "Making the Grade", TIME, 5 April 2007; Karl Kruszelnicki, "Einstein Failed School", ABC Science, 23 June 2004; Einstein's Aargau cantonal school certificate of 3 October 1896, as transcribed on einstein-website.de.
Jean-Jacques Rousseau, Les Confessions, Book VI (French text on Wikisource; English translation on Project Gutenberg); The Phrase Finder, "Let them eat cake".
G. H. Jacobs, J. F. Deegan and J. Neitz, "Photopigment basis for dichromatic color vision in cows, goats, and sheep", Visual Neuroscience, 1998; Live Science, "Why Do Bulls Charge When They See Red?".
National Museum of Denmark, "Viking helmets"; Snopes, "Did Vikings Actually Have Horns on Their Helmets?"; museum-digital, "Carl Emil Doepler (1824-1905)".
BBC Wildlife (Discover Wildlife), "Do goldfish really only have a 3 second memory?".
NPR Planet Money, "Episode 555: Why Is The Milk In The Back Of The Store?", 21 September 2016.
Snopes, "Bison Are Only Animals to Walk Through Blizzards…?".
Antoine-Henri Jomini, Vie politique et militaire de Napoléon, racontée par lui-même (Paris, 1827), volume 2 (Internet Archive scan); Archibald Alison and Austin Bierbower as quoted by Quote Investigator, "Never Interfere With an Enemy While He's in the Process of Destroying Himself".
US National Weather Service, "Lightning Myths".
Scientific American, "Do People Only Use 10 Percent of Their Brains?" (quoting Barry Gordon).
Natasha Geiling, "The Science Behind Honey's Eternal Shelf Life", Smithsonian, 22 August 2013; NHS, "Botulism".

Checked September 2026. Where a claim could not be settled, it is marked rather than guessed at. If you can show any of these wrong — with a source — we want to see it: hacksvitae@gmail.com

Real or Myth? Animal Mindsets and Shop Tricks, Checked

Hacks Vitae — Wed, 16 Sep 2026 00:00:00 GMT

Two kinds of post travel with a lot of confidence and very few sources. One is the animal mindset post: be the wolf, be the eagle, be the lion. The other is shop psychology: the secret tricks stores use on you.

They share a structure: a striking claim, a lesson attached, no source. So we took twelve of them and looked for the research.

The animal lessons mostly don't hold up. The shop tricks do better: several have real research behind them, though the effects are smaller or more conditional than the posts suggest.

Why the animal lessons often don't hold up

Every animal in these posts is doing something a person wants to be told to do. The wolf is dominant. The eagle suffers and is reborn. The lion ignores critics. Our reading: these stories start with a feeling and then go looking for an animal to carry it.

The alpha wolf is the clearest case, because the correction came from the scientist whose book spread the term. L. David Mech's The Wolf was published in 1970, when most research on pack behaviour had been done on "non-natural assortments of captive wolves", as he later put it. He then spent 13 summers watching wild wolves on Ellesmere Island in the Canadian Arctic, and in 1999 he concluded that "the typical wolf pack is a family". The wolves that lead most packs, his own website now says, are "merely breeders, or parents".

That is science working as it should, and it may be a better lesson than the one in the post: the person best placed to defend an idea was the one who dropped it when the evidence changed.

Why the shop tricks hold up better

Shop claims can be tested against real sales, and several have been. That is why the 9-ending price does better here than the eagle does.

There is a pattern in the partly-true ones. The decoy effect and the slow-music effect both began with a striking early study, and later, larger work found them weaker or more conditional. In both cases, the first result was the most dramatic one, which is part of why it is the one that gets shared.

How to check the next one

Ask whether anyone observed the animal. A claim about eagles should come from someone who studies eagles. If it comes from a slideshow, it is a parable.
Look for the year. A famous result from decades ago is worth checking for what has happened since.
Ask who the 95% are. A statistic that compares you with "the world" usually includes billions of people who never tried.
Prefer the correction to the claim. When the original researcher revises an idea, the revision is the thing to repeat.

Sources

L. David Mech, "Alpha status, dominance, and division of labor in wolf packs", Canadian Journal of Zoology 77 (1999): 1196–1203; Mech's website, "Wolf News and Information".
Snopes, "The Rebirth of the Eagle" (quoting the University of Minnesota's Raptor Center); National Eagle Center, "Fascinating Eagle FAQ".
Whit Gibbons, "The Legend of the Boiling Frog is Just a Legend", Ecoviews, Savannah River Ecology Laboratory (quoting Victor Hutchison); Fast Company, "Next Time, What Say We Boil a Consultant" (1995, quoting Douglas Melton); Theo Offerman and Ailko van der Veen, "Does the frog jump out of the boiling water?", University of Amsterdam working paper (2010), Appendix A, on the 19th-century experiments.
McGill Office for Science and Society, "Lobsters Are Not Immortal but the Myth That They Are Seems to Be"; Natural History Museum, London, "Are lobsters immortal?".
Barry Popik, "A lion doesn't concern himself with the opinions of a sheep".
Pliny the Elder, Natural History 10.1 (tr. John Bostock and H. T. Riley, on Perseus); San Diego Zoo Wildlife Alliance, "Ostrich".
Robynne Boyd, "Fact or Fiction?: Birds (and Other Critters) Abandon Their Young at the Slightest Human Touch", Scientific American.
Manoj Thomas and Vicki Morwitz, "Penny Wise and Pound Foolish: The Left-Digit Effect in Price Cognition", Journal of Consumer Research 32 (2005); Eric Anderson and Duncan Simester, "Effects of $9 Price Endings on Retail Sales", Quantitative Marketing and Economics (2003), and "Mind Your Pricing Cues", Harvard Business Review (September 2003); Antonio Filippin, "The Effect of .99 Price Endings on Consumer Demand", Journal of Economics and Management (2013).
Joel Huber, John Payne and Christopher Puto, "Adding Asymmetrically Dominated Alternatives", Journal of Consumer Research (1982); Shane Frederick, Leonard Lee and Ernest Baskin, "The Limits of Attraction", Journal of Marketing Research (2014); Huber, Payne and Puto, "Let's Be Honest About the Attraction Effect", Journal of Marketing Research (2014).
Ronald Milliman, "Using Background Music to Affect the Behavior of Supermarket Shoppers", Journal of Marketing (1982); Klemens Knoeferle and colleagues, "It is all in the mix", Marketing Letters (2012); Fabienne Krywuczky, Peeter Verlegh and Baris Kocaman, "Background Music in Retail: When taking it slow may benefit you", EMAC conference proceedings (2025).
K. Anders Ericsson, Ralf Krampe and Clemens Tesch-Römer, "The role of deliberate practice in the acquisition of expert performance", Psychological Review 100 (1993); Ericsson and Robert Pool, excerpt adapted from Peak (2016), Salon, 10 April 2016; Brooke Macnamara, David Hambrick and Frederick Oswald, "Deliberate practice and performance in music, games, sports, education, and professions: a meta-analysis", Psychological Science (2014), and its corrigendum (Psychological Science, 2018); Hambrick, Macnamara and Oswald, "Is the Deliberate Practice View Defensible?", Frontiers in Psychology (2020); Ericsson, "Summing up hours of any type of practice versus identifying optimal practice activities", Perspectives on Psychological Science (2016), and his response to Macnamara and Hambrick in Psychological Research (2021).
Game of Thrones, "You Win or You Die" (HBO, 2011).

Checked September 2026. Nothing on this page is financial or veterinary advice. If you can show any verdict here is wrong, with a source: hacksvitae@gmail.com

Your Fears, Answered by the Stoics — With the Actual Passages

Hacks Vitae — Wed, 16 Sep 2026 00:00:00 GMT

Stoic writers had a great deal to say about fear, and their lines are shared widely, sometimes without a book, a section or a translator attached.

So here are eight ordinary fears, each with what a Stoic wrote about it and where to find it: the book and section, in a translation old enough to be free to read. That way you can read each line in context rather than take our word for it.

Six of the eight answers are in the texts. Two popular answers are not Stoic: amor fati is Nietzsche's phrase, and "you have power over your mind, not outside events" is not in the Meditations.

The English quotations come from these translations: George Long for Marcus Aurelius, Thomas Wentworth Higginson for Epictetus, Richard M. Gummere for Seneca's letters, and John W. Basore for Seneca's On the Shortness of Life.

The method, briefly

Every real passage below has a book and a section number. The two marked as myths have neither. For the Marcus Aurelius line, we searched George Long's translation of the Meditations and the older Casaubon translation on Project Gutenberg for "power over your mind" and "outside events", and found neither phrase.

Where a shared English version has drifted from the original, the entry says what was dropped. In the two cases on this page where that happens, the words that went missing are the qualifying ones: "more often" in Seneca, and the calm second half of Marcus's sentence about the last day. In our reading, a qualifier makes a sentence longer, and a shorter line is easier to share.

The one worth reading in full

Seneca's line about imagination is widely shared, and the shared version is often trimmed in a way that changes what it claims.

Plura sunt, Lucili, quae nos terrent quam quae premunt, et saepius opinione quam re laboramus.

"There are more things, Lucilius, likely to frighten us than there are to crush us; we suffer more often in imagination than in reality."

— Seneca, Letters to Lucilius 13.4, translated by Richard M. Gummere

A version shared widely online reads "We suffer more in imagination than in reality." Two things go missing.

"More often" becomes "more." The Latin saepius is the comparative of saepe, "often", so it means "more often". In our reading, Seneca is counting occasions, not measuring intensity: he does not claim that imagined suffering is worse than real suffering.

The comparison disappears. The first half of his sentence does the work: more things frighten us than crush us. That is an observation about the gap between threat and outcome.

Seneca also says straight away how he means it: "I am not speaking with you in the Stoic strain but in my milder style." And in the next section, 13.5, he sorts the problem into three: "some things torment us more than they ought; some torment us before they ought; and some torment us when they ought not to torment us at all."

The Letters to Lucilius come from the last years of Seneca's life, the period from 62 to 65 CE that he spent in retirement, according to the Stanford Encyclopedia of Philosophy. Gummere's translation gives this one the title "On groundless fears".

What this does not do

Stoicism is a philosophy, not a treatment. It offers ways of thinking about fear. It is not a substitute for care.

If fear is running your life — if you are avoiding things you need to do, not sleeping, or feeling it in your body most days — that is something a doctor or a therapist can help with.

Cognitive behavioural therapy (CBT) is one such treatment, and it grew partly out of these ideas. Albert Ellis and Aaron Beck, the founders of rational emotive behaviour therapy and cognitive behavioural therapy, both acknowledged Stoicism as the philosophical precursor of their approaches. Beck opened his first book on cognitive therapy by writing that its "philosophical underpinnings go back thousands of years, certainly to the time of the Stoics, who considered man's conceptions (or misconceptions) of events rather than the events themselves as the key to his emotional upsets". The treatments were then tested: a 2012 review of 106 meta-analyses found that "the strongest support exists for CBT of anxiety disorders", among other problems.

Putting the passages to use

The passages above describe a few checks you can run on a fear of your own. None of them needs any belief in Stoicism.

Sort it. Epictetus opens the Enchiridion by dividing things into two kinds. Within our power are "opinion, aim, desire, aversion"; beyond it are "body, property, reputation, office". He puts reputation beyond our power, which bears directly on the first fear in the table.
Name which of Seneca's three it is. Is it tormenting you more than it ought, before it ought, or when it ought not to at all (Letters 13.5)?
Look for the judgement. Marcus, at 8.47, locates the pain not in the external thing but in "thy own judgment about it", and says it is in your power to wipe that judgement out.
Read the passage before you share the line. A book and section number lets you read a quotation in context in a few minutes. The two myths in the table are what that check catches.

Sources

Marcus Aurelius, Meditations 7.69, 8.47, 9.3 and 12.4, trans. George Long (Project Gutenberg #15877); the Casaubon translation (Project Gutenberg #2680), searched.
Epictetus, Enchiridion 1, 8, 9 and 11, trans. Thomas Wentworth Higginson (Project Gutenberg #45109).
Seneca, Epistulae Morales 13.4–5, Latin text (The Latin Library); Moral Letters to Lucilius, Letter 13, trans. Richard M. Gummere (Wikisource).
Seneca, On the Shortness of Life, chapters 1 and 7, trans. John W. Basore (Loeb, 1932; Wikisource), with Basore's note on Paulinus.
Friedrich Nietzsche, The Joyful Wisdom (The Gay Science) §276, trans. Thomas Common (Project Gutenberg #52881); Ecce Homo, "Why I Am So Clever" §10, trans. Anthony M. Ludovici (Project Gutenberg #52190); Beyond Good and Evil §9, trans. Helen Zimmern (Project Gutenberg #4363).
Stanford Encyclopedia of Philosophy, "Epictetus" and "Seneca".
Wiktionary, "saepius".
Andrea E. Cavanna and colleagues, "The Western origins of mindfulness therapy in ancient Rome", Neurological Sciences (2023), on Ellis and Beck.
Stefan G. Hofmann and colleagues, "The Efficacy of Cognitive Behavioral Therapy: A Review of Meta-analyses", Cognitive Therapy and Research (2012), abstract via Europe PMC.

Checked September 2026. The Stoic and Nietzsche passages rest on the free online translations above: Long, Higginson, Gummere, Basore, Common, Ludovici and Zimmern. The dates and background rest on the Stanford Encyclopedia of Philosophy; the Ellis and Beck lines on Cavanna and colleagues; and the research on CBT on Hofmann and colleagues. If you can show any of these wrong, with a source, we want to see it: hacksvitae@gmail.com.

The Web Design Resource Library: 150+ Tools, Components and Libraries, Reviewed

Hacks Vitae — Sat, 12 Sep 2026 00:00:00 GMT

Design resource lists age fast. Links die, licences change, prices move, and a project can stop being maintained while its website keeps loading. This guide keeps the list long and describes each tool mostly from its own pages, with the price and licence as they read in September 2026; where a fact comes from somewhere else, such as Wikipedia, the method section at the end names it.

How to read it: where a tool is free, it says so; where it is paid, the price is given in the currency the vendor showed, US dollars unless noted. Where a price only appears after you log in, or a site would not let us read it, the guide leaves the price out rather than printing an old number, and the last section lists those gaps. It also explains how everything was checked.

Start here — if you read nothing else

Six resources that give a beginner a lot of ground for little effort:

MDN Web Docs — "Resources for Developers, by Developers", documenting CSS, HTML and JavaScript since 2005. Mozilla describes it as "a free, high-quality, and comprehensive platform".
Refactoring UI — $99 (The Essentials) or $149 (The Complete Package), US dollars plus local taxes. A design book and video set "from the creators of Tailwind CSS", with tactics "explained from a developer's point-of-view".
Realtime Colors — "100% free! Forever." It "distributes the colors on a real website", so you judge a palette and fonts on a page rather than as a row of swatches.
Baseline — gives "clear information about which web platform features are ready to use in your projects today".
Astro hosted on Cloudflare — Astro is MIT-licensed and ships "zero JS, by default"; on Cloudflare Workers, "requests to static assets are free and unlimited".
Utopia — calculators that "elegantly scale type and space without breakpoints".

Design inspiration galleries

Awwwards gives out web design awards, and entering is paid: a standard submission costs $65 for one website, and a site has to be submitted to win a Site of the Day or its other awards.

Gallery	In its own words	Price, where the site states one
SiteInspire	"A showcase of the web's finest design + talent"; new entries added through late September 2026	—
One Page Love	One-page websites and page sections, "hand-picked since 2008"	—
Lapa Ninja	"over 7,300 landing page designs since 2015, along with 15,000+ full-page website screenshots"	—
Minimal Gallery	"a curated source of website design inspiration for creatives. Since 2013."	—
Land-book	"a curated website design gallery for creatives, updated daily"	A Pro plan removes ads
Mobbin	a "library of real mobile and web app screenshots"	Pro $10 a month, billed yearly
Are.na	"no ads, no personalized recommendations"	Free up to 200 blocks; Premium $7 a month or $70 a year
Savee	"a curated place for visual inspiration", "without ads or clutter"	—
Cosmos	"a visual inspiration platform for creatives" in "a distraction-free, ad-free environment"	—
Httpster	Website design inspiration, filterable by style and type	—
Brutalist Websites	A collection of brutalist sites; its header describes brutalism as "a reaction by a younger generation to the lightness, optimism, and frivolity of today's web design"	—
Best Website Gallery	"the most beautiful websites handpicked for you"; picks added through September 2026	—
Behance	A portfolio network "founded in 2006", with "over 50 million members"	Behance Pro available

Changed name: godly.website now redirects to recent.design, a design inspiration site covering web, interface, branding, typography, motion and more.

Dribbble is a platform where designers show their work and find clients. Its paid plans start at $4 a month, billed yearly (Pro Lite), and add features such as receiving leads and sending proposals.

3D, WebGL and motion

GSAP is free, under its own licence

GSAP's pricing page says it plainly: "GSAP is now 100% free for all users, thanks to Webflow's support." Its licence FAQ adds that "all of GSAP including the plugins that were formerly 'members-only' like SplitText and MorphSVG can be used in commercial projects at no charge".

Two details are worth knowing before you build a product on it:

It is Webflow's own "no charge" licence (effective 30 April 2025), not an open-source licence. Without Webflow's written consent, it does not allow GSAP in tools that let users build animations without code in a way that competes with Webflow's visual animation building.
The same licence page still shows an older plain-English summary saying that if end users are charged a fee, you should sign up for a "Business" Club GSAP membership. The FAQ below it says commercial use is covered at no charge. If your product charges its users, read the full licence text rather than either summary.

The stack

Tool	What it is	Cost and licence
Three.js	The JavaScript 3D library; its site showed revision r186	Free, MIT
ThreeUI	"copy-ready Three.js components, complete website templates, WebGL backgrounds, hero sections, UI effects", by Meng To	Community library free under MIT (50 components, 111 routes, 141 free variants); Pro $199 a year or $299 lifetime
Motion	"Motion (prev Framer Motion)", "a fast, production-grade animation library for React, JavaScript and Vue"	Free, MIT; paid Motion+ adds premium APIs and examples
Lenis	A smooth-scroll library for JavaScript, React, Vue and more, by darkroom.engineering	Free, MIT
Spline	A 3D design tool in the browser	Free plan with watermarked web exports; Hobby $12 and Pro $25 per seat a month, billed annually
Rive	"The interactive experience engine": an editor for designing and animating interactive graphics, driven by state machines	"Free to create", $9 a month per seat "to ship" (Cadet plan)
Lottie	Airbnb's library that renders After Effects animations exported as JSON on the web, Android and iOS	Free, MIT
p5.js	"a friendly tool for learning to code and make art"	Free and open source (LGPL 2.1)

Three.js Journey by Bruno Simon is a paid course: $95 for the Three.js course, $45 for the WebGPU & TSL course, or $120 for both, VAT included, as a one-time payment with access for life. The site lists 116 hours of video across 87 lessons.

Check the release date before you install

Before installing an animation library, it is worth checking when it last shipped. The last version each of these published on npm, as the registry showed it:

Library	Last version published on npm
Velocity	2.0.6, April 2020
Popmotion	11.0.5, August 2022
Vanta.js	0.5.24, September 2022
Curtains.js	8.1.6, May 2024
Theatre.js	0.7.2, May 2024
Locomotive Scroll	5.0.1, January 2026

An old release date is not a verdict on its own: a small, finished library can keep working. It does mean no fix has been published to npm since, which matters most for anything touching scrolling, input or accessibility.

Visual styles, with their dates

Liquid Glass

Apple announced Liquid Glass on 9 June 2025, describing it as a translucent material that "reflects and refracts its surroundings". A year later, announcing iOS 27 on 8 June 2026, Apple said the design "continues to evolve, with improvements to legibility, customizability, and consistency", and added "a new transparency slider in Settings". Its iOS page says the slider sets Liquid Glass anywhere "from ultraclear to fully tinted".

If you want to borrow the look, Liquid Glass Design is a gallery of 170+ liquid-glass and glassmorphism examples, each with a copy-ready AI style prompt. It explains the difference from ordinary frosted glass: "rather than simply blurring what sits behind it, a Liquid Glass surface refracts that content like a real lens". It also links Liqui UI, "Liquid Glass React components, free and open source". It is maintained by seergb.design.

Other styles

Glassmorphism — the Interaction Design Foundation credits the term to Michał Malewicz and says the style "became popular after Apple updated macOS Big Sur in 2020".
Neumorphism — named in 2019, but the sources disagree on by whom: the Interaction Design Foundation says Malewicz, Wikipedia says Jason Kelly. Wikipedia adds that, partly because of accessibility issues, take-up "had quickly declined by 2021".
Neubrutalism — neubrutalism.com calls itself "the definitive reference on neubrutalist web design", and its guidance includes keeping body text at a 4.5:1 contrast ratio (WCAG AA), which loud palettes often miss.
Textmode and ASCII — Textmode Overlay is a browser extension that turns canvas and video elements on a page into live ASCII art. It is "free and open source under the MIT license", and its site says all media processing runs locally on your device.

Typography

Fontsource — "Download & self-host open-source fonts": free, open-source and variable fonts, which you can "self-host web fonts with npm packages and CSS", with CDN links also available.
Google Fonts — its FAQ says "all Google Fonts are open source and are without cost", and that "you can use them commercially", under each font's licence (most often the SIL Open Font License). It also explains self-hosting: "When a font is loaded from the website operator's servers, Google does not receive any kind of data related to the visits to the website." The same FAQ lists drawbacks, starting with a bigger download, because you serve the whole font file rather than the pieces Google's API would send.
Fontshare — free fonts from the Indian Type Foundry. Fonts under its own ITF Free Font License may be used "for personal or commercial purposes, free of charge", but "you may not create a Derivative Work from the Font Software without the prior written consent of the Licensor". Some Fontshare fonts use the SIL Open Font License instead, so check which licence a font carries.
Utopia — fluid type and space calculators that generate CSS clamp() values.
Modern Font Stacks — "system font stack CSS organized by typeface classification for every modern OS", so text renders with fonts already on the device. Released under CC0.
Typewolf — a site on "what's trending in type", running since 2013. Its newest "site of the day" entry, when we read it, was dated 14 December 2025.
v-fonts.com — a catalogue of variable fonts. The newest release dates listed on its home page are from April 2022.

One legal reason people self-host: in January 2022 a Munich regional court (case 3 O 17493/20) held that loading Google Fonts dynamically from Google's servers without the visitor's consent breached EU data-protection law, and awarded the visitor €100 in damages. That is one German ruling, described here as reported by a German law firm; it is not advice on what your site must do.

Colour

Realtime Colors — "100% free! Forever." It previews a palette and fonts on a realistic page, and says it has "300k+ users".
OKLCH Color Picker — picks and converts colours in OKLCH, with a fallback for ordinary sRGB screens. Made at Evil Martians by Andrey Sitnik and Roman Shamin; the code is MIT-licensed.
Leonardo — Adobe's open-source tool (Apache 2.0) that can "generate colors based on a desired contrast ratio", for accessible colour systems.
WebAIM Contrast Checker — enter two colours and it reports the contrast ratio against the WCAG AA and AAA levels for normal and large text.
Coolors — a fast palette generator, with its own contrast checker and palette visualiser. The free tier shows ads and is limited (for example, 10 saved palettes); Pro removes ads and allows palettes of more than five colours. Its pop-up offered Pro at $3 a month.

A palette generator picks colours that look good together; it does not know which one will sit behind your body text. That is why a contrast check on the actual text-and-background pair is worth the extra minute, whichever tool picked the colours.

Icons and illustration

Icons

Lucide is a good default: "beautiful & consistent icon toolkit made by the community", released under the ISC licence (the icons it inherited from Feather keep Feather's MIT licence), and "a fork of Feather Icons". Its site showed version 1.48.0.

Set	What it says about itself	Licence and cost
Tabler Icons	"6200+ free vector icons for web design"	Free, MIT; a compiled bundle (PNG, PDF, webfont) is $9
Phosphor	"a flexible icon family", in six weights: thin, light, regular, bold, fill and duotone	Free, MIT
Iconoir	"an open-source library with 1600+ unique SVG icons, designed on a 24x24 pixels grid"	Free, MIT
Material Symbols	Google's icon set, served from Google Fonts	Free, Apache 2.0
Iconify	"All popular icon sets, one framework": "over 300,000 open source vector icons"	Free; each icon set keeps its own licence
Simple Icons	SVG icons for popular brands	The project is CC0, with the caveats below

Three licence details worth knowing

Feather Icons has gone quiet. Its GitHub repository was last pushed on 11 March 2025 and had 510 open issues when we read it. It is not archived, so it can look active at a glance. Lucide started as its fork.

Remix Icon changed its licence in January 2026. The repository's licence file is now the "Remix Icon License v1.0", dated January 2026. It forbids using the icons "to create, distribute, or sell a competing icon library or icon set", and using them "as a logo, trademark, or brand identifier". The npm package, however, still declares "license": "Apache-2.0" (version 4.9.1), so a licence scanner that reads package metadata will report the old licence.

Simple Icons' CC0 does not cover the brands. Its disclaimer says the project "is released under CC0 - though that doesn't mean to imply that all icons within the project are also CC0", and asks users to "seek the correct permissions to use the icons relevant to their project" and to follow each brand's guidelines.

Illustration

unDraw — free for "noncommercial and commercial purposes", and "you do not need to ask permission from or provide credit". Its licence "explicitly prohibits the use of unDraw assets, vectors, and images for training, fine-tuning, or developing artificial intelligence", and asks you not to "redistribute in packs the illustrations".
Humaaans and Open Peeps — mix-and-match illustrations of people by Pablo Stanley, "free for commercial and personal use under CC0 license".
DrawKit — free assets, then Pro at $79 a year or $295 for lifetime access. Its licence bars using the illustrations to train AI models and selling merchandise with them "on posters, prints, mugs".
Storyset — free customisable, animatable illustrations, on condition that use "is credited to the company/website". Using them without credit needs a Flaticon Premium subscription, which Flaticon's pricing page listed at €9.99 a month or €89.99 a year (US prices $11.99 and $99).
Icons8 illustrations, including the Ouch! collection — free if "you set a link to icons8.com wherever you use them"; Icons8 also sells subscriptions under a separate paid licence.
Absurd Design — "a singular, hand-drawn artistic universe for digital use". Its free licence is for "personal, educational, and other non-commercial purposes only" and requires the credit "Illustration(s) from absurd.design" with a link. Commercial use comes with a paid membership ("Absurd at Core" is €19 a month or €190 a year), and that licence lasts only as long as the membership: when it ends, the illustrations have to come out of anything that uses them.

Gone:

Glaze — glazestock.com now redirects to an unrelated gambling site.
Popsy illustrations — popsy.co/illustrations redirects to an app page that returned a Cloudflare "origin DNS error".

Free photography — and the licence risk

Source	Who runs it	What it offers
Unsplash	Owned by Getty Images since 2021	Free photos under the Unsplash License, which does not include the right "to compile images from Unsplash to replicate a similar or competing service"
Pexels	Owned by Canva	"All photos and videos on Pexels are free to use", no attribution required
Pixabay	Acquired by Canva in November 2018	Free images, video and music under its Content License
Openverse	A WordPress project, successor to Creative Commons' CC Search	Searches "more than 800 million images and audio tracks" that are openly licensed or public domain
Smithsonian Open Access	Smithsonian Institution	Millions of collection images released as CC0
Flickr Commons	Flickr Foundation	Historical photography from "over 100 cultural institutions in 24 countries", launched in 2008 with the Library of Congress

A free licence covers the photographer's rights, not everything in the picture. The sites say this themselves:

Pixabay: "It is your responsibility to check whether you require the consent of a third party or a license to use Content." Its licence also says that if an image contains recognisable trademarks, logos or brands, "you cannot use that Content for commercial purposes in relation to goods and services".
Pexels: "Don't imply endorsement of your product by people or brands on the imagery", and "Identifiable people may not appear in a bad light or in a way that is offensive".
The Smithsonian: even for CC0 content, "a third party may claim rights in the content such as trademark, privacy, or publicity rights".

Licences change, too. Before June 2017 Unsplash photos were released under CC0; they now carry Unsplash's own licence. Pixabay changed its terms on 17 April 2023, when the "Pixabay License" became the "Content License". What a site says today is not proof of what it said on the day you downloaded, so it helps to keep a record of the source, the date and the licence wording for each image you use commercially.

On AI images: in Part 2 of its report on copyright and AI (29 January 2025), the US Copyright Office concluded that "the outputs of generative AI can be protected by copyright only where a human author has determined sufficient expressive elements", and that this does not include "the mere provision of prompts".

Frameworks and CSS

Astro — free, MIT. "The web framework for building content-driven websites like blogs, marketing, and e-commerce." It was at 7.3.5 on npm. Note the pace: Astro 6.0 came out on 10 March 2026 and Astro 7.0 on 22 June 2026, two major versions within four months, which means planning time for upgrades.
Eleventy — free, MIT. "Eleventy is a simpler static site generator"; version 3.1.6 on npm.
Hugo — free, Apache 2.0. "A static site generator written in Go, optimized for speed and designed for flexibility".
Next.js — free, MIT, made by Vercel; version 16.3.7 on npm. It runs server code, which brings server security updates with it: in March 2025 Vercel patched CVE-2025-29927, which it called "a critical severity vulnerability in Next.js", and self-hosted apps using next start and output: 'standalone' were affected. A self-hosted Next.js site depends on keeping up with patches like this one.
Jekyll — free, MIT. "Transform your plain text into static websites and blogs." Its latest GitHub release was v4.4.1, published on 29 January 2025.

Who funds what: Vercel makes Next.js, hired Svelte's creator Rich Harris in November 2021, and NuxtLabs (the company behind Nuxt) announced it "is joining Vercel". If you are picking a framework partly to avoid depending on one company, that is worth knowing.

Tailwind — and its September 2026 news

Tailwind CSS is free and MIT-licensed, at version 4.3.3 on npm.

On 9 September 2026, Tailwind Labs announced it is joining Shopify. Its post says: "Nothing changes with Tailwind CSS or any of our other open-source projects. Everything will always be MIT-licensed." On the paid side, existing customers keep access to Tailwind Plus and ui.sh, "but we're closing sign ups for new customers".

Alternatives worth knowing:

Open Props — "supercharged CSS variables": design tokens rather than utility classes. Free, MIT.
Pico CSS — styles your HTML tags directly, "using fewer than 10 .classes overall", with a class-less version. Free, MIT.
Every Layout — a book and site on composable CSS layouts: "Buy Every Layout for $69", with several chapters free to read.

Check the date before linking: Water.css, a classless stylesheet, last published a version on npm in August 2021.

Modern CSS you may not need a library for

These features are "Baseline Widely available" in the web-features data behind Baseline (version 3.40.0), meaning they have worked across the core browsers for at least 30 months. The date each crossed that line:

Feature	What it can replace	Widely available since
Cascade layers	Specificity hacks and `!important` fights	September 2024
`<dialog>`	Modal libraries	September 2024
Container queries	Element-query polyfills	August 2025
`color-mix()`	Sass colour functions	November 2025
`subgrid`	Nested-grid alignment hacks	March 2026
CSS nesting	A common reason to install Sass	June 2026
`:has()`	"Add a class with JavaScript" for parent styling	June 2026

Not Baseline yet: anchor positioning, scroll-driven animations and cross-document view transitions all have limited availability in the same data. For scroll-driven animations and cross-document view transitions, the browser missing from the support list is Firefox.

Components, templates and code you can copy

Newer tools built for AI coding agents (Cult UI, Componentry, Refero Styles, Taste Skill, Impeccable, Playwright CLI and Awesome DESIGN.md) are covered separately in AI design tools and component libraries, checked.

Three different things get called "component libraries", and it helps to keep them apart:

Registries copy source code into your project. You own it and edit it, and nothing updates it for you.
Libraries are npm dependencies. They update, and you inherit both the fixes and the breaking changes.
Templates are finished pages you start from and change.

Copy-paste registries

Registry	What you get	Cost and licence
shadcn/ui	Accessible components and "a code distribution platform": its CLI adds each component's source to your project	Free, MIT
Magic UI	"150+ free and open-source animated components and effects", billed as a "perfect companion for shadcn/ui"	Free, MIT; a paid Magic UI Pro adds "50+ blocks and templates"
Aceternity UI	A "React & Tailwind CSS component library" of visual-effect components, blocks and templates	Free components; paid all-access $169 a year or $199 lifetime
React Bits	"200+ high quality, animated, interactive & fully customizable React components"	Free, under "MIT + Commons Clause" (see below)
21st.dev	"12,000+ hand-crafted React and Tailwind CSS components, templates, and shadcn themes" from many authors	Browsing is free with "2 free component copies every day"; membership unlocks more

(ThreeUI, the Three.js registry, is in the 3D section above.)

Two licence details in that table:

React Bits is "MIT + Commons Clause", which is not the same as MIT. Its licence says "you may use this software, including for any commercial purpose, so long as you do not sell, sublicense, or redistribute the components themselves". A dependency scanner that only reads the letters "MIT" can miss the restriction.

On 21st.dev, each component's licence is its author's. The site's terms say the underlying component code "remains separately owned or licensed by that author", and that the demos and previews belong to 21st Labs. So the licence that matters is each component's, not the site's.

Headless primitives — the behaviour layer

These ship no visual design: you bring the styling, and they handle behaviour and accessibility. Base UI's own description is "unstyled UI components for building accessible user interfaces".

Library	Notes	Licence
Base UI	"From the creators of Radix, Floating UI, and Material UI". Version 1.0.0 shipped on 11 December 2025; v1.8.0 on 4 September 2026. shadcn/ui's changelog lists "Base UI as the default" in July 2026.	Free, MIT
Radix Primitives	Made by WorkOS. Its GitHub repository was last pushed on 8 August 2026.	Free, MIT
React Aria	Adobe's "over 50 components with built-in behavior, adaptive interactions, top-tier accessibility, and internationalization out of the box"	Free, Apache 2.0
Headless UI	Tailwind Labs' "completely unstyled, fully accessible UI components". Latest release: v2.2.10, 7 April 2026.	Free, MIT

Tailwind component and block libraries

Library	What it is	Licence
daisyUI	A "Tailwind CSS component library" with semantic class names; "pure CSS. No JS dependency". Version 5.7.46.	Free, MIT
HeroUI	"Previously NextUI": "accessible React UI components built on React Aria and Tailwind CSS v4"	Free, Apache 2.0 since v3.0.3 (April 2026; MIT before that). Its npm package still declares MIT
Preline UI	"An open-source Tailwind CSS UI component library", with blocks, templates and plugins	Free for personal and commercial use under two licences, MIT and the "Preline UI Fair Use License", which bars using it to build a product "that directly competes with Preline UI"
Flowbite	Tailwind CSS components, with a paid Pro version	Released code under MIT; the name and logos are trademarks of Bergside Inc.

Tailwind Plus (renamed from Tailwind UI in March 2025) is closed to new customers after Tailwind Labs' move to Shopify; its page now opens a log-in screen for existing customers.

Full component systems

Library	Notes	Cost and licence
Mantine	"More than 120 customizable components and 70 hooks"; its Mantine UI collection of premade responsive components is "free forever for everyone"	Free, MIT
Chakra UI	"Accessible React components for building high-quality web apps and design systems"	Core free, MIT; Chakra UI Pro blocks $299 personal or $899 team, "one-time purchase - lifetime access"
Ant Design	A React design system made by Ant Group and the Ant Design community	Free, MIT

Templates and whole layouts

Source	What you get	Cost and licence
Cruip	"Tailwind CSS templates for your next project": landing pages, sites and dashboards	A free-templates section, and paid templates such as $49, $79 and $129
HTML5 UP	Responsive HTML5 and CSS3 site templates	Free under Creative Commons Attribution 3.0: credit required
Astro Themes	Astro's official directory of "free and premium Astro themes"	Both
Landingfolio	Landing-page inspiration plus Webflow, Figma and Tailwind components	—
Vercel Templates	Starter projects you can deploy to Vercel	—

The HTML5 UP licence asks for one thing in return. Its templates are "licensed under the Creative Commons Attribution 3.0 License", which means you can use them for personal and commercial work and change them however you like; "in exchange, just give HTML5 UP credit for the design". The Creative Commons legal code says that credit "may be implemented in any reasonable manner", so it need not sit in any particular place on the page. (The author sells attribution-free versions through a separate site, Pixelarity.)

Learning by reading other people's code

Resource	What it is
Codrops	Tutorials, demos and case studies; "fueling web creativity since 2009"
CodePen	"A social development environment for designers and developers"
Frontend Mentor	Real-world HTML, CSS and JavaScript challenges built from professional designs; "learn to code for free"
StackBlitz	"Instant dev environments" in the browser
CodeSandbox	"Instant cloud development environments"

How to choose between these

A rough rule, since the options overlap:

You want to own the code and never fight an upgrade → a registry (shadcn/ui, ThreeUI, Magic UI), accepting that upstream fixes will not reach you automatically.
You want accessibility handled and will do your own styling → headless primitives (Base UI, Radix, or React Aria if internationalisation matters).
You want to ship fast and do not need a custom look → a full system (Mantine, daisyUI, HeroUI).
You want a finished page to change → templates (Cruip, HTML5 UP, Astro Themes).
You want 3D without writing shaders → ThreeUI, or Spline if you would rather not touch code.

Accessibility — what a million home pages show

The WebAIM Million tests the home pages of the top one million websites each year. Its 2026 report (data from February 2026) found:

95.9% of home pages had detected WCAG 2 failures, up from 94.8% in 2025, "reversing a trend of small improvements each of the previous 6 years".
An average of 56.1 errors per page, up 10.1% on 2025.
96% of all detected errors fell into six categories:

Failure	Share of home pages
Low-contrast text	83.9%
Missing alternative text for images	53.1%
Missing form input labels	51%
Empty links	46.3%
Empty buttons	30.6%
Missing document language	13.5%

WebAIM adds that "these most common errors have been the same for the last 7 years". It also notes that "addressing just these few types of issues would significantly improve accessibility", which makes the six a sensible first list.

Free tools: the axe DevTools browser extension (Deque describes its Chrome extension as free) · WAVE, WebAIM's "suite of evaluation tools" · Accessibility Insights, Microsoft's tools for Chrome, Edge and Windows, now open source · NVDA, "a free, high quality screen reader".

Standards, as they stand:

In the United States, the Justice Department's 2024 rule makes WCAG 2.1 Level AA "the technical standard for state and local governments' web content and mobile apps".
WCAG 2.2 is a W3C Recommendation; the current edition is dated 12 December 2024.
WCAG 3.0 is still a W3C Working Draft (latest: 10 September 2026), and the draft itself says it "still has several years of work".

Accessibility overlays

The Overlay Fact Sheet describes overlays as "technologies that aim to improve the accessibility of a website" by applying third-party code to it. Three things are on the record:

The Overlay Fact Sheet asks site owners who use overlays to switch to "more robust, independent, and permanent strategies"; its signatories include contributors and editors of the WCAG, ARIA and HTML specifications.
In WebAIM's survey of accessibility practitioners (January 2021), 67% rated these tools "not at all or not very effective", and respondents with disabilities were less favourable still, at 72%.
On 3 January 2025 the US Federal Trade Commission announced that accessiBe would pay $1 million to settle allegations that it misrepresented its AI tool's ability to make any website WCAG-compliant, and that it promoted the tool in articles on third-party websites "formatted to look like impartial and objective reviews". The FTC approved the final order on 22 April 2025.

Performance

Core Web Vitals, as Google's own pages state them:

Metric	Good
Largest Contentful Paint (LCP)	within 2.5 seconds
Interaction to Next Paint (INP)	200 milliseconds or less
Cumulative Layout Shift (CLS)	0.1 or less

Both web.dev and Google Search Central still give 2.5 seconds for LCP, so a claim that the threshold has been tightened should come with a Google link. INP replaced First Input Delay as a Core Web Vital on 12 March 2024, so a page that still talks about FID is out of date.

Tools:

Lighthouse — built into Chrome DevTools; its report covers performance, accessibility, best practices and SEO.
PageSpeed Insights — "provides both lab and field data about a page": Lighthouse lab results plus real-user data from the Chrome UX Report.
Unlighthouse — "a free, open-source CLI that runs Google Lighthouse on every page of your site in parallel". MIT-licensed.
WebPageTest — free tests from a choice of browsers, locations and devices. Created by Patrick Meenan in 2008, it is "now part of LogicMonitor through Catchpoint".

One thing that catches small sites: the field data comes from the Chrome UX Report, which only includes pages and sites that are "sufficiently popular", meaning they have "a minimum number of visitors". A low-traffic site may have no field data at all, only lab results.

Hosting — and what a traffic spike costs

Platform	Free tier	First paid plan	When traffic spikes
Cloudflare Workers	100,000 requests a day; requests to static assets are free and unlimited	Workers Paid, from $5 a month	"No additional charges for data transfer (egress) or throughput (bandwidth)"
Vercel	Hobby: the first 100 GB of fast data transfer; non-commercial use only	Pro, $20 a month per developer seat, with $20 of usage credit included	Pro's Flat Rate CDN: "a one-day traffic spike doesn't change your bill"
Netlify	300 credits a month	Personal, $9 a month (1,000 credits)	A site that reaches its credit limit is paused until the next billing cycle
GitHub Pages	A soft bandwidth limit of 100 GB a month	—	Not for running an online business or e-commerce site (see below)
Amazon CloudFront	A $0/month Free plan	Flat-rate plans "with no overage charges", or pay-as-you-go	Depends on the plan you pick

What changed recently: Vercel's Pro plan now includes a "Flat Rate CDN": "you pay for a capacity tier instead of per unit of usage, so a one-day traffic spike doesn't change your bill". If a whole billing cycle's usage exceeds your tier, Vercel "moves you to a higher tier at the start of the next cycle"; the included tier is 1 million CDN requests and 1 TB of data transfer. Amazon CloudFront has also added flat-rate plans.

Two sets of terms worth reading before you choose:

Vercel's free Hobby plan is non-commercial. Its fair-use guidelines say "Hobby teams are restricted to non-commercial personal use only", and list as commercial, among other things, sites where "affiliate linking is the primary purpose" and "the inclusion of advertisements, including but not limited to online advertising platforms like Google AdSense". Asking for donations does not count as commercial.
GitHub Pages "is not intended for or allowed to be used as a free web-hosting service to run your online business, e-commerce site, or any other website that is primarily directed at either facilitating commercial transactions or providing commercial software as a service (SaaS)".

Netlify's limit applies to the whole account: its pricing FAQ says that "if one project exceeds its limits, all projects on your account will be paused".

Vercel vs Cloudflare

The two are good at different things, and the choice depends mostly on whether your site runs code on every request.

Where Vercel is strong

Preview deployments by default. Vercel "creates a preview deployment" when you push a commit to a branch that is not your production branch, or open a pull request, so you can "deploy and test changes in a live setting, without affecting your production site".
Next.js. Vercel makes Next.js.
Spend controls. On Pro, spend management can notify you, trigger a webhook, or "pause the production deployment of all your projects" at a set amount. Vercel notes that "setting a spend amount does not stop usage on its own"; pausing has to be switched on.
Predictable CDN pricing on Pro, through the Flat Rate CDN tiers above.

Where Cloudflare is strong

No bandwidth bill. Cloudflare's pricing page says "there are no additional charges for data transfer (egress) or throughput (bandwidth)", and "requests to static assets are free and unlimited".
Low entry price for code. Workers Paid has "a minimum charge of $5 USD per month", including 10 million requests; after that, requests cost $0.30 per million, and CPU time beyond the 30 million milliseconds included costs $0.02 per million milliseconds.

The fine print on each side

Vercel: Hobby is non-commercial; Pro is priced per developer seat; the Flat Rate CDN excludes some workloads, such as "bulk file distribution or download services".
Cloudflare: the free plan's 100,000 requests are counted per day, not per month. And "when Workers caching is enabled, requests served from the Worker's cache are billed at the same per-request rate as requests that invoke the Worker", which the page says "includes requests to static assets". Cloudflare Pages still works, but its own docs now say: "Start new projects with Workers."

A rough rule, as we read it:

If your site is…	Worth a close look
Static pages, a blog or a content library	Cloudflare, where static asset requests are free and bandwidth is not billed
A Next.js application that uses its server features	Vercel, where the framework's own maker runs the platform
Running ads or affiliate links on a free plan	Not Vercel Hobby, which its terms reserve for non-commercial use
A team that lives in preview deployments	Vercel, with spend management and production pausing set up on day one

SSL — shorter certificates, so automate renewal

Certificate lifetimes are shrinking on a published schedule. From the CA/Browser Forum's Baseline Requirements:

Certificates issued from	Maximum validity
Before 15 March 2026	398 days
15 March 2026	200 days (current)
15 March 2027	100 days
15 March 2029	47 days

The period for reusing domain-validation data shrinks on the same dates, reaching 10 days from 15 March 2029, so from then on almost every renewal will need fresh proof that you control the domain.

At 47 days, a certificate has to be replaced roughly every six weeks. Meanwhile Let's Encrypt stopped sending expiration notification emails on 4 June 2025, saying that "more and more of our subscribers have been able to put reliable automation into place for certificate renewal". If your renewal breaks, nothing from them will warn you.

What this means in practice: renewal has to be automated, and it helps to have some separate expiry monitoring, because the automation is now the only thing between a site and a browser warning. Caddy, an open-source web server under the Apache 2.0 licence, was "the first web server to use HTTPS automatically and by default": it provisions certificates for your sites "and keeps them renewed".

No-code builders — can you take your site with you?

A visual builder is quick to start with. The question that matters later is what you can take with you if you leave. Here is what each builder's own help pages or pricing pages say:

Builder	What its own pages say about leaving
Webflow	Code export is available "on paid Workspace plans" (not on Site plans). CMS content, user accounts, ecommerce, form processing and site search are not included in the exported code.
Framer	"Framer does not provide a way to export published sites as HTML files or static website bundles."
Wix	"In order for your site to work properly, it needs to be hosted and operated on Wix's servers." You can connect your own domain.
Squarespace	Exports "certain content" to an XML file for WordPress; "not everything will export".
Bubble	"Bubble apps can only be run on the Bubble platform; there's no way of exporting your application as code." Data can be exported as CSV.
Readymag	Lists "Code export" on its Advanced plan ($58.50 a month billed yearly).
Dorik	Lists code export on its Business plan, not its Personal plan.
Carrd	One-page sites. Pro Lite ($9 a year) removes branding; custom domains start with Pro Standard at $19 a year.

A plan name can hide the feature you need. On Webflow's pricing page, Basic ($15 a month billed yearly) is "for simple sites that don't need a CMS". On Carrd, Pro Lite includes no custom domain.

When a builder is retired: Wix says Editor X "completed the transition to Wix Studio" in January 2025, with all Editor X sites moved to the Studio editor; editorx.com now goes to wix.com/studio.

AI app builders work differently. v0, Bolt, Lovable and Replit all generate real source code, and each documents a way to keep it in a GitHub repository:

v0: "connect a v0 project to GitHub to keep its code in a repository".
Bolt: "connect a GitHub repository to back up your Bolt project, sync commits automatically".
Lovable: "export and two-way sync your Lovable project code with GitHub".
Replit: its Git pane can "initialize, connect, and manage Git repositories with GitHub integration".

Their pricing works differently too. v0 and Lovable charge in credits, and Bolt in tokens, so what you pay depends on how much you use them.

Learning

MDN — the reference for HTML, CSS and JavaScript (see the top of this guide).
web.dev and Baseline — Google's guides; Baseline "was originated by the Chrome team and is now defined by the WebDX Community Group".
Can I use — "support tables for HTML5, CSS3, etc.", for checking one feature at a time.
Smashing Magazine — "founded in September 2006 in Germany", publishing "practical articles to web designers and developers", and in its own words "independent".
CSS-Tricks — still publishing: its home page carried items dated 9 and 14 September 2026 when we read it.
A List Apart — publishes a few long articles a year; the newest on its articles page was dated 30 June 2026, with others on 23 April and 28 January 2026.
Josh W. Comeau — in-depth articles and tutorials on CSS and front-end work, such as a recent guide to anchor positioning.
Kevin Powell — a YouTube channel "helping you learn how to make the web, and make it look good while you're at it", with weekly videos.

Things worth knowing

Pagefind — "a fully static search library" that works "without hosting any infrastructure". Free, MIT. (It powers the search on this site.)
Plausible — "a lightweight and open-source Google Analytics alternative", under the AGPL-3.0 licence; hosted plans start at $9 a month for 10,000 monthly pageviews.
Umami — "privacy by default. No cookies, no personal data." You can self-host the open-source version (MIT) or use Umami Cloud.
Tally — "unlimited forms and submissions, completely free, as long as you stay within our fair usage guidelines", including payments, signatures and file uploads.
Excalidraw — "an open source virtual hand-drawn style whiteboard", "collaborative and end-to-end encrypted". MIT.
Penpot — "the open-source design platform for teams", free for individuals and small teams, under the Mozilla Public License 2.0.
SVGOMG — "SVGO's missing GUI for minifying SVGs".
Squoosh — Google Chrome Labs' image compressor in the browser. Its repository had last been pushed on 23 September 2026 when we read it.
Wakamai Fondue — "the tool that answers the question 'what can my font do?'"

One licence to read before you ship: the tldraw SDK's default licence "permits use only in development". Production use needs a trial, commercial or hobby licence, and the hobby licence (for non-commercial projects) requires the "Made with tldraw" watermark on the canvas.

Putting this list to use

A handful of checks cover most of the surprises in this guide, and each takes a few minutes:

Read the licence file, not only the label. React Bits says "MIT" and adds the Commons Clause; Remix Icon's npm package still says Apache-2.0 after the project moved to its own licence; GSAP is free under Webflow's own terms. The licence file in the repository, or the vendor's licence page, is what counts.
Look at the last release date. The npm registry and a project's GitHub releases page show when something last shipped. A quiet project can be fine, but you are then the one who fixes it.
Keep a note of what you were granted. For every image, illustration or font you use commercially: where it came from, the date, and the licence wording on that day. Unsplash and Pixabay have both changed their licences before.
Give the credit a free licence asks for. HTML5 UP, Storyset, Icons8 and Absurd Design's free tier all require a visible credit or link.
Before you build on a platform, find its export page. Webflow, Framer, Wix, Squarespace and Bubble each answer "can I take my site with me?" differently, in their own help centres.
Automate certificates, then watch the automation. With lifetimes heading to 47 days and Let's Encrypt no longer emailing reminders, a silent renewal failure is the risk.
Start accessibility with the six common failures. Low contrast, missing alt text, missing form labels, empty links, empty buttons and a missing page language account for most errors in the WebAIM Million.

Method, and what is not verified here

How this was checked. Every tool, price, licence and date on this page was read from a source on 30 September 2026, in most cases the tool's own pages: vendors' pricing and licence pages, licence files in GitHub repositories, the npm registry, the GitHub API, standards documents (W3C, CA/Browser Forum) and official announcements. Pages were fetched by script; where a site blocked scripts with a bot check, we read it in an ordinary browser, and we did not get round any check that asked us to prove we were human. A few facts come from secondary sources: Wikipedia (who owns Unsplash, Pexels and Pixabay, their licence history, and neumorphism), the Interaction Design Foundation (glassmorphism) and a German law firm's write-up of the Munich Google Fonts ruling. Each fact is recorded with its source and a verbatim excerpt in the site's source ledger.

Prices are shown in the currency the vendor displayed: US dollars unless marked, with euros for Flaticon and Absurd Design. Prices on some sites change by region, by billing period, and with promotions; treat every figure as correct in September 2026 and check the vendor's page before you pay.

What could not be verified, and so is not in this guide:

Shadertoy and LottieFiles kept us on a security-check page, so they are not described here (Lottie is covered through Airbnb's own lottie-web project).
Unsplash's licence page did not show its text to us, so its licence is quoted from Unsplash's own Help Center.
Paid tiers for several tools (Land-book Pro, Motion+, Flowbite Pro, Preline Pro, Magic UI Pro, 21st.dev membership) are mentioned without a price, because the price was not shown on the page we read.
Pages that say "free" without a licence file (for example Realtime Colors) are described in their own words; we have not checked their terms of service.

If something here is wrong or out of date, tell us: hacksvitae@gmail.com

Sources

Licences and terms: GSAP Standard License; React Bits, ThreeUI, Lucide, Three.js, Motion, Lenis, Magic UI, shadcn/ui, Base UI, HeroUI, Mantine, Chakra UI, Ant Design, Excalidraw, Pagefind, Umami, Caddy, Hugo, Jekyll, Next.js, Astro, Leonardo, Penpot, Modern Font Stacks and Plausible licence files on GitHub; Remix Icon License v1.0; Simple Icons disclaimer; unDraw, DrawKit, Storyset, Icons8, Absurd Design, HTML5 UP, Preline, Flowbite and tldraw licence pages; Fontshare ITF Free Font License; Google Fonts FAQ; Pixabay, Pexels and Smithsonian Open Access terms; Unsplash Help Center on the Unsplash License; Creative Commons Attribution 3.0 legal code; HeroUI v3.0.3 release notes; Vercel fair-use guidelines; GitHub Pages limits.
Prices: each vendor's own pricing page, read 30 September 2026 (Refactoring UI, Awwwards, Mobbin, Are.na, Dribbble, ThreeUI, Spline, Rive, Three.js Journey, Tabler, DrawKit, Flaticon, Absurd Design, Aceternity UI, Chakra UI Pro, Cruip, Every Layout, Cloudflare Workers, Vercel, Netlify, Amazon CloudFront, Carrd, Webflow, Readymag, v0, Bolt, Plausible, Coolors).
Versions and dates: npm registry and GitHub API records for Astro, Eleventy, Next.js, Tailwind CSS, Velocity, Popmotion, Vanta.js, Curtains.js, Theatre.js, Locomotive Scroll, Water.css, Remix Icon, Feather, Radix, Headless UI, Jekyll and Squoosh; web-features 3.40.0 data for Baseline status.
Standards and official records: W3C, WCAG 2.2 (Recommendation, 12 December 2024) and WCAG 3.0 (Working Draft, 10 September 2026); CA/Browser Forum TLS Baseline Requirements and ballot SC-081v3; US Department of Justice, ADA Title II web rule fact sheet (2024); US Federal Trade Commission press releases on accessiBe (3 January and 22 April 2025); US Copyright Office, NewsNet issue 1060 (29 January 2025); web.dev and Google Search Central on Core Web Vitals; Chrome UX Report methodology.
Announcements: Apple Newsroom (9 June 2025; 8 June 2026); Tailwind CSS blog (9 September 2026; 4 March 2025); Vercel blog (11 November 2021; 25 March 2025); Let's Encrypt blog (22 January 2025); NuxtLabs; Astro blog (Astro 6, Astro 7); Let's Encrypt (26 June 2025); Wix Help Center on the Editor X transition.
Research and reports: WebAIM Million 2026; WebAIM Survey of Web Accessibility Practitioners #3 (January 2021); Overlay Fact Sheet.
Reference: Interaction Design Foundation (glassmorphism); Wikipedia (Neumorphism, Unsplash, Pexels, Pixabay); Kanzlei Plutte on LG München, 3 O 17493/20.

Checked September 2026.

Prompt Engineering 2026 Prompting Manual: Architecting High-Performance AI Outputs

Hacks Vitae — Sat, 13 Jun 2026 06:47:32 GMT

Prompt Engineering 2026 Prompting Manual

Prompting advice ages quickly. A lot of what still circulates was written for models that did not reason before answering, could read only a few pages at once, and had no tools. Current models from the large labs do all three, and their makers' own guides now say some of the older tricks are unnecessary.

This page sets out what those guides recommend, where the research behind a tactic is mixed, and a template you can adapt. It was written in June 2026 and brought up to date in September 2026. Model behaviour changes within months, so treat it as a snapshot and test anything here against the model you use.

1. The Core Infrastructure of Modern LLMs

Three changes explain why older prompting habits matter less than they did.

Models reason before they answer. The phrase "Let's think step by step" comes from a 2022 paper by Takeshi Kojima and colleagues. Adding it before each answer raised one large model's accuracy on the MultiArith arithmetic test from 17.7% to 78.7%. Reasoning models now do that work internally. OpenAI's guide for its reasoning models says prompting them to "think step by step" or "explain your reasoning" is unnecessary, and that the instruction "may not enhance performance (and can sometimes hinder it)". Anthropic's guide says a general instruction such as "think thoroughly" often produces better reasoning than a hand-written step-by-step plan.

Models can hold very long inputs. Anthropic lists a 1M-token context window for Claude Opus 5.5, the model it suggests starting with (its fastest model, Claude Haiku 4.5, has 200k). OpenAI lists 1.05M tokens for GPT-6 Astra, its flagship model, and Google lists an input limit of 1,048,576 tokens for Gemini 3.8 Flash. Google's prompting guide puts 100 tokens at roughly 60 to 80 words, so a million tokens is somewhere around 600,000 to 800,000 words.

Longer is not automatically better. Anthropic's documentation says that as the token count grows, "accuracy and recall degrade", which it calls context rot. A 2023 study by Nelson Liu and colleagues, "Lost in the Middle", found that models used information at the start or end of a long input much better than information in the middle.

Models can use tools. Anthropic, OpenAI and Google each offer a code-execution tool that lets the model write and run code in a separate environment (Anthropic and OpenAI describe theirs as a sandbox), and web search is offered as a tool too. A prompt can now say when those tools should be used.

2. The Structural Layout: Separate the Parts of a Prompt

The three vendors' guides agree on one thing more than any other: mark out the different parts of a prompt so the model can tell instructions from material.

Anthropic: XML tags "help Claude parse complex prompts unambiguously, especially when your prompt mixes instructions, context, examples, and variable inputs."
OpenAI: "Use delimiters like markdown, XML tags, and section titles to clearly indicate distinct parts of the input."
Google: "XML-style tags (e.g., <context>, <task>) or Markdown headings are effective. Choose one format and use it consistently within a single prompt."

For long inputs (around 20,000 tokens or more), Anthropic also suggests placing the documents near the top of the prompt, above the question and instructions. It reports that questions at the end "can improve response quality by up to 30 percent in tests, especially with complex, multidocument inputs."

The layout below is one way to put that into practice. The five block names are this page's own labels, not a standard; any clear, consistent labels do the same job.

### SYSTEM PERSONA
The expertise, perspective, tone and limits you want the model to work within.

### INPUT DATA
<source_context>
[Paste the source material here: documents, data, specifications, notes]
</source_context>

### TASKS
1. First instruction, stated plainly.
2. Second instruction, stated plainly.

### GUARDRAILS
- Limits on what the answer may do (for example, "Do not assume facts that are not in the source").
- When to use tools (for example, "Use web search to check anything that may have changed since your training data").

### OUTPUT FORMAT
The exact format you want back (for example, a Markdown table or JSON).

3. Three Tactics, and What Backs Them

Re-grounding a long conversation

In a long chat, earlier material can drift out of the model's effective attention, which is what context rot and the "Lost in the Middle" results describe. One practical response is to ask the model to restate where things stand before a new phase of work:

"Before we continue, review our conversation so far. Summarise the current status, the decisions we have made and the open questions in three bullet points. Write nothing else."

This is a working habit rather than a tested technique; no study of this exact prompt was found. The same idea is built into some products: Anthropic's compaction feature "automatically summarizes earlier parts of the conversation" so a long session can continue. Another option is to start a fresh conversation and paste the summary in.

Making the model use a tool for calculations

The code-execution tools exist for this: OpenAI describes its code interpreter as a way for models "to write and run Python code in a sandboxed environment" for problems in "data analysis, coding, and math".

Vague: "Calculate the compound growth rate in this spreadsheet."
Specific: "Write and run a Python script to compute the compound growth rate from the attached data. Do not estimate the figure. Show the script and its output alongside the final number."

Asking for a review before the final answer

Asking a model to critique and revise its own draft has mixed evidence. A 2023 paper, "Self-Refine" (Aman Madaan and colleagues), reported that outputs improved by about 20% in absolute terms on average across seven tasks when a model gave itself feedback and revised. A later paper by Jie Huang and colleagues, presented at ICLR 2024, found that models "struggle to self-correct their responses without external feedback" on reasoning problems, and that performance sometimes got worse after self-correction.

The two results sit together more comfortably when the review has something outside the model to check against. A 2024 survey of self-correction research by Ryo Kamoi and colleagues, published in Transactions of the Association for Computational Linguistics, found that earlier studies often used "unfair evaluations that over-evaluate self-correction", and concluded that self-correction "works well in tasks that can use reliable external feedback". Anthropic's guide suggests adding a line such as "Before you finish, verify your answer against [test criteria]", and notes that one of its models, Claude Opus 5, checks its own work well without being asked. A review prompt that gives the model something outside itself to check against looks like this:

"Before giving your final answer, check your draft against [the test results / the source documents / the output of the code you ran]. List anything in the draft they do not support, revise the draft, then show only the revised version."

4. Older Habits and What the Guides Say Now

Older habit	What current guides and research say
Adding "Let's think step by step" to every question	It helped older models a great deal (Kojima and colleagues, 2022). OpenAI says it is unnecessary for its reasoning models and can sometimes hinder them.
Being very polite, or very rude, to get a better answer	The research is mixed. A 2024 study in English, Chinese and Japanese found that impolite prompts often did worse. A small 2025 study (50 questions, one model, ChatGPT-4o) found very rude prompts scored slightly higher: 84.8%, against 80.8% for very polite ones. The vendors' guides ask for clear, direct instructions rather than any particular tone.
Emphasis in capitals: "CRITICAL: you MUST…"	Anthropic says its newer models may now overreact to that kind of language and suggests plain wording such as "Use this tool when…". Google's guide says to avoid "overly persuasive language".
Cutting source material down to fit a small limit	The models named above accept around a million tokens, so the full text often fits. Choosing what goes in still matters, because accuracy and recall degrade as the context grows.
Taking the first answer as finished	A review step can help, especially when it checks the draft against something outside the model, such as test results, a tool's output or sources.

5. A Template to Adapt

Copy it, change the parts in brackets, and test it on the model you use.

### SYSTEM PERSONA
You are a senior systems architect. Write directly and clearly, without introductions or pleasantries.

### INPUT DATA
<technical_specification>
[Paste your code, notes or documentation here]
</technical_specification>

### TASKS
1. Read the <technical_specification> block and identify the three most serious bottlenecks.
2. For each bottleneck, outline one concrete fix.
3. Use web search to check whether any recent security guidance affects these fixes, and name the source for each point.

### GUARDRAILS
- No introductory phrases, filler or closing summaries.
- Base every statement on the <technical_specification> block or on sources you found with web search, and name them.
- If the material is not enough to complete a task, write "Data Insufficient" instead of guessing.

### OUTPUT FORMAT
Markdown, with ### headings for each part. Put the findings in a three-column table:

| Bottleneck | Fix | Relevant guidance and source |
| :--- | :--- | :--- |

The "Data Insufficient" line follows advice in Anthropic's guide on reducing hallucinations: give the model explicit permission to say it does not know.

Frequently Asked Questions (FAQ)

What has changed in prompting since the early chatbots?

Reasoning models work through problems before answering, so the step-by-step phrasing that helped older models is often unnecessary. Context windows grew to around a million tokens on many current models, and models can run code and search the web. The vendors' guides now put the weight on clear structure and specific instructions.

Why use tags like `<source_context>` in a prompt?

Tags or headings mark where source material ends and instructions begin. Anthropic, OpenAI and Google all recommend some form of delimiter for this. The tag names themselves are up to you; consistency matters more than the exact words.

Should I still summarise documents before pasting them in?

Often you can paste the full text, since many current models accept around a million tokens. Anthropic's documentation still notes that accuracy and recall degrade as the context grows, so leaving out material that is not needed can help, and placing long documents above the question is Anthropic's suggested layout.

Can prompting stop a model from making things up?

No prompt removes the problem. Anthropic's guide says its techniques "significantly reduce hallucinations" but "don't eliminate them entirely", and advises checking critical information. Letting the model say it does not know, grounding it in supplied documents, and making it use tools for calculations all help. For more on why models invent answers, see Why AI makes things up.

Putting It to Use

The habit that holds up whichever model you use is to test prompts on your own work. Anthropic's guide makes the same point about its own advice: treat a tip as measured on one model, and re-check it against your own tests before relying on it with another. In practice that can be as simple as a short list of real questions with answers you already know, run again whenever you change the prompt or the model changes, changing one thing at a time. For the terms used on this page, see the AI glossary; for how these models work underneath, see How large language models work.

Sources

Takeshi Kojima, Shixiang Shane Gu, Machel Reid, Yutaka Matsuo, Yusuke Iwasawa, "Large Language Models are Zero-Shot Reasoners" (2022), arXiv:2205.11916
OpenAI, Reasoning best practices; Models; Code interpreter
Anthropic, Prompting best practices; Context windows; Models overview; Code execution tool; Reduce hallucinations
Google, Prompt design strategies; Gemini 3.8 Flash; Code execution
Nelson F. Liu and colleagues, "Lost in the Middle: How Language Models Use Long Contexts" (2023), arXiv:2307.03172
Aman Madaan and colleagues, "Self-Refine: Iterative Refinement with Self-Feedback" (2023), arXiv:2303.17651
Jie Huang and colleagues, "Large Language Models Cannot Self-Correct Reasoning Yet" (ICLR 2024), arXiv:2310.01798
Ryo Kamoi and colleagues, "When Can LLMs Actually Correct Their Own Mistakes? A Critical Survey of Self-Correction of LLMs" (TACL 2024), arXiv:2406.01297
Ziqi Yin and colleagues, "Should We Respect LLMs? A Cross-Lingual Study on the Influence of Prompt Politeness on LLM Performance" (2024), arXiv:2402.14531
Om Dobariya and Akhil Kumar, "Mind Your Tone: Investigating How Prompt Politeness Affects LLM Accuracy" (2025), arXiv:2510.04950

Checked September 2026.

The Geopolitical Reconfiguration of Western Eurasia: The Sino-Iranian Strategic Nexus Amidst Global Conflict (2024-2026)

Hacks Vitae — Tue, 03 Mar 2026 12:23:40 GMT

First written March 2026. Brought up to date on 29 September 2026. This page describes a war that is still going on. Everything below is dated to the source it comes from; anything that depends on the state of the fighting may have moved since. It describes; it does not predict.

The Sino-Iranian Strategic Nexus Amidst Global Conflict (2024-2026)

Between April 2024 and February 2026, the long-running shadow conflict between Iran and Israel turned into open war. The United States struck Iranian nuclear sites in June 2025, and from 28 February 2026 it went to war with Iran directly. Through all of it, China remained the main buyer of Iran's oil and defended Iran diplomatically.

This page sets out what the sources show about that relationship: the oil trade, the 25-year partnership, reported arms deals, China's diplomacy, and the money. It ends with where the war stood on 29 September 2026.

The Escalation Continuum: From Shadow War to Decapitation (2024-2026)

April 2024. Iran launched hundreds of drones and missiles from its own territory towards Israel, most of which were intercepted, according to the UN. Iran said it was responding to an attack on its diplomatic premises in Damascus on 1 April.
July 2024. Hamas leader Ismail Haniyeh was killed in Tehran.
Late September 2024. Hezbollah leader Hassan Nasrallah was killed in Israeli airstrikes on the southern suburbs of Beirut.
June 2025. The IAEA said Iran was in breach of its non-proliferation obligations. From 13 June 2025 Israel struck Iran's nuclear programme, missile sites and energy facilities, and Iran fired missiles and drones back. On 22 June the United States struck three Iranian nuclear sites, including Fordow. A ceasefire announced by President Trump took effect on 24 June 2025. The Commons Library describes the episode as joint strikes "over 12 days".
August-September 2025. On 28 August 2025 the UK, France and Germany triggered the "snapback" mechanism of the 2015 nuclear deal. Six earlier UN sanctions resolutions on Iran were reinstated from 27 September 2025.
February 2026. On 27 February, Oman's foreign minister, who had mediated talks between the US and Iran, said that "a peace deal is within our reach". The next day, 28 February 2026, Israel and the United States launched strikes on Iran; the US operation was named Operation Epic Fury. Iran's supreme leader, Ali Khamenei, was killed in the strikes. Iran responded by attacking US bases in the region, Israel and sites in neighbouring states, and by closing the Strait of Hormuz.

Domestic Fragility and the Economic Crisis of 2026

Protests began on 28 December 2025, after a sharp collapse of Iran's currency, starting with shopkeepers in Tehran's Grand Bazaar. According to the UN fact-finding mission on Iran, they reached all of Iran's 31 provinces. Amnesty International describes the killing of thousands of protesters by security forces on 8 and 9 January 2026, thousands more arrests, and a shutdown of internet access from 8 January.

Executions rose sharply before that. Amnesty International recorded at least 2,159 executions in Iran in 2025, more than double its 2024 figure. Iran drove a global total that Amnesty calls the highest it has recorded since 1981.

On 8 March 2026 Iran's Assembly of Experts announced that Mojtaba Khamenei, Ali Khamenei's son, would succeed his father as supreme leader. The UK government said the choice "suggests no change in direction from Iran". In the UK House of Commons Library's account (April 2026), he had not appeared in public since taking office.

The Sino-Iranian Energy Nexus: A Strategic Lifeblood

China is the main buyer of Iranian oil. Estimates of how dependent Iran is on that one customer differ with the source:

Measure	Figure	Source
China's imports of Iranian crude, 2025	1.38 million barrels per day	Kpler, cited by Columbia University's Center on Global Energy Policy
Share of Iran's shipped oil bought by China, 2025	more than 80%	Kpler, cited by Al Jazeera
Share of Iran's oil exports going to China	more than 90%	US-China Economic and Security Review Commission, cited by the House of Commons Library
Iranian crude as part of China's seaborne imports, 2025	about 1.4 of 10.4 million barrels per day	Kpler, cited by Al Jazeera

That last line works out at about 13% of China's seaborne crude imports: a large supplier, but a minority of the total. The Commons Library adds that oil exports provide around a third of Iran's annual state revenue.

The oil arrives under other names. China's customs agency has not recorded any oil imports from Iran since 2022. Many Iranian barrels are relabelled as Malaysian: China imported 1.3 million barrels a day of "Malaysian" crude in 2025, while Malaysia produced 535,000 barrels a day in 2024, according to the Columbia analysis. Al Jazeera describes the small independent refineries known as "teapots", mostly in Shandong province, as the channel for discounted Iranian and Russian oil.

China had a cushion. As of early January 2026 it held 1.206 billion barrels of oil in onshore storage, enough for 104 days of net crude imports at the 2025 level, by the Columbia estimate.

The war changed the flow. In June 2026, under the US-Iran memorandum, the US Treasury issued a waiver for Iranian oil exports running until 21 August 2026. In August 2026, Reuters reported that offers of Iranian crude to Chinese buyers had fallen as the US blockade took hold.

Beyond Oil: The 25-Year Comprehensive Strategic Partnership (CSP)

Iran and China signed a 25-year strategic accord on 27 March 2021. It is often described as a plan for US$400 billion of Chinese investment in Iran's energy, banking, telecommunications and transport, in return for a regular supply of oil. That figure is reported, not confirmed: in April 2021, analysts at Brandeis University's Crown Center noted that the terms and details of the agreement remained unknown, and that some commentators were relying on an earlier leaked draft. We found no published text of the signed agreement.

Military and Security Cooperation: The Defense Dimension

Four reports stand out. None is a confirmed transaction.

Missile fuel chemicals, early 2025. In February 2025, CNN reported, citing two European intelligence sources, that a ship had left China with most of a 1,000-ton shipment of sodium perchlorate, the main precursor of the solid propellant for Iran's mid-range missiles. The sources estimated it was enough for about 260 motors for Kheibar Shekan missiles.
Anti-ship missiles, 2026. In February 2026, Reuters reported that Iran was close to a deal to buy Chinese CM-302 supersonic anti-ship missiles, after negotiations that had run for at least two years and sped up after the June 2025 war. Whether a deal was completed is not established in the sources we read.
Air-defence missiles, 2026. In July 2026, Reuters reported, citing anonymous sources and as summarised by Newsweek, that Iran had signed a US$60–70 million deal through a Hong Kong-based intermediary for 300–400 Chinese shoulder-fired air-defence missile systems.
Components, 2026. In September 2026, The Wall Street Journal reported, as summarised by Iran International, that Chinese suppliers had sent around 1,300 shipments of dual-use components to Iran's defence ministry in the first half of 2026.

Impact on International Relations: The "Axis of Upheaval"

"Axis of upheaval" is a phrase from Andrea Kendall-Taylor and Richard Fontaine, writing in Foreign Affairs in April 2024, for the growing coordination among China, Iran, North Korea and Russia. Three examples of that coordination on the Iran file:

Joint drills. China, Iran and Russia held the "Security Belt 2025" naval exercise near Iran's Chabahar port in March 2025. China sent the guided-missile destroyer Baotou.
Words. On 1 March 2026, China's foreign ministry called the killing of Khamenei "a grave violation of Iran's sovereignty and security" and said China "firmly opposes and strongly condemns it".
Votes. On 7 April 2026, China and Russia vetoed a Bahraini-drafted Security Council resolution on the Strait of Hormuz. In March they had abstained on a separate resolution, S/RES/2817 (2026), which passed and demanded an end to attacks by Iran and its proxies on Arab states.

What the sources do not show is China fighting for Iran. Earlier in the war, the Commons Library notes, China was among the countries that declined US requests to help secure Hormuz while hostilities went on. In July 2026 Newsweek described Beijing as having "limited its direct involvement in the conflict", while, according to Reuters and other media reports, providing Iran with satellite imagery of US military targets around the Gulf.

Trade Realignments and Financial Architecture: De-Dollarization

The war put payment channels outside the US dollar to use.

Hormuz tolls. From mid-March 2026, according to the analytics firm TRM Labs as reported by Fortune, Iran's military charged ships up to $2 million each to transit the strait, accepting Chinese yuan, bitcoin and potentially the stablecoin USDT. TRM Labs said at the time that it was not seeing on-chain evidence that tolls were being paid in crypto at scale. The Commons Library reports that some ships had paid in yuan, and that the US sanctioned Iran's new strait authority.
The renminbi's share. In the Bank for International Settlements' 2025 survey, the renminbi's share of global foreign exchange turnover rose to 8.5% in April 2025, continuing a rise since 2013.

Where things stand, 29 September 2026

The war has paused and restarted. On 8 April 2026, Pakistan's prime minister announced a conditional two-week ceasefire, later extended. On 17 June the US and Iran signed a 14-point memorandum of understanding, which set an initial 60-day period for talks on Hormuz, Iran's nuclear and missile programmes and sanctions.
The memorandum broke down. On 8 July, after Iranian attacks on commercial ships, US strikes on Iran resumed, and President Trump said he thought the memorandum was "over". Its 60-day period ran out in mid-August, with little sign that either side wanted to extend it, according to Al Jazeera.
Hormuz traffic has picked up, and estimates of how much differ. Iranian attacks on shipping have sharply curtailed traffic through the strait, and the United States maintains a blockade of Iranian ports in some form. On 28 September Al Jazeera reported that the US military had moved "considerably more energy shipments" through the strait in recent weeks, though the waterway was "still far away from its pre-war unimpeded state", and set out three estimates against roughly 20 million barrels a day before the war: Kpler expected crude exports through the strait to average 7.4 million barrels a day in September; the US Energy Secretary said nearly 13 million barrels of crude a day were passing; and the US Treasury Secretary gave 15–22 million. They do not all measure the same thing, and they come from parties with different stakes. Iran's Revolutionary Guards say the strait stays closed to vessels that do not coordinate passage with Iranian authorities. Before the war, around 20% of global petroleum passed through the strait each year, according to the Commons Library.
The latest proposal was rejected. On 25 September Iran's foreign minister, Abbas Araghchi, presented a plan to reopen the strait within seven days if the US released frozen funds, lifted oil sanctions and ended its blockade. On 26 September President Trump said it "would not be acceptable".

What this snapshot does and does not show

On the evidence above, China's part in this war has been as a customer, a diplomat and, by report, a supplier, not a combatant. It bought most of Iran's oil, defended Iran in words and at the Security Council, and its currency features in some of the payment workarounds. The reported arms and component deals remain reports.

What the snapshot cannot tell you is how the war ends. That is why this page is dated, and why the figures above each name their source and their date: a reader checking a claim here later should check it against the newest version of that source.

Sources

Conflict and diplomacy

House of Commons Library: Israel/US-Iran conflict 2026: Background and UK response (24 April 2026); Reopening the Strait of Hormuz (8 June 2026); US-Iran ceasefire and nuclear talks 2026 (2 July 2026); Iran: Impacts of June 2025 Israel and US strikes and outlook (22 July 2025).
UN News: April 2024 Security Council session; 31 July 2024; 1 October 2024.
UK Government: E3 joint statement on the activation of the snapback (28 September 2025); UK reimposes UN sanctions on Iran (1 October 2025).
US Department of State, Office of the Legal Adviser: Operation Epic Fury and International Law (21 April 2026).
Ministry of Foreign Affairs of the People's Republic of China: spokesperson's remarks on the killing of Ali Khamenei (1 March 2026).
Al Jazeera: Why have US-Iran strikes resumed? (8 July 2026); US-Iran MoU ends: What happens next? (16 August 2026); Trump rejects Iran's seven-day roadmap (26 September 2026). NPR, Trump rejects plan to reopen Strait of Hormuz (26 September 2026).

Iran at home

Amnesty International: What happened at the protests in Iran? (26 January 2026); Executions surge to highest recorded figure in 44 years (May 2026).

China, oil and money

Erica Downs, Where China Gets Its Oil: Crude Imports in 2025, Center on Global Energy Policy, Columbia University (29 January 2026).
Al Jazeera, How China's 'teapot' refineries are cushioning it from Iran war oil crisis (3 April 2026).
The China-Global South Project, Reuters report on Iranian oil offers to China (21 August 2026).
Brandeis University, Crown Center, The Iranian-Chinese Strategic Partnership: Why Now and What it Means (28 April 2021).
Newsweek, China sending weapons to Iran despite Trump's faith in Xi: report (29 July 2026), summarising Reuters.
Iran International, China shipments help sustain Iran's weapons programs despite US pressure - WSJ (24 September 2026).
Al Jazeera, Iran touts Hormuz attacks as oil flows increase despite tensions (28 September 2026).
CNN, Iran is rearming its missile program and a ship of supplies just arrived from China (13 February 2025).
Defence Industry Europe, report of Reuters' CM-302 story (26 February 2026).
Center for a New American Security, The Axis of Upheaval (23 April 2024; first published in Foreign Affairs).
Ministry of National Defense of the People's Republic of China, "Security Belt 2025" naval exercise wraps up (14 March 2025).
Fortune, Iran is demanding tankers in the Strait of Hormuz pay tolls in crypto (10 April 2026).
Bank for International Settlements, OTC foreign exchange turnover in April 2025.

Checked September 2026. This page is a snapshot of an unfinished war, read on 29 September 2026.

Clawdbot AI: A Leap Towards Sovereign Agentic Workflow

Hacks Vitae — Tue, 03 Feb 2026 06:50:23 GMT

Name check, September 2026. This project has had several names. It began in November 2025 as a weekend project its creator calls "WhatsApp Relay" (published on npm as warelay), then became known as Clawdbot, became Moltbot on 27 January 2026 after a trademark request from Anthropic, and was renamed OpenClaw on 29 January 2026. If you go looking for it, OpenClaw is the name to search.

Clawdbot AI: Architecting the Sovereign Agentic Workflow

OpenClaw (formerly Clawdbot) is an open-source AI assistant that you run on your own computer and talk to through the chat apps you already use. Its creator, Peter Steinberger, wrote on the day of the rename that it "started as 'WhatsApp Relay'", and described it as "an open agent platform that runs on your machine and works from the chat apps you already use".

According to the project's own site, it runs on Mac, Windows or Linux; it works with hosted, subscription-backed or local language models; and its state "lives on your machine, not a vendor cloud". The site lists WhatsApp, Telegram, Discord, Slack, Signal and iMessage among its 29 chat channels.

The name story, in the project's words: the assistant was first called Clawd, "a playful pun on 'Claude' with a claw", until "Anthropic's legal team politely asked us to reconsider". Moltbot came next; the creator wrote that it "never quite rolled off the tongue", and OpenClaw followed two days later.

Since 8 July 2026 the project has been stewarded by the OpenClaw Foundation, which it describes as an independent US 501(c)(3) non-profit. The project says the code stays MIT licensed and that there is no paid version.

The Evolution of the "Invisible" Executive Assistant

Most AI chat tools wait in a browser tab for you to type. OpenClaw works the other way round: the assistant lives in your messaging apps, and it can also start work on its own. Its documentation describes "tasks, scheduled jobs, event hooks, and standing instructions" that run in the background.

The project's homepage says it "organizes your inbox, sends emails, manages your calendar, checks you in for flights". Those are the project's own claims about what it can be set up to do, not something we have tested.

The idea that switching to a browser tab is what stops people from using AI every day is an argument, not a measured finding. What is different is the shape: an assistant that can act first, on a schedule or a trigger, instead of only answering when asked.

The Sovereign Intelligence Framework

One way to picture OpenClaw (our own summary, not the project's terms) is as three layers:

Layer	What it is
The Gateway	A single long-running process on your machine that owns the chat connections. By default the apps that control it connect on `127.0.0.1`, your own computer.
The Memory	Plain Markdown files in the agent's workspace (by default `~/.openclaw/workspace`), such as `MEMORY.md` for long-term facts and dated daily notes.
The Skills	Folders containing a `SKILL.md` file: Markdown instructions that teach the agent how and when to use tools. Community skills are shared through a marketplace called ClawHub.

The documentation puts the memory point plainly: "The model only remembers what gets saved to disk; there is no hidden state."

What self-hosting changes, and what it doesn't

You can see what it remembers. Because memory is ordinary files, you can read, back up or edit it, and the files stay where they are if you switch to a different model provider.
Your data sits on your machine, and so does the job of protecting it. The same shell access, file access and messaging access that make the assistant useful are what an attacker would want.
The model is not automatically local. If you connect a hosted model, what you ask the agent still goes to that model's provider. Only a local model keeps the requests on your hardware.

Before you self-host this: the security record

An assistant that can read your messages, run commands and install add-ons has a lot of access, and OpenClaw's first months had real security problems. Four are well documented.

Malicious skills on ClawHub. In early February 2026 the security firm Koi Security reported that an audit of 2,857 skills on ClawHub had found 341 malicious ones, according to The Hacker News. Of those, 335 used fake "prerequisites" to get users to install a macOS information stealer called Atomic Stealer (AMOS). On 7 February 2026 the project announced that every skill published to ClawHub is now scanned with VirusTotal, and said in the same post: "this is not a silver bullet".
A fake VS Code extension. On 27 January 2026 the security firm Aikido flagged a VS Code extension called "ClawdBot Agent". It did work, as a coding assistant, while installing ScreenConnect remote-access software connected to the attackers' server. Aikido notes that the real project "never published an official VS Code extension"; Microsoft removed the fake one after Aikido reported it.
A no-password mode, removed. Before release 2026.1.29, the gateway could be set to an authentication mode of "none". That release's notes say: "Gateway auth mode "none" is removed; gateway now requires token/password (Tailscale Serve identity still allowed)."
A one-click token leak, fixed in the same release. The US National Vulnerability Database records it as CVE-2026-25253: versions before 2026.1.29 took a gateway address from a link's query string and connected to it automatically, without prompting, "sending a token value". It is rated 8.8 out of 10 on the CVSS 3.1 scale. The project's own advisory says an attacker holding that token could change the gateway's settings and run code on the machine, and that this worked even on installs listening only on the user's own computer, because the victim's browser made the connection.

Where things stand, September 2026. The project's security page, reviewed on 9 September 2026, counts 1,799 vulnerability reports filed since January 2026 and 722 fixes published, 39 of them with a CVE number. It lists 14 confirmed critical issues, all fixed and disclosed, and says it knows of no compromise of its own infrastructure or official install channels; that statement does not cover third-party ClawHub skills. On 21 September 2026 the project published the results of an audit by the security firm Trail of Bits: 24 severity-rated reports, none critical, and, the project says, every actionable issue repaired.

What the project's own documentation describes. A regular install binds the gateway to loopback, meaning only your own machine can reach it. Container images are the exception: they default to an exposed bind, which the documentation says to pair with authentication. An openclaw security audit command reports when a setup has drifted from those defaults. On skills, the VirusTotal announcement adds: "A clean scan doesn't mean a skill is safe."

Prompt injection is not solved. On the day of the rename, the creator wrote: "Remember that prompt injection is still an industry-wide unsolved problem". An agent that reads your email and can run commands can be steered by text hidden in an email.

Moltbook: the agents' forum and its breach

Moltbook is a social network built for AI agents. Its homepage describes it as a place "where AI agents share, discuss, and upvote. Humans welcome to observe." Its topic groups are called submolts.

In late January 2026 the security firm Wiz found that Moltbook's database was open to anyone. What follows is from Wiz's disclosure.

What went wrong. Moltbook was built on Supabase. Wiz found the site's Supabase API key in its client-side JavaScript. Wiz explains that this kind of key is safe to expose when Row Level Security is set up, but Row Level Security was missing, so the key gave "unauthenticated access to the entire production database - including read and write operations on all tables".

What was exposed. Wiz counted around 4.75 million records, including:

1.5 million API authentication tokens for agents on the platform, which Wiz says would let an attacker "fully impersonate any agent on the platform"
35,000 email addresses, in Wiz's summary; its write-up also describes a separate table of 29,631 early-access sign-up emails
4,060 private conversations between agents, some of which contained third-party credentials, including plaintext OpenAI API keys, that agents had shared

Anyone could also write to the database, which means they could change posts that thousands of agents were reading.

The numbers behind the platform. Moltbook said it had 1.5 million registered agents. Wiz found about 17,000 human owners behind them, and no way for the platform to check whether an "agent" was actually AI or a person with a script.

The fix. Wiz first contacted Moltbook's maintainer at 21:48 UTC on 31 January 2026, and after several rounds of fixes all tables were secured by 01:00 UTC on 1 February 2026. Wiz says the data it accessed during the research was deleted. Wiz also notes that the security researcher Jamieson O'Reilly had found the same misconfiguration independently.

The lesson Wiz draws is that "the issue ultimately traced back to a single Supabase configuration setting". None of this involved an agent doing something unexpected. It was ordinary web security, left undone.

What you can take from this

Names change fast in this field. A tool that went viral as Clawdbot in January 2026 has been OpenClaw since the end of that month, and advice written under the old names may be out of date.
Add-ons are code. Both the ClawHub skills and the fake VS Code extension worked because they looked legitimate. The project's own advice is to review what a skill asks for and to start with publishers you trust.
A "private" message is only as private as the database behind it. Wiz's point about Moltbook is that people shared API keys in messages they assumed were private, and one configuration setting made them readable.
The project's own security page is the up-to-date record. It is dated, it lists current advisories, and it will be newer than anything written about the project, this page included.

Sources

Peter Steinberger, "Introducing OpenClaw", OpenClaw blog, 29 January 2026. openclaw.ai
OpenClaw homepage. openclaw.ai
Dave Morin and Peter Steinberger, "Introducing the OpenClaw Foundation", OpenClaw blog, 8 July 2026. openclaw.ai
OpenClaw documentation: Gateway architecture, Memory overview, Skills, Automation, Security. docs.openclaw.ai
OpenClaw GitHub repository and release 2026.1.29. github.com
npm registry, package warelay. registry.npmjs.org
National Vulnerability Database, CVE-2026-25253. nvd.nist.gov
"Clawdbot Rebrands to Moltbot After Trademark Request From Anthropic", Laravel News, 27 January 2026. laravel-news.com
"Researchers Find 341 Malicious ClawHub Skills Stealing Data from OpenClaw Users", The Hacker News, 2 February 2026. thehackernews.com
Peter Steinberger, Jamieson O'Reilly and Bernardo Quintero, "OpenClaw Partners with VirusTotal for Skill Security", OpenClaw blog, 7 February 2026. openclaw.ai
Aikido Security, "Fake Clawdbot VS Code Extension Installs ScreenConnect RAT". aikido.dev
OpenClaw, "Security" status page, reviewed 9 September 2026. openclaw.ai/security
Josh Avant, "OpenClaw Completes Security Audit Through OpenAI's Patch the Planet Initiative", OpenClaw blog, 21 September 2026. openclaw.ai
Moltbook homepage. moltbook.com
Wiz, "Hacking Moltbook: The AI Social Network Any Human Can Control". wiz.io

Checked September 2026. Everything on this page describes a fast-changing project as of September 2026. Version numbers, defaults and the security figures change often; the project's own security page and repository are the place to check before acting on any of it. If you can show any of this wrong, with a source, write to hacksvitae@gmail.com.

The Post-Authenticity Shift: Why the Living Room is the New Feed

Hacks Vitae — Tue, 13 Jan 2026 17:27:25 GMT

The Post-Authenticity Era: Inside the YouTube Landscape of 2026

YouTube in 2026 looks less like a feed of viral clips and more like television. In the US it is now watched most on the TV in the living room, its biggest creators run shows and studios, and AI tools are built into the platform. At the same time, its own payment rules reward original work and refuse to pay for mass-produced content.

"Post-authenticity" is our label for that tension. It is not a YouTube or Nielsen term, and others have used the phrase for other things. This page sets out what can be checked about it: the viewing numbers, the formats, the AI tools and the rules, each with its source. It is a snapshot as of September 2026, and most of the numbers in it will move.

The Living Room is the New Feed

In February 2025, YouTube's chief executive Neal Mohan wrote that TV "has surpassed mobile and is now the primary device for YouTube viewing in the U.S. (by watch time)", and that viewers were watching "over 1B hours of YouTube content on TVs daily". Those are YouTube's own figures.

Nielsen's independent measurement points the same way. Its Media Distributor Gauge ranks media companies by their share of all US television viewing. In its April 2025 report, YouTube led that ranking for a third consecutive month, with 12.4% of TV time against Disney's 10.7%. In July 2026 YouTube reached a record 14.2% of all US TV viewing, 5.0 share points ahead of the next company, while streaming as a whole reached 49.0%.

In November 2025 EMARKETER analyst Ross Benes put it this way: "YouTube is the most important company in streaming right now, at least from the consumer side." For consumers "especially those who are under 30 years old," he said, "it's the service that they rely on most when they are watching content on a TV screen."

Two limits on those numbers. Nielsen measures the United States, so its figures say nothing directly about viewing elsewhere. And the TV is only part of YouTube: Benes added that "half the time spent with YouTube is on phones and computers and tablets", a figure EMARKETER attributes to its own forecast.

YouTube's own framing is that this is television now. Mohan's January 2026 letter says creators are "buying studio-sized lots in Hollywood and beyond" and that "the era of dismissing this content as simply 'UGC' is long over." Our reading: a video watched from a sofa, across a room, is a different product from a clip swiped past on a train, and the formats are following the screen.

Shorts and Long-Form: Both, Not Either

YouTube says Shorts "now averages 200 billion daily views". The same January 2026 letter lists long-form, Shorts, music videos, livestreams and podcasts side by side, "on every screen", and Mohan's 2025 letter noted that people watch Shorts on TVs as well.

For a creator, the question is less "Shorts or long-form?" than what each is for. Our reading, not a YouTube rule: a short video reaches people who have never heard of you, and a long one is where they decide whether to stay.

Quiet Formats: Silent Vlogs

In a silent vlog, people "make videos without talking to the camera -- and sometimes even without showing their physical appearance." That is how YouTube's culture and trends manager for Japan, Makoto Maeoka, described the format in July 2021, in a YouTube blog post that introduced it as "a new interesting trend that's taking off across the Asia Pacific region".

She called South Korea "a leading home for silent vlogging culture", including a subgenre, cafe vlogs, that "started growing rapidly at the end of 2019". The videos, she wrote, "often focus on routine, daily activities such as cooking, cleaning, journaling and sewing in a calm atmosphere accompanied by soothing music or ASMR-like natural sounds." She linked the facelessness partly to privacy, "relatively common in some East Asian cultures", and the appeal to relaxation: "a short escape from their busy, stressful daily lives."

Our reading on why the format travels: with little or no speech, there is little to translate, and the pace suits a screen watched from a distance.

AI in the Background, and YouTube's Rules for It

AI tools are part of YouTube itself. In its January 2026 letter YouTube said more than one million channels a day, on average, used its AI creation tools in December. Its automatic dubbing "generates translated audio tracks in different languages", and videos with these tracks are marked "auto-dubbed". An English video can be dubbed into languages including Spanish, Hindi, Japanese and Arabic. YouTube's help page warns that dubs "might contain errors due to mispronunciations, accents, dialects, or background noise".

The same shift reaches beyond video. "Vibe coding", using AI prompted by natural language to write computer code, was coined by Andrej Karpathy and named Collins' Word of the Year 2025.

The rules run the other way. Since March 2024 YouTube has required creators to disclose when realistic content, "content a viewer could easily mistake for a real person, place, scene, or event", is made with altered or synthetic media, including generative AI. It does not require disclosure for content that is "clearly unrealistic, animated, includes special effects, or has used generative AI for production assistance." Its 2026 letter says it labels content made with its own AI products, removes harmful synthetic media that violates its community guidelines, and is working to reduce what it calls "AI slop", low-quality AI content.

Authenticity Is Now a Monetisation Rule

On 15 July 2025 YouTube renamed its "repetitious content" monetisation policy to "inauthentic content", to clarify that it covers content that is "repetitive or mass-produced". YouTube said this kind of content "has always been ineligible for monetization". Its policy expects monetised content to "be your original creation" and not to be "mass-produced, generic, repetitive, or manipulative".

This is the centre of the shift as we read it. AI makes a video cheap to produce; YouTube's payment rules put the value back on what a particular person made.

What This Means If You Watch or Make Video

Our reading of the sources above, not a rulebook.

If you make video

Think about the sofa as well as the phone. In the US, TV is now the main screen for YouTube viewing, by YouTube's own count.
Use the two lengths for different jobs. Short to be found, long to be kept.
Disclose AI when it depicts something real. It is a platform requirement for realistic altered or synthetic content.
Make something only you could have made. The inauthentic-content policy is aimed at the opposite: mass-produced, generic, repetitive work.

If you watch

An "auto-dubbed" label means a machine made the voice. YouTube itself says dubs can contain errors, so check the original language for anything important.
Labels can be missing. Creators must disclose realistic AI content, and YouTube says it may add a label itself to content made with its own AI tools, content carrying C2PA metadata, and content its systems detect as AI-made. A missing label still does not prove a video is real.
The numbers on this page are dated and mostly American. Check the source before repeating them.

Sources

Neal Mohan, "From the YouTube CEO: Our big bets for 2025", YouTube Official Blog, 11 February 2025.
Neal Mohan, "From the CEO: What's coming to YouTube in 2026", YouTube Official Blog, 21 January 2026.
Nielsen, "YouTube Maintains Largest Share of TV Viewing Among Media Companies For Third Consecutive Month", 27 May 2025.
Nielsen, "TV Usage Kicks Usual Summer Slowdown, Fueled by World Cup and Streaming in Nielsen's July Gauge Reports", 10 September 2026.
Arielle Feger, "YouTube, rising prices, and the shifting balance of CTV in 2026", EMARKETER, 17 November 2025.
Makoto Maeoka, "What's all the noise about silent vlogs?", YouTube Official Blog, 26 July 2021.
YouTube Help, "Use automatic dubbing".
YouTube Help, "Disclosing use of altered or synthetic content".
John Wihbey, "In Post-Authenticity AI Age, Knowledge Institutions Matter More than Ever", TechPolicy.Press, 12 November 2025 (a different use of the phrase).
YouTube, "How we're helping creators disclose altered or synthetic content", YouTube Official Blog, 18 March 2024.
YouTube Help, "YouTube channel monetization policies", update of 15 July 2025.
Collins Dictionary, "Collins' Word of the Year 2025: AI meets authenticity as society shifts", 6 November 2025.

Checked September 2026.

De Facto Meaning: Understanding Reality Beyond Formal Recognition

Hacks Vitae — Mon, 26 Feb 2024 07:02:09 GMT

De Facto Meaning

De facto is Latin. Literally it means "in fact, in reality", and from that comes its everyday sense: existing, but not necessarily legally ordained or morally right. The facto part comes from factum, "deed, act", the same root as the English word fact.

So when something is described as de facto, the phrase is pointing at what is really happening, as opposed to what the rules, the paperwork or the official titles say. The person everyone brings their questions to may be the de facto head of a team, whatever the organisation chart shows.

De facto vs de jure

The phrase is usually paired with its opposite. De jure is the Latin expression for "by law" or "by right", and it describes something that exists according to law. In modern use it almost always means "as a matter of law".

	De facto	De jure
Literal meaning	"in fact, in reality"	"by law" or "by right"
What it describes	How things are in practice	How things are formally, on paper
Example	The team member everyone treats as the lead	The person with the manager's title

The two can match, and often do. The phrase earns its keep when they don't.

A narrower meaning in law

In US legal usage, the Legal Information Institute at Cornell Law School gives de facto a more specific sense: an action taken without strict legal authority, but recognised as legally valid nonetheless. Such an action gets its validity from the fact that it exists and from tradition. So a lawyer's "de facto" is not only "in practice"; it can mean "in practice, and accepted by the law anyway".

Where you will meet it

Four examples, each from a primary or official source.

Segregation: in fact, or by law

The Cornell definition of de facto segregation describes a term used during the school integration efforts of the 1960s: a situation where legislation did not openly separate students by race, but segregation in schools continued. A Pennsylvania court, in Balsbaugh v. Rowland, put the distinction this way: de facto segregation is segregation that exists in fact, as distinguished from segregation imposed by law or by public authority, which is de jure.

PDF: a de facto standard that became a formal one

The Portable Document Format was introduced in 1993, and Adobe made its specification freely available from the start. According to the International Organization for Standardization (ISO), Adobe's PDF version 1.7 became the international standard ISO 32000 in 2008. For the fifteen years in between, the full PDF specification was Adobe's own published document, not an international standard, although ISO had begun standardising narrower profiles built on it, such as PDF/X in 2001. ISO itself calls PDF the worldwide de facto standard for electronic documents.

English in the United States

On 1 March 2025 the US President signed Executive Order 14224, Designating English as the Official Language of the United States. The order's own text says that English "has been used as our national language" since the founding, and that it was "long past time" it was declared official. In the terms of this article, the order took a de facto national language and declared it the official one.

Customary marriages in South Africa

South Africa's Recognition of Customary Marriages Act 120 of 1998, in force from 15 November 2000, was passed "to make provision for the recognition of customary marriages" and to set out the requirements for a valid one. The statute did not invent customary marriage; it gave marriages concluded under customary law a legal framework of their own.

De Facto: Understanding Reality Beyond Formal Recognition

The examples share a pattern. Something works in practice first; recognition, if it comes, comes later. The gap between the two can run in either direction: something can be in force on paper and ignored in practice, or followed by everyone and written down nowhere.

That gap is where most arguments about legitimacy sit. A de facto situation is not automatically fair or lawful, and a de jure one is not automatically followed. The phrase describes; it does not judge.

Reading the phrase when you meet it

When a headline or a report calls something "de facto", two questions usually make it clear:

What is happening in practice? That is the de facto part.
What does the rule, law or title say? That is the de jure part, and it may say nothing at all.

The distance between the two answers is the story. A "de facto ban" means something is blocked in practice without a law that bans it; a "de facto leader" holds power without the title. Where the two answers are the same, the phrase adds little, and plainer words would do.

Sources

Legal Information Institute, Cornell Law School, Wex: de facto, de jure and de facto segregation.
Online Etymology Dictionary: de facto.
International Organization for Standardization, The worldwide standard for electronic documents is evolving (3 August 2017), and the catalogue entry for ISO 32000-1:2008.
PDF Tools, PDF ISO standards (PDF/X, ISO 15930).
Executive Order 14224, Designating English as the Official Language of the United States, 90 FR 11363 (Federal Register).
South African Government, Recognition of Customary Marriages Act 120 of 1998.

Checked September 2026.

Did King Baldwin IV Really Offer the Crown to Balian of Ibelin? ∙ Balian of Ibelin and the Myth of Kingship

Hacks Vitae — Thu, 22 Feb 2024 21:09:32 GMT

Beyond the Hollywood Crown: Balian of Ibelin and the Myth of Kingship

None of the four chronicles we searched (William of Tyre, the chronicle attributed to Ernoul, Baha ad-Din and Ibn al-Athir) records Baldwin IV offering Balian of Ibelin the crown. The idea comes from Ridley Scott's film Kingdom of Heaven (2005), written by William Monahan. In the film Balian is a blacksmith, the unknown son of a crusader baron, who comes to Jerusalem and is drawn into palace intrigue: the dying King Baldwin does not want the kingdom to pass to Guy de Lusignan, and Balian is the obvious candidate to become Sibylla's husband and the next king. He refuses to take part in the plot.

The real Balian's story is different, and in some ways more interesting. He did not need a crown to matter: the Muslim historian Ibn al-Athir wrote that his rank among the Franks came close to the king's, and he later helped decide who wore the crown.

Who Balian was

Balian of Ibelin was a baron, not a blacksmith. William of Tyre, writing in the kingdom at the time, names him as one of the three sons of Balian the elder (also called Barisan): Hugh, Baldwin and Balian, all of whom took the surname Ibelin from the castle their father guarded.

In 1177 Balian, "the brother of Baldwin of Ramlah", married Queen Maria, the widow of King Amalric, with the king's consent. Maria was a Greek (Byzantine) princess, and with her Balian received the city of Nablus, her jointure, to hold for her lifetime. The marriage also made him stepfather to Maria's daughter by Amalric, Isabella, a half-sister of Baldwin IV. So he was not an outsider who rose: he married into the royal family, and the politics of the kingdom were his day job.

Whose line is "a kingdom of conscience"?

Many people search for this line, so it is worth being exact. It is film dialogue, and it is Balian's, not Baldwin IV's.

Early in the film, when Balian asks what a king could ask of a man like him, Godfrey of Ibelin answers: "A better world than has ever been seen. A kingdom of conscience. A kingdom of heaven."
Later, when a knight tells Balian that Jerusalem has no need of a perfect knight, Balian answers: "No. It is a kingdom of conscience, or nothing." It comes about an hour and 47 minutes into the film.

Neither line is in William of Tyre's history of the kingdom. Both belong to the screenplay.

Fact vs. Fiction: Untangling the Succession Crisis

The real succession crisis turned on marriages and a child king, not on an offer to Balian.

In April 1180 Baldwin IV married his sister Sibylla to Guy de Lusignan and made him Count of Jaffa. William of Tyre, who disapproved, notes that the wedding was held in Easter week, against custom.
Baldwin V was not Guy's son. He was Sibylla's son by her first husband, William of Montferrat, known as Longsword, who died in June 1177 leaving her pregnant.
Homage was done to the child while his uncle still lived. After Baldwin IV's death the regent was Raymond III of Tripoli, not Balian. Baldwin V held the title for scarcely a year and died in 1186, still a boy.
His death opened the crisis. In a coup led by Sibylla and Joscelin, the brother of her mother Agnes, Guy was made king, and Raymond's regency ended.

Not a rise: he was already there

The film shows Balian drawn in by love and accident. The record shows a baron at the centre of events.

When Guy was crowned in 1186, the barons who opposed him, with Raymond III and Balian's brother Baldwin of Ramla among them, planned to crown Isabella's husband, Humphrey of Toron, instead. Humphrey slipped away to Jerusalem and did homage to Sibylla, and the plan collapsed.
In 1186–87 Balian was one of the envoys Guy sent to make peace with Raymond, and the chronicle attributed to Ernoul, who served as Balian's squire, says he advised Guy to seek that reconciliation.
Maria Comnena was Baldwin IV's stepmother, since she had been his father's second wife.

Leadership in the Face of Adversity: Defending Jerusalem

Here the film is closest to the record, though the details differ.

Balian was in the rearguard at Hattin in 1187 and escaped to Tyre. He asked Saladin's leave to go to Jerusalem to bring out his wife and children, promising to stay only one night and not to bear arms.
In Jerusalem the townspeople begged him to stay and be their lord. The Patriarch absolved him of his oath to Saladin, Balian agreed, and the city did homage to him and received him as its lord. (Ernoul's 19th-century editor sums this up as Balian receiving homage "as regent".)
According to Ernoul, there were only two knights in the city, both survivors of Hattin, so Balian knighted up to sixty sons of burgesses (fifty, in some manuscripts) to have anyone at all to lead.
Saladin took up position outside the walls on 20 September 1187, and the city surrendered on 2 October 1187, about two weeks later.

The film turns the surrender into a matter of honour. In the account of the Muslim historian Ibn al-Athir, Balian's leverage was a threat: if refused terms, the defenders would kill their own families, burn their goods, pull down the Dome of the Rock and the al-Aqsa mosque, and kill the 5,000 Muslim prisoners in the city. Saladin's council advised him to accept a ransom. What Balian won was ransom terms, not freedom: those who could pay their ransom left, and those who could not were to stay as slaves. Ibn al-Athir records that Balian offered 30,000 dinars on behalf of the poor. The full account, with both sides' numbers, is in Salahuddin Ayyubi: Righteousness of the Faith.

Beyond the Crown: A Legacy of Service and Sacrifice

The film ends with Balian back in his village, telling Richard the Lionheart, "I am the blacksmith." The real Balian never held the crown, and he never held the regency of the kingdom either. What he did instead was help decide who would rule.

In 1190, after Queen Sibylla died, Isabella's marriage to Humphrey of Toron was set aside and she married Conrad of Montferrat, which gave Conrad a claim to the throne against Guy. The Itinerarium Peregrinorum, a chronicle of King Richard's crusade that is openly hostile to Conrad, names Balian and Maria among those Conrad won over to persuade Isabella to seek the separation.
In 1192 Balian took part in the peace talks between Richard and Saladin. The truce was dated 2 September 1192 and was to last three years; Saladin's biographer Baha ad-Din names "the son of Birezin", Balian, among those who were told its terms and accepted them.
After the truce, according to Ernoul, Saladin gave Balian a castle five leagues from Acre, called Caymont in some manuscripts, with its land.

The Power of Storytelling: Fact and Fiction Intertwine

Kingdom of Heaven is a film about ideas, and its best-known lines are the screenwriter's. That is worth knowing when a line from it turns up as a medieval quotation.

Some checks that help with any historical film:

Who is speaking? A line said by a character in a film is the screenwriter's, however historical the character.
What do the chronicles say? For the kingdom of Jerusalem, William of Tyre, the chronicle attributed to Ernoul, Baha ad-Din and Ibn al-Athir are all available in translation, and they often disagree.
Where does the real story differ? Here the difference is the interesting part: Balian mattered not because he refused a crown, but because he helped defend a city and decide a succession.

Sources

William of Tyre, A History of Deeds Done Beyond the Sea, trans. E. A. Babcock and A. C. Krey (Columbia University Press, 1943), Internet Archive, with the translators' introduction and notes.
Chronique d'Ernoul et de Bernard le Trésorier, ed. L. de Mas Latrie (1871), Internet Archive, including the editor's chapter summaries.
Baha ad-Din, The Life of Saladin, trans. C. R. Conder (Palestine Pilgrims' Text Society, 1897), Internet Archive.
Ibn al-Athir, Kamel-Altevarykh, French translation in Recueil des historiens des croisades: Historiens orientaux, vol. 1 (1872), Internet Archive.
Itinerary of Richard I (Itinerarium Peregrinorum), in Chronicles of the Crusades (Bohn, 1848), Internet Archive.
Kingdom of Heaven (2005), scene transcripts on Clip.Cafe; reviews by PopMatters (Cynthia Fuchs, 6 May 2005) and Plugged In.

Checked September 2026.

Lifemaxxing: A Deep Dive into Optimizing Your Life for Happiness

Hacks Vitae — Tue, 20 Feb 2024 11:47:57 GMT

Lifemaxxing Meaning

Lifemaxxing is internet slang for deliberately improving every part of your life at once — health, work, money, relationships, learning. It belongs to a family of "-maxxing" words that CNBC listed in July 2026: booksmaxxing, fibermaxxing, looksmaxxing and, for balance, relaxmaxxing.

The word has a second, narrower use too. In September 2026 CNBC used "lifemaxxing" for choosing to spend on experiences now — family, travel, meaningful memories — even if it means slower progress on financial goals, reporting a SoFi survey of more than 4,000 US adults in which 72% said they were willing to make that trade. This page is about the broader sense, and it gives no financial advice.

Beyond Self-Help Cliches

The version of lifemaxxing worth keeping is not about reaching an ideal or forcing yourself into a mould. It is closer to experimenting: trying a change, keeping what helps, dropping what does not, and adjusting towards your own picture of a good life rather than someone else's.

Where the word comes from

Worth knowing before you adopt the word, because its history explains where it can go wrong.

"-maxxing" comes from gaming. Merriam-Webster says it comes from gaming slang max, a shortening of maximize, and that its use "gained traction in online communities in the mid-2010s (especially those composed mostly of young men)" in words such as looksmaxxing. The dictionary dates the first known use of looksmaxxing to 2016, and defines it as taking "sometimes extreme or dangerous measures" to look more attractive.

It spread in words such as looksmaxxing. A 2025 study in Crime, Media, Culture by Anda Iulia Solea and Lisa Sugiura describes looksmaxxing and the "PSL scale", "a hierarchical framework used to assess facial attractiveness", as ways incel ideology is rebranded "under the guise of self-improvement", and traces how the terms travel to mainstream platforms such as TikTok. Merriam-Webster notes that "-maxxing" has since spread across mainstream social media, where it is now often used "humorously or ironically".

Why this matters: in its original setting, the framing treats a person as a set of numbers to push up, scored by other people. That is the part worth leaving behind. If a version of lifemaxxing has a scoreboard on it, it may have inherited it.

And a caveat. Optimisation and perfectionism are close neighbours. In a July 2026 CNBC piece, psychotherapist Billie Dunlevy said the maxxing trend has a "perfectionistic ideal about it, which isn't based in reality", and clinical psychologist Jennifer Hartstein said "the idea is about optimization, but the risk is about overdoing it." Dunlevy also said that aggressively searching for ways to change your appearance can lead to body dysmorphia or other body image issues. The NHS describes body dysmorphic disorder as "a mental health condition where a person spends a lot of time worrying about flaws in their appearance", flaws that are "often unnoticeable to others", and says its symptoms "can get better with treatment". The NHS advises seeing a GP if you think you might have it.

Pillars of Lifemaxxing

This page groups what people attach to the word into five areas. None of these is a claim about results; they are where the attention goes.

Health & Fitness

Sleep, movement, food and stress: the everyday habits this area covers.

Productivity & Efficiency

Getting more done in less time with better planning, simpler routines and technology, used on purpose rather than by default.

Career & Finances

Work that fits your values and skills, skills that build on each other, and money that is understood rather than ignored. This page does not give financial advice.

Relationships & Social Connection

Keeping up a few relationships that matter: the people you call, the friends you make time for, the support you can lean on.

Personal Growth & Learning

Learning new things, picking up interests, and stepping outside what is comfortable, as an ongoing process rather than a finish line.

Lifemaxxing in Action

The practices people attach to lifemaxxing are mostly familiar ones:

Dropping habits that get in the way, such as putting things off or doom-scrolling.
SMART goals. The acronym goes back to a 1981 article by George T. Doran in Management Review, and its wording has changed over time; one common version today is Specific, Measurable, Attainable, Relevant and Time-bound.
Productivity systems, such as the Pomodoro Technique, which Francesco Cirillo says he created in the 1980s "with a tomato-shaped kitchen timer", or Getting Things Done, the system created by David Allen.
Learning resources: books, courses, mentors, or therapy.
Routines, such as a morning ritual or an evening review.
Gratitude and mindfulness practices, for noticing what is already going well.

Each of these is a tool. None of them is proof that a life is going well.

Lifemaxxing Tips

The cautions matter as much as the tools:

Balance over optimisation. A life with no slack in it has no room for the unexpected good things either.
Comparing yourself with others can mean comparing your ordinary days with other people's best ones. The version that helps is measured against your own last month.
Sustainability over bursts. A habit you can keep for a year tends to matter more than one you can only keep for a week.
A process, not a destination. There is no final score.

The CNBC piece offers two checks worth borrowing. Dunlevy's: "If no one was looking or no one was able to see the quantifiable results, would I still be pushing this hard?" And Hartstein's: "What's a reasonable goal for me specifically?" Both therapists also said the healthiest pursuits are ones you can step away from when you need to and pick back up when you have the bandwidth.

Lifemaxxing is not a magic bullet. Treated as a set of experiments rather than a scoreboard, it can be a useful way to think about change.

Exploring More About Lifemaxxing

Beyond the Individual

Some people extend the idea past themselves: how they travel and consume, the causes they give time to, and the companies they choose to support.

Technology and Lifemaxxing

Technology cuts both ways. The same phone that holds your habit tracker holds the feeds that eat the time you meant to spend on the habit. Some people schedule regular screen breaks; others are more deliberate about what they follow.

Lifemaxxing Community

Some people do this with others: a small group with similar goals, a mentor a few steps ahead, or an online community around one specific skill. It is worth choosing groups that cheer effort rather than rank people.

LIFEMAXXING

Lifemaxxing, at its most useful, is a name for paying attention to how you live: change one thing at a time, and keep what helps.

A way to put it to use:

Pick one area, not five. Choose the one change that would make the others easier, and start there.
Use Dunlevy's question before adding anything new: would you still do it if nobody could see the results?
Keep the scoreboard private, or drop it. Track only what helps you decide something.
If tracking starts to feel like surveillance, or a missed day ruins the week, that is worth talking to someone about.

Sources

-maxxing — Merriam-Webster slang dictionary and looksmaxxing — Merriam-Webster, for the origin of the suffix and the first known use.
Anda Iulia Solea and Lisa Sugiura, "Digital Subcultural Diffusion Theory: Rebranding the incel ideology through Looksmaxxing, Sub5s and the PSL scale", Crime, Media, Culture (2025), doi:10.1177/17416590251387245.
Jesse Kessenheimer, "Gen Z’s Approach to Extremeness, like ‘Lifemaxxing,’ is Actually a Healthy Response to Overwhelming Societal Pressures", Yahoo Lifestyle, 29 January 2026 (an opinion piece), for the broad sense of the word.
Renée Onque, "From booksmaxxing to looksmaxxing: Why the viral trends concern some mental health experts", CNBC, 11 July 2026.
Kamaron McNair, "72% of Americans would slow financial progress to enjoy life now, SoFi survey finds", CNBC, 23 September 2026.
Body dysmorphic disorder (BDD) — NHS.
"Developing SMART Goals for Your Organization" — University of Florida IFAS Extension, on SMART goals and Doran (1981).
Francesco Cirillo — Pomodoro Technique and About GTD — David Allen Company.

Checked September 2026.

Mukbang: Beyond Mountains of Food - A Dive into the Global Eating Show Phenomenon

Hacks Vitae — Fri, 09 Feb 2024 08:58:58 GMT

Mukbang Meaning

Mukbang (먹방, meokbang) is short for 먹는 방송 (meokneun bangsong), "eating broadcast". The National Institute of Korean Language's open dictionary, Urimalsaem, defines it as a programme that mainly shows people eating, and notes that it can also mean an eating scene in a drama or film.

A typical mukbang has one host eating a large meal on camera and talking to viewers, either live with a chat running alongside or in a recorded video. The reason viewers give most often for watching is the company, and some watch for the sound of eating.

The facts, with dates


The word	먹방 meokbang, short for 먹는 방송 meokneun bangsong, "eating broadcast"
Where it started	South Korea. A 2025 research review dates the first mukbang to 2008 and its rise in popularity to 2010, on the streaming platform AfreecaTV (now called SOOP)
What a BJ is	"Broadcast jockey", the usual name for the host. On AfreecaTV, some hosts earn money from viewers' contributions in virtual currency
When it went global	Twitch launched a Social Eating category at the end of June 2016, reversing an earlier policy that had prohibited mukbang streams
Why people watch	Easing the loneliness of eating alone is the reason mentioned most often in research on viewers' motives

The dates depend on what is being counted. A 2020 study, using Google Trends, found that "mukbang" started trending as a search term in South Korea in winter 2014 and worldwide in autumn 2014, and so describes the format as originating around 2014. The earlier dates describe the first broadcasts; the later one describes when the word took off.

The Allure of Mukbang

Company at mealtimes

A 2025 scoping review of the research found that alleviating the feelings of loneliness associated with eating alone is the reason mentioned most often for watching mukbang, and noted that this was particularly relevant during the COVID-19 pandemic. Live shows make the company feel real: viewers comment, the host answers, and on AfreecaTV hosts can often be heard naming and thanking viewers in real time when they contribute money.

Food and culture

The same review found that another common reason for watching during the pandemic was to learn about food: its taste, its nutritional value and how it is cooked. A meal is an easy way into a cuisine you have never eaten.

The sound

The chomping and chewing sounds hosts make can evoke what researchers call an autonomous sensory meridian response (ASMR): a pleasantly warm, tingling sensation that starts at the crown of the head and spreads down the body.

The Dark Side of Mukbang

Watching and eating habits

The research here is mixed, and worth reading closely. A 2020 study of YouTube comments and Reddit posts found that for some viewers watching mukbang helped them eat more, avoid binge eating or feel less lonely, while for others it motivated restrictive eating or triggered a relapse into loss-of-control eating. The authors summed up the comments as "simultaneously useful and hurtful".

A questionnaire study has found a positive relationship between problematic mukbang watching and binge-eating or purging tendencies. That is an association: it does not show which comes first.

Food waste

Food waste is the concern one government has written into law: China's 2021 law on food waste bans binge-eating broadcasts (see below).

The serious part

Some of the risks are in the news, not only in studies.

Dongz Apatan, a mukbang vlogger from Iligan City in the Philippines, died in June 2024 after a hemorrhagic stroke. He was reportedly rushed to hospital after a mukbang session. In July 2024 the country's health secretary said the Department of Health was considering a ban on mukbang videos; the department's spokesperson said it would not ban them immediately.
Chen Ziyi, a Chinese mukbang streamer known online as Yingying, with more than one million followers, died on 17 August 2026, aged 24, according to her parents. In her last video, posted on 14 August, she said she was in hospital because her blood potassium had fallen as low as 2.3, and she put it down largely to her work as a mukbanger and a long-term habit of making herself vomit. Her parents' announcement did not give a cause of death.
What a doctor pointed to. A doctor at Zhejiang Hospital told the Global Times that binge eating itself does not directly cause severe potassium loss. The main risks, the doctor said, are inducing vomiting and using laxatives, which some mukbangers do off camera to keep their figures. Researchers list vomiting and laxative misuse among the compensatory behaviours of disordered eating.

What governments have done

South Korea, 2018. On 26 July 2018 the Ministry of Health and Welfare published a package of anti-obesity measures. One sentence in the 12-page report said it planned to "develop a guideline and create a monitoring system for advertisements and media that encourage overeating". A ministry official said there would be no punishment for those who did not comply. The plan still drew petitions on the presidential website, and an opposition leader called it "statism".
China, 2021. The Law on Food Waste, adopted on 29 April 2021, says "the production, release, and dissemination of competitive eating and binge-eating programs or audios and videos on food wasting shall be banned" (Article 22). A broadcaster or online video service that refuses to correct a breach, or where the circumstances are serious, can be fined between RMB 10,000 and RMB 100,000 (Article 30).
The Philippines, 2024. The Department of Health considered a ban after Dongz Apatan's death, as above.

If food feels like a problem for you. Watching this content is not itself a disorder, but if you recognise bingeing or purging in your own life, that is worth talking to a doctor about. Eating disorders are serious and can be life-threatening, and with treatment most people can recover.

If you are struggling or thinking about suicide, please reach out now. findahelpline.com lists free, confidential helplines worldwide. In the US and Canada, call or text 988. In the UK and Ireland, call Samaritans on 116 123. In Australia, call Lifeline on 13 11 14. In India, call Tele-MANAS on 14416. In the Philippines, call the National Center for Mental Health crisis line on 1800-1888-1553 (or 1553 from a Luzon landline). If you are in immediate danger, call your local emergency number.

What you can do with this

Notice what a session does for you. The research above finds the same videos helping some viewers and harming others. Whether you come away hungrier, calmer, or ready to eat less than you need is information about you, not about mukbang in general.
Separate the company from the quantity. If what you want is someone to eat with, the loneliness finding suggests that is the part doing the work. A call with a friend at dinner or a shared meal offers company too.
Read the amounts as a performance. A video shows what happens on camera. The doctor quoted above describes what can happen afterwards, off camera.

Sources

Definition of 먹방 — National Institute of Korean Language, Urimalsaem (우리말샘), entry for 먹방.
Origin (2008, 2010, AfreecaTV), motives, ASMR, and the link between problematic watching and binge eating or purging — Wang et al., "Use of Mukbang in Health Promotion: Scoping Review", Journal of Medical Internet Research, 2025; full text via NCBI.
Google Trends dating, AfreecaTV virtual currency, and viewers' experiences — Strand and Gustafsson, "Mukbang and Disordered Eating: A Netnographic Analysis of Online Eating Broadcasts", Culture, Medicine, and Psychiatry, 2020 (full text).
Twitch's Social Eating category — Tubefilter, "Twitch Chows Down With New 'Social Eating' Category", 30 June 2016.
Dongz Apatan — PEP.ph, "Mukbang vlogger passes away from hemorrhagic stroke", 29 June 2024; Rappler, "Mukbang no more? DOH considers banning 'mukbang' after death of vlogger", 8 July 2024.
Chen Ziyi and the doctor's comments — Global Times, "Mukbanger, 24, dies of cardiac arrest reportedly linked to low potassium", 8 September 2026; Mothership, "Chinese mukbang influencer, 24, dies after repeatedly inducing vomiting after binge eating", 12 September 2026; PEP.ph, "Chinese mukbang influencer dies after suffering from hypokalemia", 9 September 2026.
South Korea's 2018 measures — Korea Biomedical Review, "How bad are 'mukbang' shows, really?", 7 August 2018; The Korea Times, "Beware: Gov't wants to regulate your eating habits", 2 August 2018.
China's law — National People's Congress, Law of the People's Republic of China on Food Waste, Articles 22 and 30.
Eating disorders — NIMH, Eating Disorders; NHS, Eating disorders.
Helplines — Find A Helpline and its Philippines page; 988 Lifeline; 9-8-8 Canada; Samaritans; Lifeline Australia; Tele-MANAS on Find A Helpline; MentalHealthPH.

Checked September 2026. Figures and regulations are as they stood at that date.

Epic Sax Guy Meme: The Enduring Legacy of Epic Sax Guy in Meme Culture

Hacks Vitae — Mon, 05 Feb 2024 09:47:47 GMT

Who Epic Sax Guy is

Epic Sax Guy is Sergey Stepanov, the saxophonist of the Moldovan band SunStroke Project. The meme is his saxophone solo from Moldova's entry at the Eurovision Song Contest 2010 in Oslo. The contest's own page on the band puts it simply: "It was after the 2010 contest that saxophonist Sergey Stepanov became a YouTube sensation known all over the world as the Epic Sax Guy."

Everything below comes from the contest's official records, plus a report by SBS, the Australian broadcaster of Eurovision, from the week the band returned in 2017, and one note from the meme database Know Your Meme.

From Moldova to Eurovision: The Birth of a Meme

SunStroke Project, formed in 2008, plays a mix of violin, saxophone and dance music, in the contest's description. In 2010 the band represented Moldova together with the singer Olia Tira, with the song "Run Away".

The semi-final. On Tuesday 25 May 2010, Moldova opened the first semi-final. "Run Away" scored 52 points, finished tenth and qualified for the final.
The final. On Saturday 29 May 2010, at the Telenor Arena, "Run Away" scored 27 points and finished 22nd of 25. The contest was won by Lena, for Germany.

So the song did not do well. The saxophone did. The band's Eurovision page says the video of Stepanov performing at the contest "went viral and has been watched millions of times", and SBS reported in 2017 that "when their performance was first uploaded to YouTube, he became known as the 'epic sax guy' as the video started going viral."

Back at Eurovision in 2017

The band did not stay a one-off. SunStroke Project, this time without Olia Tira, represented Moldova again at the Eurovision Song Contest 2017 in Kyiv, with "Hey Mamma". It finished second in the first semi-final and third in the final on Saturday 13 May 2017, with 374 points. Portugal's Salvador Sobral won that year.

The contest credits the music of "Hey Mamma" to four writers, Stepanov among them, and the lyrics to Alina Galetcaia. SBS, writing in May 2017, called the saxophone "the real star of this years song contest" [sic] and reported that Stepanov and the band's violinist, Anton Ragoza, had worked together in the army, and that the band's name came from a sunstroke Ragoza got while working in the field with him.

"Epic" and "Ultra": a note on names

Epic Sax Guy is the name the contest itself uses for Stepanov. It comes from the 2010 performance, and the contest kept it in 2017: its own video of the 2017 final is titled "Hey Mamma (LIVE feat. Epic Sax Guy)".

The 2017 performance is sometimes shared as "Ultra Sax Guy". We found that name in a fan video that set the two performances side by side, as recorded by the meme database Know Your Meme. We did not find it on the contest's results and participant pages, in the title of its 2017 video, or in the SBS report. It is a nickname for the 2017 clip, not a second person: both clips show Stepanov.

How to trace a meme clip to its source

Epic Sax Guy is an easy case, because the event behind it keeps a public record. The same steps work for most clips that come from a broadcast:

Find the event. A clip from a contest, a match or a televised show usually has an official results page. For Eurovision it is the contest's own site, with every entry, running order, score and placing.
Check the numbers there. The placing, the points and the date are the easiest details to get wrong when a clip is retold. The official page settles them.
Match the name to the right clip. Nicknames drift: "Ultra" was a fan label for the 2017 clip, seven years after "Epic" began.
Separate the clip from the person. A meme records a few seconds of someone's work. For facts about the person, use what they or a reliable publication have said on the record, and leave the rest out.

For a fuller method, see how to check something online before you share it.

Sources

Eurovision Song Contest, "Grand Final of Oslo 2010" and "First Semi-Final of Oslo 2010" (results), and "Oslo 2010" (event details), archived copies: https://web.archive.org/web/20251116031421/https://eurovision.tv/event/oslo-2010/grand-final/, https://web.archive.org/web/20251118030153/https://eurovision.tv/event/oslo-2010/first-semi-final, https://web.archive.org/web/20260107053901/https://eurovision.tv/event/oslo-2010
Eurovision Song Contest, "Grand Final of Kyiv 2017" (results) and "Kyiv 2017" (event details), archived copies: https://web.archive.org/web/20251115191004/https://eurovision.tv/event/kyiv-2017/grand-final/, https://web.archive.org/web/20260107053901/https://eurovision.tv/event/kyiv-2017
Eurovision Song Contest, "Sunstroke Project" (participant page), archived copy: https://web.archive.org/web/20251114221555/https://eurovision.tv/participant/sunstroke-project
SBS Guide, "Everything you wanted to know about sax (But were too afraid to ask)", SBS (14 May 2017): https://www.sbs.com.au/whats-on/article/everything-you-wanted-to-know-about-sax-but-were-too-afraid-to-ask/k9ejkl8er
Eurovision Song Contest (official YouTube channel), "Sunstroke Project - Hey Mamma (LIVE feat. Epic Sax Guy) | Moldova | Grand Final | Eurovision 2017": https://www.youtube.com/watch?v=SWaQdHoCvYk
Know Your Meme, "Epic Sax Guy" (entry, section "2017 Appearance"): https://knowyourmeme.com/memes/epic-sax-guy

Checked September 2026.

Fight Club: A Philosophical Brawl in a Consumerist Wasteland

Hacks Vitae — Wed, 31 Jan 2024 12:39:09 GMT

Fight Club Analysis: A Philosophical Brawl in a Consumerist Wasteland

Fight Club is Chuck Palahniuk's 1996 novel and David Fincher's 1999 film of it. Its target is a life built around buying things. Its story is about a man who tries to escape that life through bare-knuckle fights, and then through something much worse.

This page covers what happens, where the famous lines come from and who says them, how the novel and the film end differently, and how the story compares with the Japanese idea of ikigai. The facts are sourced at the end. The interpretations are ours and are marked as readings; the film does not settle them, and you may read it differently.

The novel, the film and what happens

The book and the film. The novel was published in New York in 1996. The film, written by Jim Uhls and directed by David Fincher, opened in Los Angeles on 15 October 1999. Edward Norton plays the narrator, who is credited only as "Narrator"; Brad Pitt plays Tyler Durden and Helena Bonham Carter plays Marla Singer.
The narrator. He cannot sleep, and he goes to support groups for illnesses he does not have, "just so he can put his head on someone's shoulders, cry in someone's embrace, and have someone listen to him", as the film site No Film School describes it.
Tyler. After an explosion destroys the narrator's apartment, he meets Tyler Durden, who "makes and sells soaps", over a beer.
The club and what it becomes. The fight club grows, and Tyler turns it into Project Mayhem, "a vandalistic group he hopes will disrupt society in a major way". The narrator later learns that he and Tyler are split personalities, and that Tyler plans to blow up "several buildings housing credit records to wipe debts".

A Nietzschean Punch to the IKEA-Furnished Soul

In the film, the narrator's apartment stands in for his identity. In what No Film School calls "the IKEA scene", he orders furniture "sitting on the toilet, a catalog in his hands"; the site sums it up: "he is mapping his worth through brand names". Tyler's reply, once the apartment is gone, is the film's most quoted line:

"The things you own end up owning you." — Tyler Durden, Fight Club (1999 film)

Our reading. Tyler's pitch is that freedom starts with losing what you own. Another of his lines in the film makes it sharper: "It's only after we've lost everything that we're free to do anything." That invites a comparison with Nietzsche, whose "talk about the creation of values", in the words of the Stanford Encyclopedia of Philosophy, "challenges philosophical common sense". The comparison has limits: the same encyclopedia records that scholars read Nietzsche's "overman" in conflicting ways, and some hold that his position on it "is simply riven by internal contradiction". Calling Tyler a Nietzschean hero assumes an answer the philosophers have not agreed on.

Beyond the Blood and Bruises: Self-Destruction or Self-Creation?

Our reading. The film asks whether tearing a life down can be a way to build one, and Tyler says yes. The narrator's own line, early in the film, gives the question its urgency: "This is your life and it's ending one minute at a time."

The story then follows Tyler's answer to where it leads. One of the narrator's friends is killed by the police during a Project Mayhem mission, and the plan grows into bombing buildings. The film shows the pull of violence clearly and then shows its cost; whether that adds up to a critique or a celebration is a judgement each viewer makes.

Project Mayhem: An Anarchic Masterpiece or a Terrifying Climax?

This section describes both endings.

The film. The narrator shoots himself through the cheek, which "kills" Tyler. Variety describes what follows: he watches "buildings burst into flames in apparent confirmation that his plan to destroy modern civilization is being executed".
The novel. In the book, as GamesRadar summarises it, the plot fails "when the explosives malfunction". After shooting himself, the narrator "wakes up in a psychiatric facility" and learns that members of Project Mayhem are looking after him, and that the group "has no intention of slowing down".
A third ending. In January 2022, the version on Tencent Video in China stopped before the explosions and added a title card: "The police rapidly figured out the whole plan and arrested all criminals, successfully preventing the bomb from exploding." In February 2022, after a backlash, Tencent reportedly restored 11 of the 12 minutes it had cut.

Our reading. Neither the novel's nor the film's ending shows anyone better off. The film ends with buildings coming down, and the novel with the movement still going.

Fight Club and Ikigai: the same diagnosis, opposite prescriptions

The film's complaint, that a life assembled out of catalogue furniture and a job you tolerate is not a life, is not unique to it. The Japanese word ikigai is defined in the Digital Daijisen dictionary as 生きるに値するもの。生きていくはりあいや喜び。, which translates roughly as "something worth living for; the zest and joy of going on living". It starts from a similar question and goes somewhere else.

Our reading	Fight Club	Ikigai
The problem	A life built around buying things feels empty	A life needs something worth living for
The method	Destroy the false self; look for the real one in violence	Notice what already makes life feel worth it, and build on it
Where it ends	Buildings coming down	An ordinary life with a point to it

The overlap is real: both reject inherited scripts, and both treat purpose as something found rather than issued. The difference is the useful part. Read this way, Fight Club is a diagnosis dressed as a solution: precise about what is wrong, with nothing workable to put in its place.

Three lines from the film carry the diagnosis. All three are film dialogue, as transcribed on the film's IMDb page:

"This is your life and it's ending one minute at a time." — the narrator

"The things you own end up owning you." — Tyler Durden

"We've all been raised on television to believe that one day we'd all be millionaires, and movie gods, and rock stars. But we won't." — Tyler Durden

One more thing about ikigai. The four-circle diagram that circulates online, with love, skill, need and pay overlapping, is not Japanese. Marc Winn published it on 14 May 2014, and he has since added that it "began as a purpose venn diagram by the Spanish astrologer Andrés Zuzunaga".

The ikigai half, practically

Some ways to use the idea, if it appeals:

Start with a plain list of what you actually enjoy, not what you would like to enjoy.
You do not need all four circles. The diagram asks for an overlap of what you love, what you are good at, what the world needs and what pays. Since it is a 2014 adaptation rather than the Japanese idea, nothing requires all four. The dictionary sense is simpler: something that makes life worth living.
Count relationships in. In Okinawa, a moai is a private mutual-aid group for money. The Ryukyu Shimpo's dictionary of Okinawan terms dates the first record of the practice to 1733, and a 2023 book by the anthropologist Misa Hirano describes moai as still active in Okinawa as everyday finance, mutual help and circles of friends. The fight club was the film's version of belonging; a group like this is a quieter one.

Delving Deeper

Three threads to follow, each a reading rather than a fact about the film:

Nietzsche. Tyler's call to throw out inherited values invites comparison with Nietzsche's. The Stanford Encyclopedia of Philosophy is the place to see how much scholars disagree about what Nietzsche meant.
Camus and the absurd. For Camus, the absurd is the paradox "between our impulse to ask ultimate questions and the impossibility of achieving any adequate answer", in the encyclopedia's summary. He set it out in The Myth of Sisyphus, published in October 1942, and he denied being an existentialist.
Men without a place. One of Tyler's speeches in the film includes "We're the middle children of history, man. No purpose or place." The film puts men who feel that way at its centre. What it concludes about them is left to the viewer.

Sources

American Film Institute Catalog, "Fight Club (1999)": credits, literary source and release.
IMDb, "Fight Club (1999) – Quotes" (user-transcribed, with timestamps).
Vishal Wagh, "The Philosophy Behind the Most Quoted Line in 'Fight Club'", No Film School.
Variety, "'Fight Club' Censored in China, Gets New Ending in Censorship Wars" (January 2022); NME, "'Fight Club' ending restored in China after outcry" (7 February 2022); RTÉ, "Fight Club ending changed for streaming in China" (26 January 2022).
GamesRadar+, "Fight Club creator explains why he 'wasn't a big fan' of the David Fincher movie's ending" (11 September 2023).
Stanford Encyclopedia of Philosophy: "Friedrich Nietzsche"; "Albert Camus".
Digital Daijisen (Shogakukan), "生き甲斐", via Kotobank.
Marc Winn, "What is your ikigai?", The View Inside Me (14 May 2014, with a later note).
Ryukyu Shimpo, Okinawa dictionary, "模合 (もあい)"; Center for African Area Studies, Kyoto University, publication page for Misa Hirano, 沖縄のもあい大研究 (2023).

Checked September 2026. The quotations are film dialogue from the 1999 film; the novel's wording was not checked, and none of these lines is attributed to it here.

Carpe Diem: Seizing the Day, One Moment at a Time

Hacks Vitae — Tue, 30 Jan 2024 17:23:37 GMT

Carpe Diem Meaning

Carpe diem is Latin, and it is usually translated "seize the day". The verb carpo means, first of all, to pick or pluck, as you would a flower or a piece of fruit, so "pluck the day" is closer to the word. The standard Latin dictionary, Lewis & Short, files Horace's use of it under a figurative sense of the verb: "to enjoy, use, make use of".

The two words are also the end of a longer sentence, and the rest of that sentence changes what they mean.

Carpe Diem's Origin

The phrase comes from the Roman poet Horace (Quintus Horatius Flaccus), in the eleventh poem of Book 1 of his Odes. The first three books of the Odes came out together around 23 BC; the 1911 Encyclopaedia Britannica dates them to "about the end of 24 or the beginning of 23 B.C."

The same encyclopedia sums up the tone of Horace's teaching as "a refined Epicureanism", and gives the aim of his philosophy as, among other things, "to make the most of life, and to contemplate its inevitable end without anxiety". Carpe diem puts that in two words.

The line in full, and what Horace meant

The whole poem is only eight lines long. Here it is in Latin, as printed in Paul Shorey and Gordon J. Laing's edition of the Odes (1919):

Tu ne quaesieris, scire nefas, quem mihi, quem tibi finem di dederint, Leuconoe, nec Babylonios temptaris numeros. ut melius, quidquid erit, pati. seu pluris hiemes seu tribuit Iuppiter ultimam, quae nunc oppositis debilitat pumicibus mare Tyrrhenum: sapias, vina liques et spatio brevi spem longam reseces. dum loquimur, fugerit invida aetas: carpe diem quam minimum credula postero.

And in plain English. This is our own literal translation, not a published one:

Don't ask (we are not meant to know) what end the gods have given to me, or to you, Leuconoe, and don't try out Babylonian numbers. How much better to put up with whatever comes, whether Jupiter has given us more winters, or this one as the last, the one now wearing the Tyrrhenian sea down against the rocks. Be wise: strain the wine, and since our time is short, cut back long hope. While we are talking, envious time will have run away. Pluck the day, trusting as little as possible in the next one.

The famous phrase is the first half of the last line. The second half, quam minimum credula postero, is "trusting as little as possible in the next one": act now, because the future is not something you can count on. Without it, "seize the day" can sound like permission to be reckless. With it, the point is closer to the opposite.

The context is a warning about fortune-telling

The poem is addressed to a woman Horace calls Leuconoe (the Latin word credula, "trusting", is feminine). It opens by telling her to stop trying to find out when she, or he, will die. The "Babylonian numbers" are astrology: as Paul Shorey put it in his 1898 edition of the Odes, "astrology is Babylonian numbers". Whether they have many winters left or only this one, Horace says, strain the wine, cut back long hope, and take today.

So the original carpe diem is a poem about mortality and about not being frozen by what you cannot know. Read that way, it sits closer to memento mori than to the way the phrase is often used today.

And "seize" is a translator's choice. The dictionary's first meaning of carpo is to pick or pluck flowers and fruit, and it places Horace's line under the sense "to enjoy, use, make use of". "Pluck the day" keeps the picture of taking something that is ripe now and will not stay ripe.

Why the phrase is famous in English

The phrase was in English use long before the film. In a letter dated 2 January 1817, printed in Thomas Moore's Life of Lord Byron, Byron wrote: "as to the future, I never anticipate—carpe diem—the past at least is one's own".

Much of its modern fame comes from a film: Dead Poets Society (1989), directed by Peter Weir from a screenplay by Tom Schulman, with Robin Williams as the English teacher John Keating. In the script, Keating has a pupil read the opening of Robert Herrick's poem "To the Virgins, to Make Much of Time" ("Gather ye rosebuds while ye may"), and tells the class: "The Latin term for that sentiment is Carpe Diem."

He then walks them to old photographs of former pupils and whispers: "Carpe Diem. Seize the day boys, make your lives extraordinary." That is film dialogue, written for the character. The scene keeps Horace's point, though: the photographs are there to remind the boys that those pupils are dead, "now fertilizing daffodils", as Keating puts it.

A 2019 essay by the linguist Chi Luu in JSTOR Daily says the film "catapulted an obscure phrase from a long dead language into the public consciousness and mainstream popular culture".

Herrick's poem, which makes the same argument in English, was published in his collection Hesperides in 1648.

Analyzing Carpe Diem

Carpe diem is often read as a call to be impulsive. The poem points somewhere quieter: time is short, the future cannot be read in advance, so pay attention to what is in front of you and do not wait for a better moment that may not come. That can mean a small thing, like a cup of tea taken slowly, or a large one, like finally starting something you have been putting off.

Taking Hold of the Day: Practical Ways to Embrace Carpe Diem

Horace's poem gives three pieces of advice of its own, and each can be read as a habit:

"Don't try out Babylonian numbers." Leuconoe was checking forecasts about something no forecast could tell her. A modern version is noticing when you keep checking news, numbers or predictions you cannot act on, and choosing to stop checking.
"Cut back long hope." Horace is not saying "have no plans". He is saying do not hang today on a hope that stretches far ahead. One way to use this is to plan in shorter spans, and to ask what can be done this week rather than someday.
"Strain the wine." Here the verb, liquo, means to strain or clear, and the dictionary cites the Roman farming writer Columella on a saccus, quo vinum liquatur, a bag for straining wine. It is an ordinary household job. One reading is that taking the day does not have to mean anything dramatic; it can mean doing the ordinary thing in front of you, and enjoying it.

Sources

Horace, Odes 1.11, ed. Paul Shorey and Gordon J. Laing (Chicago: Benj. H. Sanborn & Co., 1919). Perseus Digital Library
Paul Shorey (ed.), Horace: Odes and Epodes (1898), introduction. Internet Archive text
Charlton T. Lewis and Charles Short, A Latin Dictionary (1879), "carpo" and "liquo". Perseus: carpo, liquo
William Young Sellar and James Gow, "Horace", Encyclopaedia Britannica, 11th edition (1911), vol. 13. Wikisource
Tom Schulman, Dead Poets Society, screenplay. Internet Movie Script Database
Andrew Bloomenthal, "Dead Poets Society Retrospective with Tom Schulman, Peter Weir, and Ethan Hawke", Script, 6 June 2019. scriptmag.com
Poetry Foundation, "Robert Herrick". poetryfoundation.org
Chi Luu, "How 'Carpe Diem' Got Lost in Translation", JSTOR Daily, 7 August 2019. daily.jstor.org
Thomas Moore, Life of Lord Byron, vol. 3, letter dated Venice, 2 January 1817. Project Gutenberg #16548

Checked September 2026. The Latin is from the 1919 Shorey and Laing edition; the English translation of the poem is our own. If you can show any of this wrong, with a source, write to hacksvitae@gmail.com.

Liquid IV vs Pedialyte: The Hydration Showdown ∙ Which is Better?

Hacks Vitae — Wed, 10 Jan 2024 17:13:52 GMT

The Contenders: Liquid IV vs Pedialyte

The short answer: once you compare the same amount of drink, the two carry similar sodium and potassium. The clearer differences are sugar (Liquid I.V. has more), what else is added (vitamins in Liquid I.V., zinc in Pedialyte Classic), the format, and what each maker says its product is for. Neither label makes one of them better for everyone, and dehydration from illness is a medical matter, covered at the end of this page.

Every figure below comes from the makers' own US product pages, read in September 2026. Formulas and labels change, and they differ by flavour and by country.

LIQUID IV

Liquid I.V. is a powdered drink mix sold in single-serving sticks and in multiserve pouches. Its own website gives the company's founding date as 2012 and lists Unilever as its parent company. The version compared here is the original Hydration Multiplier; the brand also sells sugar-free, energy and immune-support lines with different numbers.

PEDIALYTE

Pedialyte is made by Abbott, which says the brand has been "backed by science since 1966". The version compared here is Pedialyte Classic, a ready-to-drink liquid.

Round 1: Electrolyte Smackdown

The two labels use different serving sizes, so the fairest comparison is per equal volume. Liquid I.V. says: "we recommend mixing 1 stick with 16oz of water" (16 US fl oz is about 473 mL). Abbott gives Pedialyte Classic's figures per 12 fl oz (360 mL) and per litre, and its directions say: "READY TO USE. DO NOT ADD WATER OR DILUTE."

Converted to the same 16 fl oz of drink:

Per 16 fl oz of drink	Liquid I.V. Hydration Multiplier (one stick)	Pedialyte Classic
Sodium	490 to 630 mg, depending on flavour	about 520 mg (label: 390 mg per 12 fl oz)
Potassium	370 to 390 mg	about 373 mg (label: 280 mg per 12 fl oz)
Total sugars	10 to 11 g	about 7 g (label: 5 g per 12 fl oz)
Also listed	vitamin C, niacin (B3), vitamin B6, vitamin B12, pantothenic acid (B5)	zinc, 2.8 mg per 12 fl oz

How the ranges were read:

Liquid I.V.: we read the nutrition panels of 17 original Hydration Multiplier flavours on liquid-iv.com. Sodium runs from 490 mg (golden cherry) to 630 mg (pear) per stick; the other 15 flavours, lemon lime among them, list 500 to 520 mg. Potassium runs from 370 to 390 mg, and sugars are 10 or 11 g.
Pedialyte Classic: Abbott's page lists 390 mg of sodium and 5 g of sugars per 12 fl oz for every flavour except one, which lists 370 mg of sodium and 9 g of sugars.

What the table shows: sodium and potassium are close once the volumes match. Comparing one stick with one bottle, or one serving with another, compares different amounts of liquid. The sugar gap is real: a Liquid I.V. stick carries about half as much sugar again as the same volume of most Pedialyte Classic flavours.

Round 2: Taste Test

Taste is personal, and we did not run a tasting, so there is no winner in this round. Here is what goes into each:

Flavours. Liquid I.V. sells the Hydration Multiplier in flavours including strawberry lemonade, watermelon, passion fruit, lemon lime and concord grape. Abbott lists Pedialyte Classic in flavours including cherry frost, coconut burst, fruit punch, grape, iced berry, mixed fruit and strawberry, plus an unflavoured version.
Sweetening. Liquid I.V.'s lemon lime ingredient list starts with cane sugar and dextrose and includes stevia leaf extract. Pedialyte Classic is sweetened with dextrose, and its flavoured versions add the sweeteners sucralose and acesulfame potassium; the unflavoured version lists neither.

Round 3: Convenience Corner

Liquid I.V. comes as a dry stick: light to carry, but you need water and something to mix it in. Pedialyte Classic is already mixed, and Abbott sells it in litre bottles it describes as "reclosable" for "easy measuring and pouring". Which suits you depends on where you will be drinking it.

Round 4: Price Point Punch

Prices change with the retailer, the pack size and the country, so this page prints none. The fair comparison is the one Round 1 uses: cost per equal volume of drink. One Liquid I.V. stick makes 16 fl oz; for Pedialyte Classic, divide the bottle price by its volume.

What an oral rehydration solution is

The World Health Organization describes oral rehydration salts (ORS) as "a mixture of clean water, salt and sugar" that "is absorbed in the small intestine and replaces the water and electrolytes lost in the faeces". It says diarrhoea "should be treated with oral rehydration solution (ORS)".

The formula the WHO and UNICEF recommend is a reduced-osmolarity ORS with, per litre, 75 mmol of sodium, 20 mmol of potassium and 75 mmol (13.5 g) of glucose. Set beside the two products, per litre:

Per litre	Liquid I.V. (converted from 16 fl oz)	Pedialyte Classic (Abbott's 1 L label)	WHO/UNICEF reduced-osmolarity ORS
Sodium	about 1,040 to 1,330 mg	1,080 mg (one flavour: 1,030 mg)	75 mmol, about 1,720 mg
Potassium	about 780 to 820 mg	780 mg	20 mmol, about 780 mg
Sugar	about 21 to 23 g of sugars	13 g of sugars (one flavour: 25 g)	13.5 g of glucose

Both products carry less sodium than the WHO/UNICEF formula. Liquid I.V. carries more sugar than the formula's 13.5 g of glucose. Pedialyte Classic lists 13 to 25 g of sugars per litre depending on flavour: 13 g for every flavour but one, which lists 25 g.

What the makers say their products are:

Abbott says Pedialyte Classic "can help prevent mild to moderate dehydration caused by vomiting, diarrhea, exercise, travel, and extreme heat", and its label adds: "Use under medical supervision for the dietary management of mild to moderate dehydration during diarrhea and vomiting."
Liquid I.V. says its Hydration Multiplier products "are classified as an Oral Rehydration Solution (ORS)".

The NHS says that if you are being sick or have diarrhoea and are losing too much fluid, "a pharmacist can recommend oral rehydration solutions".

The verdict: where they differ

There is no overall winner, because the two differ in ways that matter to different people:

Sodium and potassium: close, once you compare the same volume.
Sugar: Liquid I.V. has more per equal volume.
Extras: Liquid I.V. adds vitamin C and B vitamins; Pedialyte Classic adds zinc.
Format: a stick you mix, or a liquid that is ready to drink.
What the maker says it is for: Abbott labels Pedialyte Classic for use under medical supervision in dehydration from diarrhoea and vomiting, as well as for exercise, travel and heat. Liquid I.V. sells a drink mix that it describes as an oral rehydration solution.

When hydration is a medical matter

Everything above is about the labels. Dehydration from illness is different. The NHS notes that babies, children and older adults are more at risk of dehydration, and Abbott's own directions for Pedialyte Classic say: "For children under 1 year of age: consult your doctor", and "Consult your doctor if vomiting, fever, or diarrhea continues beyond 24 hours".

The NHS lists signs of dehydration that need urgent help, including feeling unusually tired or a child seeming drowsy, feeling dizzy on standing and it not going away, dark yellow pee or peeing less than normal, and breathing quickly or a fast heart rate. It treats confusion, being sleepier than normal or hard to wake, difficulty breathing, and blue, grey, pale or blotchy skin as an emergency. Wherever you live, those are signs to contact a doctor or your local emergency number, not to compare flavours.

How to compare any two rehydration drinks

The method on this page works for any pair of products, and it takes a minute with the two labels:

Match the volumes. Convert both labels to the same amount of drink, per litre or per 16 fl oz, before comparing anything.
Read sodium, potassium and sugars side by side. Those three are where rehydration drinks differ most.
Check the flavour and the line. Numbers vary between flavours and between a brand's original, sugar-free and other versions.
Read the maker's directions. Who the product is for, how to mix it, and when the maker says to see a doctor are all on the label.
Work out price per equal volume, not per pack.

Sources

Abbott Nutrition, Pedialyte Classic product page (nutrition per 12 fl oz and per litre, ingredients, directions, flavours): https://www.abbottnutrition.com/our-products/pedialyte-classic
Pedialyte, Why Pedialyte (brand history): https://www.pedialyte.com/why-pedialyte
Liquid I.V., Lemon Lime Hydration Multiplier product page (mixing directions, ingredients, FAQ, company details) and the nutrition-facts panels on the Hydration Multiplier product pages: https://www.liquid-iv.com/products/lemon-lime-hydration-multiplier
Liquid I.V., Key ingredients: https://www.liquid-iv.com/pages/ingredients
World Health Organization, Diarrhoeal disease fact sheet: https://www.who.int/news-room/fact-sheets/detail/diarrhoeal-disease
World Health Organization, The Treatment of Diarrhoea: A manual for physicians and other senior health workers (2005), Annex 2, Table A: https://iris.who.int/server/api/core/bitstreams/df59ceab-2498-4a64-bd93-ad7566addb4e/content
NHS, Dehydration: https://www.nhs.uk/conditions/dehydration/

Checked September 2026.

Stoicism vs. Mindfulness: What's the Difference and How Can They Complement Each Other?

Hacks Vitae — Tue, 09 Jan 2024 18:44:31 GMT

Stoicism vs. Mindfulness

Stoicism and mindfulness are often mentioned together, and they do overlap: both train attention, and both locate a good deal of our distress in how we meet events rather than in the events alone. Where they part is in what you do once you have noticed a thought. Below is where each one comes from, what the research says, how they differ, and how people combine them.

Where each one comes from

Stoicism was founded by Zeno of Citium in Athens around 300 BC. It takes its name from the Stoa Poikile, the "painted porch", a colonnade on the north side of the Athenian agora where Zeno and his followers met; unlike the Epicureans, Zeno taught in a public space. No complete works survive from the early Stoics. What most people read today comes from the Roman period: Seneca, a statesman and dramatist; Epictetus, who was born into slavery; and Marcus Aurelius, a Roman emperor whose Meditations were private notes to himself. Stoicism is a philosophy: an account of how to live and of what is good.

Mindfulness, in the form most people now meet it, comes from Buddhist meditation. The Pali word sati, usually translated as "mindfulness", literally means "memory" or "retention"; the scholar T. W. Rhys Davids first rendered it in English as "mindfulness" in 1881.

The secular version found in apps and clinics traces largely to Jon Kabat-Zinn. In 1979 he founded the Stress Reduction Clinic at the University of Massachusetts Medical School, where he adapted Zen, vipassana, hatha yoga and Advaita Vedanta teachings into an eight-week course he later named Mindfulness-Based Stress Reduction (MBSR). He removed the salvation-oriented goals of the religious traditions it drew on, and presented it in a scientific context. His widely quoted description of mindfulness is paying attention on purpose, in the present moment, and without judging, to experience as it unfolds.

So one is an ancient ethical system with practices attached, and the other an ancient practice with its religious goals set aside. That goes a long way towards explaining the differences below.

What the evidence shows

One large review is Goyal and colleagues, JAMA Internal Medicine, 2014. It pooled 47 randomised trials with 3,515 participants, each comparing a meditation programme with an active control, in adults with a range of health conditions. It found:

Moderate evidence that mindfulness meditation programmes improved anxiety, depression and pain.
Low evidence that they improved stress or distress and mental-health-related quality of life.
Low evidence of no effect, or insufficient evidence of any effect, of meditation programmes on positive mood, attention, substance use, eating habits, sleep and weight.
No evidence that meditation programmes were better than active treatments such as drugs, exercise and other behavioural therapies.

The authors summed it up as "small to moderate reductions of multiple negative dimensions of psychological stress".

A later review points the same way. Galante and colleagues (PLOS Medicine, 2021) pooled 136 trials with 11,605 participants, looking at mindfulness-based programmes in nonclinical settings. Compared with no intervention, the programmes improved average anxiety, depression, distress and well-being "in most but not all scenarios". Compared with specific active control conditions, the review found no statistically significant evidence that mindfulness did better, and the authors rated their confidence as moderate to very low. When the trials at highest risk of bias were left out, only the effect on distress, compared with no intervention, remained.

Meditation is not free of side effects for everyone. A 2020 systematic review by Farias and colleagues in Acta Psychiatrica Scandinavica covered 83 studies and found an overall prevalence of adverse events of 8.3%: 3.7% in experimental studies and 33.2% in observational ones. The most common were anxiety and depression, and the authors note they may occur in people with no previous history of mental health problems.

Stoicism has not been tested in the same way. It is a philosophy rather than a set programme, and we did not find a systematic review of Stoic training comparable to the mindfulness reviews (we searched PubMed in September 2026 for reviews of Stoic interventions, programmes or training). The closest is a 2025 scoping review of Stoic teachings in nursing and allied health, which found 28 studies, two of them interventional, and summarised them by theme rather than pooling results. The studies that do exist are small. In one from 2022, 24 third-year medical students took 12 days of online training based on Stoic philosophy, and their scores were compared before and after the course; resilience and empathy scores rose, though the authors describe the numerical results as mixed.

Stoicism's clearest link to tested therapy is indirect. Stoic philosophy was the original philosophical inspiration for modern cognitive psychotherapy, particularly through Albert Ellis's rational emotive behaviour therapy, the major forerunner of cognitive behavioural therapy (CBT); and the cognitive therapy manual for depression by Aaron T. Beck and colleagues traces the philosophical origins of cognitive therapy back to the Stoics.

Understanding the Core Principles

Stoicism

What is in our power, and what is not. Epictetus opens the Enchiridion with it: "There are things which are within our power, and there are things which are beyond our power." Within our power, he says, are opinion, aim, desire and aversion; beyond it are body, property, reputation and office (Thomas Wentworth Higginson's translation, as are the other Epictetus quotations here).
Judgements, not events, disturb us. "Men are disturbed not by things, but by the views which they take of things." (Enchiridion 5)
Virtue as the only good. The Stoics taught that "virtue is the only good" for human beings, and practised four cardinal virtues: prudence, fortitude, temperance and justice.
Accepting what cannot be changed. "Demand not that events should happen as you wish; but wish them to happen as they do happen, and you will go on well." (Enchiridion 8)
Attention. The Stoics had their own practice of attention: Epictetus discusses prosochē, attention, an idea also found in Seneca and Marcus Aurelius.

A note on one label. The Latin phrase amor fati, "love of fate", is often attached to Stoicism, but it is known above all from Nietzsche, who used it in section 276 of The Gay Science (1882) and again in Ecce Homo. The idea is often linked to Epictetus and Marcus Aurelius, but Marcus did not use those words: he wrote in Greek. The attitude is close to Stoic; the label came later. There is more on this on our Stoicism page.

Mindfulness

Present-moment awareness: paying attention, on purpose, to what is happening now.
Observing without judging: noticing sensations, thoughts and emotions as they come, without judging them.
Practices: watching the breath and the body scan are among the common exercises, and MBSR combines mindfulness meditation, body awareness and yoga.

Key Differences

This is our reading of the sources above.

What you do with a thought. Mindfulness asks you to notice it and let it be. Epictetus asks you to examine it: "Seek at once, therefore, to be able to say to every unpleasing semblance, 'You are but a semblance and by no means the real thing.' And then examine it by those rules which you have; and first and chiefly by this: whether it concerns the things which are within our own power or those which are not; and if it concerns anything beyond our power, be prepared to say that it is nothing to you." (Enchiridion 1)
Ethics. Stoicism comes with a view of what is good: virtue. Secular mindfulness, as MBSR presents it, is a skill with its religious goals set aside.
Evidence. Mindfulness programmes have been tested in dozens of trials; Stoicism, mostly, has not.

Complementary Benefits

The two meet at one point. Marcus Aurelius put it this way: "If thou art pained by any external thing, it is not this thing that disturbs thee, but thy own judgment about it. And it is in thy power to wipe out this judgment now." (Meditations 8.47, George Long's translation)

Mindfulness is a way of noticing the judgement as it forms. The Stoic question gives you something to do with it next: is this about something within my power or not? In the 2022 Stoic training study, one of the themes the researchers drew from the students' interviews was that "Stoic mindfulness encourages students to think about how they think and feel."

Integrating Both Practices

Some ways people combine the two, described here rather than prescribed:

Noticing, then asking. A few minutes of watching the breath, followed by Epictetus's question about whatever came up: is this within my power?
Writing notes to oneself. The Meditations are exactly that: Marcus wrote them as a source for his own guidance.
Rehearsing difficulty in advance. Epictetus: "Let death and exile, and all other things which appear terrible, be daily before your eyes, but death chiefly; and you will never entertain an abject thought, nor too eagerly covet anything." (Enchiridion 21) The training course in the 2022 study promoted a related practice it called negative visualisation.
Reading the sources themselves. Higginson's Enchiridion and Long's Meditations are both free on Project Gutenberg.

Nothing here is medical advice. If anxiety or depression is affecting your life, a doctor or another qualified professional is the right person to talk to. The authors of the 2014 review wrote that clinicians "should be prepared to talk with their patients about the role that a meditation program could have in addressing psychological stress."

Sources

Wikipedia, "Stoicism" (Zeno, the Stoa Poikile, the surviving texts, the cardinal virtues, the link to REBT and cognitive therapy).
Wikipedia, "Epictetus", "Seneca the Younger" and "Meditations".
Wikipedia, "Sati (Buddhism)", "Mindfulness", "Jon Kabat-Zinn" and "Mindfulness-based stress reduction".
Wikipedia, "Amor fati".
Epictetus, The Enchiridion, translated by Thomas Wentworth Higginson (Project Gutenberg ebook 45109), sections 1, 5, 8 and 21.
Marcus Aurelius, Meditations, translated by George Long (Project Gutenberg ebook 15877), 8.47.
Goyal M. et al., "Meditation programs for psychological stress and well-being: a systematic review and meta-analysis", JAMA Internal Medicine 174(3), 2014 (PubMed 24395196).
Galante J. et al., "Mindfulness-based programmes for mental health promotion in adults in nonclinical settings", PLOS Medicine 18(1), 2021 (PubMed 33428616).
Farias M. et al., "Adverse events in meditation practices and meditation-based therapies: a systematic review", Acta Psychiatrica Scandinavica 142(5), 2020 (PubMed 32820538).
Brown M. E. L. et al., "Can stoic training develop medical student empathy and resilience? A mixed-methods study", BMC Medical Education 22, 2022 (PubMed 35505329).
Boostani A. et al., "Application of Stoic Philosophy Teachings in Nursing and Allied Health: A Scoping Review", SAGE Open Nursing 11, 2025 (PubMed 41458012).

Checked September 2026. The history rests on Wikipedia's articles on Stoicism, mindfulness, sati and Jon Kabat-Zinn; the quotations on Higginson's and Long's translations; and the evidence on the Goyal (2014), Galante (2021), Farias (2020), Brown (2022) and Boostani (2025) papers, read as their PubMed abstracts, with Galante's risk-of-bias finding from its full text.

18 Timeless Stoicism Quotes to Navigate Life's Challenges with Wisdom and Grace

Hacks Vitae — Sun, 17 Dec 2023 20:18:18 GMT

18 Timeless Stoicism Quotes

These are eighteen lines from four Stoic philosophers: Epictetus, who was a slave in Rome and later taught at Nicopolis in Epirus; Marcus Aurelius, the Roman emperor who wrote the Meditations; Seneca, who wrote letters of advice to his friend Lucilius; and Zeno of Citium, who founded the school.

Each quote is given in a named English translation that is now in the public domain, with the book and section number, so you can open the full text and read what comes before and after. Where a famous English line is not what the translation says, we go with the translation.

1. "There are things which are within our power…" — Epictetus

"There are things which are within our power, and there are things which are beyond our power."

Epictetus, Enchiridion 1, translated by Thomas Wentworth Higginson

The first line of the Enchiridion, the short handbook of Epictetus's teaching. He goes straight on to say what falls on each side: within our power are opinion, aim, desire and aversion; beyond it are body, property, reputation and office. Most of the other lines on this page build on that split.

2. "No longer talk at all about the kind of man that a good man ought to be…" — Marcus Aurelius

"No longer talk at all about the kind of man that a good man ought to be, but be such."

Marcus Aurelius, Meditations 10.16, translated by George Long

It is written as an order, in the second person: stop describing what a good man is and be one.

3. "Externals are not in my power: will is in my power" — Epictetus

"Thus then in life also the chief business is this: distinguish and separate things, and say, Externals are not in my power: will is in my power."

Epictetus, Discourses 2.5, translated by George Long

The chapter's title is "How magnanimity is consistent with care". Epictetus compares life to a game of dice: the throw is not yours, but how carefully you play it is.

4. "Things do not touch the soul" — Marcus Aurelius

"Things do not touch the soul, for they are external and remain immovable; but our perturbations come only from the opinion which is within."

Marcus Aurelius, Meditations 4.3, translated by George Long

This comes in a passage about retreat. People look for getaways in the country or by the sea, Marcus writes, when "it is in thy power whenever thou shalt choose to retire into thyself".

5. "If it is not right, do not do it…" — Marcus Aurelius

"If it is not right, do not do it: if it is not true, do not say it."

Marcus Aurelius, Meditations 12.17, translated by George Long

One of the shortest entries in the book. Long prints a broken-off clause after it, "[For let thy efforts be--]", and notes that "there is something wrong here, or incomplete".

6. "That which is an obstacle on the road helps us on this road" — Marcus Aurelius

"The mind converts and changes every hindrance to its activity into an aid; and so that which is a hindrance is made a furtherance to an act; and that which is an obstacle on the road helps us on this road."

Marcus Aurelius, Meditations 5.20, translated by George Long

Marcus is talking about people who get in his way. They may impede his action, he writes, "but they are no impediments to my affects and disposition".

7. "Lay hold of to-day's task" — Seneca

"Hold every hour in your grasp. Lay hold of to-day's task, and you will not need to depend so much upon to-morrow's."

Seneca, Letters to Lucilius 1.2, translated by Richard M. Gummere

The first letter of the collection is about time. In the next sentence Seneca writes: "Nothing, Lucilius, is ours, except time."

8. "We suffer more often in imagination than in reality" — Seneca

"There are more things, Lucilius, likely to frighten us than there are to crush us; we suffer more often in imagination than in reality."

Seneca, Letters to Lucilius 13.4, translated by Richard M. Gummere

Note the words "more often": Seneca's claim is about how often fear runs ahead of the facts, not a promise that every fear is groundless. He goes on: "Some things torment us more than they ought; some torment us before they ought; and some torment us when they ought not to torment us at all."

9. "We should strive, not to live long, but to live rightly" — Seneca

"We should strive, not to live long, but to live rightly; for to achieve long life you have need of Fate only, but for right living you need the soul."

Seneca, Letters to Lucilius 93.2, translated by Richard M. Gummere

Seneca is answering a letter in which Lucilius lamented the death of the philosopher Metronax, as if he ought to have lived longer.

10. "If thou art doing thy duty" — Marcus Aurelius

"Let it make no difference to thee whether thou art cold or warm, if thou art doing thy duty; and whether thou art drowsy or satisfied with sleep; and whether ill-spoken of or praised; and whether dying or doing something else."

Marcus Aurelius, Meditations 6.2, translated by George Long

This is the real Stoic line on duty. The better-known "Do your duty, and leave the rest to the gods" is not from Marcus; see the section below.

11. "You need a change of soul rather than a change of climate" — Seneca

"You need a change of soul rather than a change of climate."

Seneca, Letters to Lucilius 28.1, translated by Richard M. Gummere

Written to Lucilius after long travels had not lifted his gloom. In the next section Seneca repeats a remark he credits to Socrates: "Why do you wonder that globe-trotting does not help you, seeing that you always take yourself with you?"

12. "The soul is dyed by the thoughts" — Marcus Aurelius

"Such as are thy habitual thoughts, such also will be the character of thy mind; for the soul is dyed by the thoughts."

Marcus Aurelius, Meditations 5.16, translated by George Long

The image is of cloth taking a colour. What you think about again and again, he says, becomes what you are.

13. "Two ears and only one mouth" — Zeno of Citium

"The reason why we have two ears and only one mouth is that we may listen the more and talk the less."

Zeno of Citium, as reported by Diogenes Laertius, Lives of the Eminent Philosophers 7.23, translated by R. D. Hicks

No complete work by Zeno survives. This line comes down to us because Diogenes Laertius, a later writer, collected stories and sayings about him. In the story Zeno says it to a young man who was talking nonsense.

14. "He is free who lives as he wishes to live" — Epictetus

"He is free who lives as he wishes to live; who is neither subject to compulsion nor to hindrance, nor to force."

Epictetus, Discourses 4.1, translated by George Long

The opening of the chapter "About freedom". The better-known "No man is free who is not master of himself" is not in the Discourses or the Enchiridion: W. A. Oldfather's translation prints it as Fragment 35, in the section he headed "Doubtful and spurious fragments". This chapter is the traceable version of the same idea.

15. "It is circumstances (difficulties) which show what men are" — Epictetus

"It is circumstances (difficulties) which show what men are."

Epictetus, Discourses 1.24, translated by George Long

The chapter goes on to compare a hard situation to a wrestling trainer who has matched you with a rough young opponent, so that you can become an Olympic winner, which "is not accomplished without sweat".

16. "Life in agreement with nature" — Zeno of Citium

"Zeno was the first (in his treatise On the Nature of Man) to designate as the end 'life in agreement with nature' (or living agreeably to nature), which is the same as a virtuous life."

Diogenes Laertius, Lives of the Eminent Philosophers 7.87, translated by R. D. Hicks

This is the Stoics' definition of the goal of life, and it is reported, not quoted: Zeno's treatise does not survive. "Nature" here does not mean the countryside. The section just before explains that for rational beings, "life according to reason rightly becomes the natural life", which is why it equals a virtuous life.

17. "Men are disturbed not by things…" — Epictetus

"Men are disturbed not by things, but by the views which they take of things."

Epictetus, Enchiridion 5, translated by Thomas Wentworth Higginson

His example follows at once: death is nothing terrible, or it would have seemed so to Socrates. The terror is in the opinion that death is terrible.

18. "Take away thy opinion…" — Marcus Aurelius

"Take away thy opinion, and then there is taken away the complaint, 'I have been harmed.' Take away the complaint, 'I have been harmed,' and the harm is taken away."

Marcus Aurelius, Meditations 4.7, translated by George Long

The same thought as number 17, from a different writer: the event happens, but the sense of injury comes from the judgment you add to it.

Three famous lines not found in Stoic texts

These three lines are sometimes presented as Stoic. None of them can be traced to a Stoic text:

"Do your duty, and leave the rest to the gods." This is from Pierre Corneille's tragedy Horace, the last line of Act II, spoken by the elder Horace: Faites votre devoir, et laissez faire aux Dieux. The Stoic passage on duty is Meditations 6.2, number 10 above.
"The good life is one inspired by love and guided by knowledge." This is Bertrand Russell, in his 1925 book What I Believe, chapter "The Good Life".
"You have power over your mind — not outside events. Realize this, and you will find strength." Credited to Marcus Aurelius, it is not in George Long's translation of the Meditations, and we have found no ancient source for it. The real passage closest to it is Meditations 8.47: "If thou art pained by any external thing, it is not this thing that disturbs thee, but thy own judgment about it. And it is in thy power to wipe out this judgment now."

Navigate Life's Challenges with Wisdom and Grace

A few ways to get more out of lines like these:

Check the number. A real Stoic quote can be traced to a book and a section. If a line comes with only a name, search a public-domain translation of that author for a distinctive phrase. If it is not there, treat the attribution as unproven.
Read the paragraph around it. Several of the lines above mean something narrower in context: Seneca's line about fear is about frequency, and Marcus's "obstacle" is about people who block him.
Try the sorting exercise. The core Stoic habit, in Epictetus's own terms, is to take a worry and ask which part of it is within your power (your opinion, aim, desire and aversion) and which is not (body, property, reputation, office). The Stoics' claim is that effort belongs on the first list.
Expect older English. Long's translations use "thou" and "thy". Newer translations read more smoothly, but when a modern line and an old translation disagree, compare both with the passage number before you decide which to trust.

Sources

Epictetus, The Enchiridion, translated by Thomas Wentworth Higginson, Project Gutenberg #45109, chapters 1 and 5.
Epictetus, The Discourses of Epictetus; with the Encheiridion and Fragments, translated by George Long (1877), Wikisource: Book 1, Chapter 24, Book 2, Chapter 5, Book 4, Chapter 1.
Epictetus, Fragments, translated by W. A. Oldfather, Wikisource, fragment 35, under "Doubtful and spurious fragments".
Marcus Aurelius, Meditations, translated by George Long, Project Gutenberg #15877: 4.3, 4.7, 5.16, 5.20, 6.2, 8.47, 10.16, 12.17.
Seneca, Moral Letters to Lucilius, translated by Richard M. Gummere, Wikisource: letters 1, 13, 28, 93.
Diogenes Laertius, Lives of the Eminent Philosophers, Book VII, translated by R. D. Hicks, Wikisource, sections 23 and 87.
Pierre Corneille, Horace, Act II, in Œuvres, ed. Marty-Laveaux (1862), vol. 3, French Wikisource.
Bertrand Russell, What I Believe (1925), chapter II, Project Gutenberg #73782.

Checked September 2026.

Memory Reboot Meaning: Reclaiming Clarity and Freshness

Hacks Vitae — Tue, 12 Dec 2023 06:45:52 GMT

If you searched "memory reboot meaning", you are probably here for one of two things: the song, or the phrase. This page covers both, starting with the song.

The song: "Memory Reboot" by VØJ & Narvent

"Memory Reboot" was released on 13 January 2023, credited to VØJ & Narvent, who are also listed as its producers. Official slowed and "over slowed" versions came out on the same day, and a sped-up version followed on 3 March 2023. An album also called Memory Reboot, holding the original, those three versions and five other tracks, was released on 1 September 2023.

Apple Music files the single under Electronic. On Narvent's own Bandcamp page, the album's tags include synthwave, phonk, darkwave, cyberpunk and blade runner 2049.

It charted in the US and the UK:

Chart	Result
US Hot Dance/Electronic Songs (Billboard)	debuted at No. 42 on the chart dated 5 August 2023; peaked at No. 13 on the chart dated 30 December 2023
US Hot Dance/Electronic Songs, 2023 year-end (Billboard)	No. 47
UK Independent Singles Breakers (Official Charts Company)	peaked at No. 20, two weeks in September 2023
Certification	Gold from the RIAA, as listed on Billboard's chart page

"You look lonely. I can fix that."

This line is dialogue from Blade Runner 2049 (2017). In the film, a giant hologram advertisement of Joi, played by Ana de Armas, speaks to the replicant K (Ryan Gosling): "You look lonely. I can fix that." In the scene, Joi is a product, and this is her advertisement.

It is film dialogue, not a lyric we could find. Whether the line appears in the officially released recording is something we could not confirm. What we can say is that the connection is not only a listener's invention: Narvent's own Bandcamp page tags the album blade runner 2049.

So what is the song "about"?

We found no statement of meaning from either artist. Listeners may hear it as a song about lost love or revisiting the past. That describes how it makes them feel, which is a fair response, though it is not the same as a stated meaning.

One reading, and it is only a reading: the title and the Blade Runner tag both point toward memory and distance.

The phrase, in everyday use

Outside the song, "memory reboot" is used in two ordinary ways.

Literally: refreshing your recall

Revisiting old photos, notes or places to bring a memory back into focus. It also happens without being asked for, when a smell or a piece of music drops you into a moment you had not thought about in years.

About the computer comparison. People sometimes describe this as "clearing your mental RAM", by analogy with restarting a computer. It is a figure of speech, not a description of how memory works: nothing is wiped and reloaded. Sleep does play a part in memory, but not as a restart. The US National Institute of Neurological Disorders and Stroke puts it this way: memory consolidation "most likely requires both non-REM and REM sleep".

Metaphorically: a fresh start with the past

Used this way, a memory reboot means deciding to stop being governed by an old story:

Letting go of a past hurt or a grudge.
Questioning a belief about yourself that has never been examined.
Coming back to something with a fresh perspective.

Where the idea gets used

Personal development: revisiting an experience in order to learn from it.
Creativity: mining your own past for material.
Relationships: re-reading an old conflict for what the other person may have meant.
Mental health: memories that are intrusive or distressing are something clinicians treat. The NHS, for example, lists trauma-focused cognitive behavioural therapy and EMDR among the treatments for post-traumatic stress disorder, planned with a GP and a mental health specialist.

Putting the idea to use

Used as a metaphor, the phrase points at a few ordinary habits:

Writing things down. A note or a journal is a record that lives outside your head.
Changing the setting. Revisiting a memory somewhere new, or talking it through with someone.
Reframing rather than erasing. A memory reboot is not deletion; it is deciding what a memory is allowed to mean now.

What we could not confirm

Any statement of meaning from VØJ or Narvent. If one exists, we could not find it.
Whether the Blade Runner 2049 line is in the officially released recording. We could not listen to the releases for this check, and the store and chart pages we read carry no sample credit.
The song's 2024 year-end position. Billboard's public 2024 year-end page lists the top 50, and "Memory Reboot" is not among them.

Sources

Billboard, Hot Dance/Electronic Songs, chart dated 13 January 2024 (entry for "Memory Reboot": debut, peak, credits, RIAA certification).
Billboard, Hot Dance/Electronic Songs, 2023 year-end and 2024 year-end.
Official Charts Company, Memory Reboot by VØJ & Narvent, chart history.
Apple, iTunes lookup for the single and the slowed and sped-up versions (release dates, genre).
Narvent, Memory Reboot album page on Bandcamp (track list, release date, tags).
Blade Runner 2049 (2017): Know Your Meme, You Look Lonely (the scene and the advert's dialogue); dialogue transcript at Scripts.com; Ian Failes, More Joi: how the giant 'Blade Runner 2049' hologram shot was pulled off, vfxblog, 24 October 2017.
National Institute of Neurological Disorders and Stroke, Brain Basics: Understanding Sleep.
NHS, Post-traumatic stress disorder (PTSD): treatment.

Checked September 2026.

MOASS Meaning: Mother of All Short Squeezes

Hacks Vitae — Sat, 09 Dec 2023 22:18:59 GMT

What MOASS means

MOASS stands for "Mother Of All Short Squeezes". People use it for a short squeeze bigger than any that has happened before. Since January 2021 it has been used above all about GameStop.

The older phrase has been used after the fact, for squeezes that happened (see below). In the GameStop era, MOASS names a squeeze people expect, not one that has happened: a prediction. This page explains the term and what the US Securities and Exchange Commission's staff found about the trading it grew out of. It does not say whether any stock will squeeze, and nothing here is investment advice.

Understanding Short Squeezes

Short selling is selling a stock you do not own. The usual way is to borrow shares, sell them, and hope to buy them back later at a lower price to return to the lender. The SEC staff report on 2021 points out why this is risky: a stock's price can in principle keep rising, so a short seller "can lose more than the value of their original investment". Brokers usually require collateral in a margin account of at least 50% of the shorted position, on top of the cash from the sale.

A short squeeze is what can happen when a heavily shorted stock rises fast. In the report's words, if the price jumps, short sellers "would face margin calls requiring them either to post additional collateral or to exit their position". Those who exit have to buy the stock back, and that buying pushes the price up further, which can force out other short sellers in turn.

A real example: on 28 October 2008, Reuters reported that Volkswagen had briefly become the world's biggest company by market value, as short sellers betting on a fall scrambled to find shares "after a buying spree by Porsche".

Where the phrase came from

The phrase is older than GameStop. The etymologist Barry Popik found "the mother of all short squeezes" in the Star Tribune of Minneapolis on 16 February 1994, in a money manager's prediction about NordicTrack's stock, and in USA Today in 1995 about the bond market. The Reuters report on Volkswagen in 2008 quotes an analyst calling it "the mother of all short squeezes". Popik traces the "mother of all" formula to the Gulf War of 1990–91.

The acronym MOASS appears in social-media posts from 2009 onward, by Popik's record, and it went viral in January 2021 with GameStop.

What the SEC staff found about GameStop in 2021

In October 2021 the SEC staff published the Staff Report on Equity and Options Market Structure Conditions in Early 2021. It is the SEC staff's account of the episode, and it is a staff report: its own disclaimer says the Commission "has expressed no view regarding the analysis, findings, or conclusions contained herein".

The price. GameStop (GME) had ended 2020 at a little under $20 a share. By 27 January 2021 it closed at $347.51, more than 1,600% above its close on 11 January, and the next day it reached an intraday high of $483.00. By 19 February it had fallen as low as $40.59.

Short interest. GameStop's short interest hovered around 100% of shares outstanding from 2019 to early 2021, peaking at 109.26% on 31 December 2020. Measured against the public float, it reached 122.97% in January 2021. The two numbers differ because they divide by different things: all shares outstanding, or only the shares available to trade.

Was it a short squeeze?

Partly, at moments. The staff found discrete periods when GME rose sharply as known major short sellers covered their positions, and that "during these times, short sellers covering their positions likely contributed to increases in GME's price."

But the report's overall finding runs the other way. Buying by traders with large short positions was "a small fraction of overall buy volume", and the price stayed high "after the direct effects of covering short positions would have waned". Its conclusion:

"Whether driven by a desire to squeeze short sellers and thus to profit from the resultant rise in price, or by belief in the fundamentals of GameStop, it was the positive sentiment, not the buying-to-cover, that sustained the weeks-long price appreciation of GameStop stock."

Elsewhere the staff put it more briefly: "a short squeeze did not appear to be the main driver of events".

Not everyone who studied the trading agrees. In February 2022 six academics writing as the Ad Hoc Academic Committee on Equity and Options Market Structure Conditions in Early 2021 (Joshua Mitts, Robert Battalio, Jonathan Brogaard, Matthew Cain, Lawrence Glosten and Brent Kochuba) published a critique of the staff's method. The staff built their list of short sellers from trading records that begin on 24 December 2020 and kept only the more heavily shorted half; the academics argue that this start date means the method counts "zero short positions on December 23, 2020", and that keeping only the more heavily shorted half may leave out a large volume of covering purchases. Extending the sample and adding securities-lending data, they report "evidence that a nontrivial fraction of the trading volume in GameStop's stock consisted of purchases by short sellers covering their positions".

Where that leaves it: open. Both sides agree that short covering pushed the price up at some moments. They disagree on how much of the whole rise it explains. The staff had non-public, deanonymised data the academics could not see; the academics used a longer window and lending records the staff left out, and they say themselves that without the staff's data they "are unable to pin down the magnitude of a short squeeze or gamma squeeze". Neither account settles the size of the squeeze.

Short interest above 100%

Short interest can pass 100% through ordinary share lending. The report explains that the same shares can be "lent multiple times by successive purchasers": if someone buys a share from a short seller and lends it out again, it counts as sold short twice. GME was the only stock the staff saw with short interest above its shares outstanding in January 2021.

Naked short selling and fails to deliver

A "naked" short sale is one made without arranging to borrow the shares, so the seller fails to deliver them on time. The staff did see spikes in GME's fails to deliver. But fails can come from ordinary (long) sales as well as short sales, which makes them "an imperfect measure of naked short selling", and GME "did not experience persistent fails to deliver at the individual clearing member level". Most clearing members cleared their fails within a few days.

Gamma squeeze

A gamma squeeze happens when market makers who have written call options buy the stock to hedge them, pushing its price up. The staff "did not find evidence of a gamma squeeze in GME during January 2021". The jump in options trading by individual customers was mostly in put options, not calls, and market makers were buying call options rather than writing them. The academic critique above argues that this test is too narrow: "what matters is not the type of option at issue – put or call – but rather the direction of the hedging", so market makers hedging put options can also end up buying the stock.

Why brokers restricted buying

On 28 January 2021 several brokerages restricted trading in GME. The report ties this to the clearing system. On 27 January, the National Securities Clearing Corporation (NSCC) made intraday margin calls totalling $6.9 billion on 36 clearing members. The report notes that NSCC's rules do not let it tell firms to stop trading particular stocks; some brokers chose to restrict trading "in reaction to margin calls and capital charges imposed by NSCC". It records that witnesses at a Congressional hearing testified that the restrictions did not come from pressure by hedge funds.

Not just GameStop. AMC, another stock caught up in the January 2021 trading restrictions, had short interest of 11.4% of its float in January 2021, against GME's 122.97%. And on 27 January, four other stocks, AMC among them, had bigger one-day price rises than any single-day rise in GME.

What changed afterwards

The report's conclusions named four areas for further consideration. Under the first, why brokers restricted trading, it suggested shortening the settlement cycle; under the fourth, short selling, it called for better reporting of short sales and said "the interplay between shorting and price dynamics is more complex than these narratives would suggest." On 15 February 2023 the SEC adopted rules shortening the standard settlement cycle for most broker-dealer trades from two business days after the trade (T+2) to one (T+1), with a compliance date of 28 May 2024. SEC Chair Gary Gensler said the change addressed one of the four areas the staff recommended "in response to the meme stock events of 2021".

How to read a squeeze claim

The 2021 record gives a few questions worth asking of any post that says a squeeze is coming:

Short interest of what, and on what date? Percent of float and percent of shares outstanding are different numbers, and the short-interest figures the SEC staff used are reported by the exchanges every two weeks, so a figure can be out of date.
Does above 100% prove something illegal? Not by itself: the SEC staff showed how the same shares, lent again, are counted twice.
Who was buying? In January 2021, covering by short sellers was “a small fraction of overall buy volume” by the SEC staff's count; the academics found “a nontrivial fraction” of trading volume and argued the staff's method may have missed a large volume of it. When two studies of the same trades disagree, the size is not settled.
Is a fail to deliver the same as a naked short? No: the staff called fails an imperfect measure, since they also come from ordinary sales.
Who is making the claim, and what do they hold? A prediction from someone who owns the stock is not a finding from someone who studied the trades.

This page describes what happened and what the regulator's staff found. It does not recommend buying, selling or holding anything.

Sources

U.S. Securities and Exchange Commission staff, Staff Report on Equity and Options Market Structure Conditions in Early 2021 (14 October 2021), sections 1, 3.2, 3.4–3.6 and 4: https://www.sec.gov/files/staff-report-equity-options-market-struction-conditions-early-2021.pdf
U.S. Securities and Exchange Commission, "SEC Finalizes Rules to Reduce Risks in Clearance and Settlement", press release 2023-29 (15 February 2023): https://www.sec.gov/newsroom/press-releases/2023-29
Sarah Marsh, "Short sellers make VW the world's priciest firm", Reuters (28 October 2008), archived copy: https://web.archive.org/web/20210123004118/https://www.reuters.com/article/us-volkswagen/short-sellers-make-vw-the-worlds-priciest-firm-idUSTRE49R3I920081028
Joshua Mitts, Robert H. Battalio, Jonathan Brogaard, Matthew D. Cain, Lawrence R. Glosten and Brent Kochuba, "An Academic Critique of the SEC's GameStop Report", CLS Blue Sky Blog, Columbia Law School (22 February 2022): https://clsbluesky.law.columbia.edu/2022/02/22/an-academic-critique-of-the-secs-gamestop-report/
Barry Popik, "Mother Of All Short Squeezes (MOASS)", The Big Apple (31 January 2021): https://www.barrypopik.com/new_york_city/entry/mother_of_all_short_squeezes/

Checked September 2026.

Toma Meaning

Hacks Vitae — Sat, 09 Dec 2023 21:35:23 GMT

Toma Meaning

People reach this word from a Spanish sentence, a song lyric, a menu or a name, and each has a different answer.

In Spanish, toma is a form of the verb tomar: "take". It is both the command ("take!", "here, have this") and the present tense for he, she or you-formal ("she takes", "he drinks"). Shouted on its own, ¡toma! is closer to "take that!" or "serves you right". The same is true in Portuguese.

Outside those two languages the word is unrelated: an Italian cheese, the Romanian form of the name Thomas, and in Tagalog a drinking session.

Toma in Spanish: the verb

Tomar covers much more ground than English "take". The Real Academia Española's dictionary lists 39 senses, from "to take hold of something with the hand" to taking a bus and taking something seriously.

Toma is two forms of it, spelled the same:

The command, for tú: toma el vaso, "take the glass"; toma, es para ti, "here, it's for you".
The present tense, for él, ella and usted: ella toma el autobús, "she takes the bus".

The sentence around the word tells you which one you are reading.

The drinking sense

One of the dictionary's senses of tomar is simply "to eat or drink": tomar un desayuno, to have breakfast. The dictionary also marks a sense used in the Americas and Equatorial Guinea: to drink alcohol. So ¿qué tomas? is "what are you having?", and in those places tomar on its own can mean drinking.

As a shout: "¡Toma!"

On its own, ¡toma! is an interjection, and the dictionary gives it two colloquial uses:

pointing at something as a punishment, a comeuppance or a disillusion. Its example: ¿No te dije que corrías peligro? Pues ¡toma!, "Didn't I tell you that you were in danger? Well, take that!"
signalling that something is no news or of little importance, closer to "so what?".

A related expression, tómate esa, is said when someone takes a blow, to show they deserved it.

Figurative uses

Tomar una decisión — to make a decision. The dictionary notes that tomar with some nouns means the same as the verb they come from.
Tomar en serio — to take seriously.
Tomar el pelo — literally "to take the hair": to tease someone with fake praise, promises or flattery.

Toma as a noun

As a noun, la toma is a taking. The dictionary's senses include:

A conquest or occupation by force of a town or stronghold, as in la toma de la Bastilla, the storming of the Bastille.
A portion taken at one time, and each dose of a medicine taken by mouth.
The point where a flow of fluid or electricity is drawn off. In the Americas a power socket is a tomacorriente.
The act of photographing or filming, as in a film take.

Toma the song: Pitbull featuring Lil Jon, and the phonk tracks

Several songs share the title. The best-documented is "Toma" by Pitbull featuring Lil Jon. It is on Pitbull's 2004 album M.I.A.M.I., released by TVT Records, and it charted on Billboard's Hot Rap Songs, reaching No. 21 in March 2005.

Pitbull's "Toma" is not an Enrique Iglesias song. Pitbull and Iglesias have recorded together, and Billboard lists "Messin' Around" (2016) as Pitbull featuring Enrique Iglesias, but "Toma" is credited to Pitbull and Lil Jon.

More recent tracks use the title too, including phonk tracks: "Toma Phonk" by Mia Hill (2023) and "Toma Toma (Brazilian Phonk)" by SXBER (2024) are two. Whatever the track, a shouted "toma!" in a Spanish or Portuguese lyric carries the meanings above: the command "take it!", or "take that!".

Toma as a name

Toma is a form of Thomas. The name goes back to an Aramaic word for "twin", and the Gospel of John calls the apostle "Thomas, which is called Didymus", the Greek for twin. In the Romanian Bible the same verse reads Toma, zis Geamăn: "Toma, called the Twin".

As a surname

The Dictionary of American Family Names gives several separate origins for the surname Toma, almost all of them forms of Thomas: Romanian and Croatian from the personal name Toma; Italian, French, Polish, Slovak and Hungarian from pet forms of Thomas; Albanian from Tomë; Assyrian/Chaldean from the Syriac Tōmā; and Arabic, from a variant of Tuma, "Thomas". It also lists an unrelated Japanese surname, Tōma.

None of the origins it gives is the Spanish verb tomar.

Toma in other languages

Portuguese — the same as Spanish. Toma is the command form and the he/she form of tomar, which covers taking, eating and drinking. The Priberam dictionary also lists toma! as an informal interjection showing satisfaction at someone's defeat or setback, with the example toma, foi bem feito o que lhe aconteceu: "take that, what happened to him served him right".
Italian — not a verb. The Treccani dictionary lists toma as a cheese typical of Piedmont and the Aosta Valley, made from cow's, sheep's or mixed milk, and as a word found only in the old saying promettere Roma e toma, "to promise great things", and a few like it. One variety, Toma Piemontese, is a protected designation of origin, and its rules require cow's milk.
Romanian — Toma is the name Thomas, as above.
Tagalog — toma is a drinking session, and a tomador is a drinker of alcohol, or a heavy drinker.

How to tell which toma you have

Ask where you found it:

In a Spanish or Portuguese sentence, check what comes after it. A thing to take (toma el vaso) means the command; a person before it (ella toma) means the present tense; a drink after it means drinking.
Alone, with exclamation marks, it is the interjection: "take that!" or "serves you right".
After la or una, it is the noun: a capture, a dose, an outlet or a take.
On a cheese board or an Italian menu, it is the cheese.
As someone's name, it is Thomas.
In a song title, check the artist: the 2004 Pitbull track with Lil Jon is one of several called "Toma".

Sources

Real Academia Española, Diccionario de la lengua española: tomar (senses, the interjection ¡toma!, tómate esa, conjugation), toma (the noun), pelo (tomar el pelo), tomacorriente.
Priberam, Dicionário da Língua Portuguesa: toma, tomar, conjugation of tomar.
Treccani, Vocabolario: toma (1) and toma (2).
Disciplinare di produzione della denominazione di origine protetta "Toma Piemontese", Gazzetta Ufficiale, 5 March 2015.
Billboard, Pitbull chart history, Hot Rap Songs.
Apple Music, M.I.A.M.I. by Pitbull and "Toma Toma (Brazilian Phonk)" by SXBER; Spotify, "Toma Phonk" by Mia Hill.
Strong's Concordance, G2381, Thomas; John 11:16 in the Cornilescu Romanian Bible.
Dictionary of American Family Names, 2nd edition (2022), via Ancestry: Toma surname meaning.
Tagalog.com dictionary: toma and tomador.

Checked September 2026.

Leper King Baldwin IV of Jerusalem: Life and Legacy

Hacks Vitae — Tue, 28 Nov 2023 05:31:16 GMT

Life and Legacy of the Leper King

The Leper King was Baldwin IV, king of Jerusalem from 1174 to 1185. He was crowned on 15 July 1174, when he was "barely thirteen years old", and he had leprosy for his whole reign. At sixteen he defeated Saladin at Montgisard. By 1183 he was blind and could no longer use his hands or feet, and he died in 1185, aged 23 or 24. Two years after his death, Jerusalem fell to Saladin.

Most of what we know about him comes from one man who knew him well: William of Tyre, his tutor, later chancellor of the kingdom and archbishop of Tyre, whose chronicle stops early in 1184. That gives us an unusually close witness. It also means the main witness was a friend with his own politics, which is worth keeping in mind below.

How the Leprosy Was Found

William took charge of the prince's education around 1170, when the boy was about nine. One day Baldwin was playing with other noble boys, and they began, "as playful boys often do, to pinch each others' arms and hands with their nails." The others cried out. Baldwin did not. When William questioned him, he found that the boy's "right arm and hand were partially numb, so that he did not feel pinching or even biting in the least."

Physicians tried "repeated fomentations, oil rubs, and even poisonous remedies", without result. As Baldwin reached his teens it became clear that he had leprosy. "The extremities and the face were especially attacked," William writes, and yet the boy became "an excellent horseman".

Was it really leprosy? In the Middle Ages the word covered several skin diseases, and some have suggested Baldwin had endemic syphilis instead. The medical historian Piers Mitchell weighed the symptoms William describes, above all the numbness and the destruction of the hands and feet, and concluded they fit lepromatous leprosy, the most severe form, better than endemic syphilis.

A Boy King and a Regent

Baldwin's father, King Amaury, died in July 1174. On 15 July, "the fourth day after the death of his father", Baldwin was consecrated and crowned in the Church of the Holy Sepulchre.

A thirteen-year-old could not rule alone. Raymond III, count of Tripoli, claimed the regency as the king's next of kin, and from then on the kingdom's politics turned on two questions: who governed while the king was ill, and who would succeed a sick king with no children.

Legacy and Impact

Montgisard, 25 November 1177. Saladin's army was raiding the kingdom while Baldwin was at Ascalon. The king, then sixteen, led his forces out along the coast, joined by the Templars from Gaza, and met Saladin before all of his scattered raiders could regroup. William names the lords with the king, among them Reynald of Châtillon and the brothers Baldwin of Ramla and Balian of Ibelin, and says their men, "including all ranks and conditions, were barely three hundred and seventy-five." After asking around, he put Saladin's force at twenty-six thousand light cavalry. Those are William's figures, and chroniclers' numbers deserve caution. William also says the king led out "all his forces, both cavalry and infantry", so the whole Christian army was larger than those 375. By his account Saladin's lines broke and the Christians chased them "for twelve miles and more".

Two details complete the picture. William was probably with the army, so this is close to an eyewitness account. And the king had appointed Reynald of Châtillon regent and commander in chief a few months earlier; William's modern translators call Reynald "the real commander" of the victory. Baldwin led, and he was there, but a sixteen-year-old king did not win the battle alone.

How he governed

Baldwin's eleven-year reign was held together by alliances, building and marriage politics as much as by battle.

He looked for allies abroad. In 1177 the kingdom had a joint campaign against Egypt "arranged with the emperor of Constantinople", Manuel I Komnenos. Count Philip of Flanders, whom the barons asked to lead it, refused to go to Egypt.
He fortified the frontier. In October 1178 he began a castle at Jacob's Ford, beyond the Jordan and ten miles from Banyas, built "within six months". Saladin took it in 1179.
He managed the succession. In April 1180 he married his sister Sibylla to Guy of Lusignan and made Guy count of Jaffa.
He changed course when it failed. In 1183, with the leprosy much worse, he named Guy regent. By November he had taken the government back, and he had Sibylla's son from her first marriage, a boy "scarcely five years old", crowned as co-king Baldwin V, cutting Guy out of the succession.

"Kingdom of Heaven" (2005): what the film gets right, and what it invents

Baldwin IV is a central character in Ridley Scott's Kingdom of Heaven, written by William Monahan, in which he is played, masked and uncredited, by Edward Norton. The film is worth watching. Several of its most famous touches are its own.

The silver mask. Neither of the two main chronicles of his reign, William of Tyre and the chronicle attributed to Ernoul, mentions a mask. William describes a face and hands attacked by the disease.
"A king may move a man, a father may claim a son…" This speech is film dialogue, spoken by Baldwin to Balian. It is not a medieval quotation.
"It is a kingdom of conscience, or nothing." In the film this is Balian's line, not Baldwin's.
Balian of Ibelin. In the film he is a blacksmith. The real Balian was a baron of the kingdom and the brother of Baldwin of Ramla. He fought at Montgisard in 1177, and in 1187 he led the defence of Jerusalem.
What the film gets right is the shape of it: a young king holding a fragile kingdom together while his body failed, with rival factions already fighting over what came next.

How historians judge the reign

There are two main ways to read his reign, and they are a matter of interpretation rather than fact.

The traditional view, as the historian Bernard Hamilton summarises it, sees the reign as "a period of decline" in which the king's illness let "unsuitable men" take power and make the wrong decisions. Hamilton's book The Leper King and His Heirs (Cambridge University Press, 2000) argues the opposite: that peace with Saladin was never a real option, that Baldwin "was an excellent battle leader", and that he "had to remain king in order to hold factions in check." It is one historian's argument, made at book length by a specialist, and the evidence above lets readers weigh both.

What does not survive William's account is the popular picture of a figurehead. Even blind, it was Baldwin who, on the advice of his counsellors and his mother, took the government back from Guy and had Baldwin V crowned.

How it ended

Kerak, 1183. Late in 1183 Saladin besieged the castle of Kerak, beyond the Jordan, in the middle of a wedding: the young Humphrey of Toron was marrying the king's younger sister, Isabella. The chronicle attributed to Ernoul adds that Humphrey's mother sent Saladin bread, wine, oxen and sheep from the wedding feast, and that Saladin ordered his men not to attack the tower where the couple were staying. The king, blind by now, took the True Cross and marched to relieve the castle, placing Raymond of Tripoli in command of the army. Saladin withdrew after a siege that William puts at "an entire month"; the chronicle attributed to Ernoul says eight weeks (six in some manuscripts).

Neither William nor Ernoul says how he travelled; William simply says he "marched thither himself". The historian Bernard Hamilton judges that he was "almost certainly too ill to ride, and would have had to be carried in a litter", and later describes him being carried into battle "in a litter slung between two horses" when he went to the relief of Kerak in 1183 and again in 1184. William does record him being carried in a litter when he was ill in 1177.

The end. Baldwin died in 1185, and the exact day is uncertain. The medical historian Piers Mitchell gives 16 March 1185; Bernard Hamilton says only that he died "at some time before 16 May 1185", "not quite twenty-four years old". Baldwin V died the following year, and Guy of Lusignan became king. On 4 July 1187 Guy's army was destroyed at Hattin, near Tiberias, and Guy was taken prisoner. Jerusalem surrendered to Saladin on 2 October 1187, after a defence led by Balian of Ibelin. Those who paid their ransom were escorted to Tyre, which was still in Christian hands, but the city Baldwin had ruled from was lost two years after his death.

What His Story Is Useful For

Baldwin's story is often told as pure inspiration. It is more useful read carefully:

Separate the chronicle from the film. The mask, the speeches and the blacksmith are Hollywood's. The pinching game, Montgisard and Kerak are William's. When you meet a moving line attributed to a medieval king, check whether it comes from a screenplay.
Ask who the witness was. William was devoted to his pupil and opposed to some of the men around him. His translators point out places where his opinions follow his politics. A single close witness is valuable, and it is still a single witness.
Look at what he built around himself. A king who did not expect to live long spent his reign on regents, alliances, castles and a successor. The lasting interest of the reign is less his courage, which is real, than the arrangements he made for a future he would not see.

Checked September 2026.

Sources

William of Tyre, A History of Deeds Done Beyond the Sea, translated by Emily Atwater Babcock and A. C. Krey (Columbia University Press, 1943), Books XXI–XXII. Internet Archive.
Chronique d'Ernoul et de Bernard le Trésorier, ed. L. de Mas Latrie (Paris, 1871). Internet Archive.
Bahā' al-Dīn Ibn Shaddād, The Life of Saladin, translated by C. W. Wilson (Palestine Pilgrims' Text Society, 1897). Internet Archive.
Piers D. Mitchell, "Leprosy and the case of King Baldwin IV of Jerusalem", International Journal of Leprosy 61 (2), 1993, pp. 283–291. Journal page.
Bernard Hamilton, The Leper King and His Heirs (Cambridge University Press, 2000), pp. 196, 210, 241. Publisher's page.
Kingdom of Heaven (2005), scene transcripts: "A king may move a man" and "It is a kingdom of conscience, or nothing".
3DVF, He isn't even in the credits, so who plays Kingdom of Heaven's masked king? (24 June 2026).

100 Reasons to Stay Alive and Embrace the Beauty of Life

Hacks Vitae — Thu, 23 Nov 2023 11:36:04 GMT

If you are struggling or thinking about suicide, please reach out now. findahelpline.com lists free, confidential helplines worldwide. In the US and Canada, call or text 988. In the UK and Ireland, call Samaritans on 116 123. In Australia, call Lifeline on 13 11 14. In India, call Tele-MANAS on 14416. In the Philippines, call the National Center for Mental Health crisis line on 1800-1888-1553 (or 1553 from a Luzon landline). If you are in immediate danger, call your local emergency number.

A list is not treatment, and this page knows it. If you are in real difficulty, the numbers above are worth more than anything below. Find A Helpline describes contacting a helpline as "free, anonymous and confidential", and says you might just need to talk: you do not have to be at a breaking point to use one.

What a list like this is good for is narrower and still real: when everything has flattened out, it can be hard to call to mind that anything is worth staying for. These are prompts, not an argument. Take the ones that land. Leave the rest — some of them will not be yours, and that is expected.

100 Reasons to Stay Alive

Some of these are big moments and most of them are small ones, on purpose: small things are often the ones within reach on a bad day. Most of the reasons follow a list that has been shared widely online; we could not find who first wrote it. The notes under each one are ours.

1. Making Your Parents Proud: The look on a parent's face when they see who you have become, or the version of that you give yourself if they are not around to see it.

2. Conquering Your Fears: The small, surprising fact of having done the thing you were sure you could not do.

3. Reuniting with Family: The first minutes back with people who knew you before you knew yourself.

4. Seeing Your Favorite Artist Live: Standing in a crowd while a song you have played a hundred times happens in the same room as you.

5. Listening to Music: The song that says it better than you could, on the day you need it said.

6. Experiencing New Cultures: Food you cannot name yet, a greeting you get slightly wrong, a place that works differently from yours.

7. Making New Friends: Someone who was a stranger in spring and knows your coffee order by winter.

8. Inspiring Others: Somebody, somewhere, getting through their own hard stretch a little more easily because you got through yours.

9. Having Your Own Children: If you want them: a small person who did not exist before and now does, partly because of you.

10. Adopting a Pet: An animal that has decided you are its person.

11. Making Yourself Proud: Achievements, big and small, that nobody else noticed but you did.

12. Meeting Your Idols: Finding out that someone you admired from a distance is a person, and sometimes a kind one.

13. Laughing Until You Cry: The joke that should not have been that funny, and the friend who could not stop either.

14. Experiencing Tears of Happiness: The kind of crying that comes from something going right.

15. Eating Your Favorite Food: The dish that tastes like a particular kitchen, or a particular person.

16. Watching Siblings Grow: Seeing a brother or sister become someone you would choose as a friend.

17. Passing School: The results you were dreading turning out to be fine, or better than fine.

18. Getting a Tattoo: Something you chose to carry, with a story only you fully know.

19. Smiling Until Your Cheeks Hurt: A day so good that your face gets tired.

20. Meeting Internet Friends: Hearing the real voice of someone you have only ever read.

21. Finding Unconditional Love: Being with someone who is glad you exist on your bad days as well as your good ones.

22. Enjoying Ice Cream on a Hot Day: The race to finish it before it melts.

23. Drinking Hot Chocolate on a Cold Day: Both hands around a warm mug while it is cold outside.

24. Seeing Untouched Snow in the Morning: A whole street before anyone has walked on it.

25. Witnessing a Sunset that Sets the Sky on Fire: The evenings when the sky does something nobody could have planned.

26. Seeing Stars Light Up the Sky: A clear night far enough from the lights to see how many there are.

27. Reading a Life-Changing Book: The book you finish as a slightly different person from the one who started it.

28. Experiencing Spring Flowers: The first colour coming back after a grey season.

29. Watching Leaves Change from Green to Brown: A season changing in front of you, slowly enough to notice.

30. Traveling Abroad: Waking up somewhere where nothing is familiar and everything is worth looking at.

31. Learning a New Language: The first joke you understand in a language that is not your own.

32. Developing Artistic Skills: Drawing, painting, making: watching your hands get better at something over months.

33. Sharing Your Story to Help Others: The day your worst stretch becomes something that helps someone else through theirs.

34. Puppy Kisses: Being greeted as though you had been gone for years.

35. Baby Kisses: Clumsy, open-mouthed affection from someone who has no reason to fake it.

36. Swearing and Release: The relief of saying exactly how you feel, loudly, somewhere nobody minds.

37. Jumping on Trampolines: Feeling ten years old again for a few minutes.

38. A Good Night's Sleep: Waking up rested after a night when sleep finally came easily.

39. Finding Out You Were Wrong About Something: The small jolt of learning that a thing you were sure of is not true, and the world getting a little bigger for it.

40. Cloud Watching: Lying on your back and letting the sky rearrange itself.

41. Enjoying a Refreshing Shower and Clean Sheets: Ten of the nicest minutes in an ordinary evening.

42. Receiving Thoughtful Gifts: Something that shows a person was paying attention.

43. “I Saw This and Thought of You": Being on someone's mind when you were not even in the room.

44. The Feeling When Someone Says, "I Love You": Hearing it from someone who means it.

45. The Relief After Crying: The quiet, lighter feeling that sometimes comes afterwards.

46. Sunshine: Warm light on your face on a day that needed it.

47. The Feeling When Someone Listens to You: Being heard all the way to the end of a sentence.

48. Your Future Wedding: If it is something you want: a day built around you and someone you chose.

49. Your Favorite Candy Bar: A small, cheap, reliable pleasure.

50. New Clothes: Wearing something for the first time and liking how you look in it.

51. Witty Puns: The groan a good bad pun gets out of a whole room.

52. Really Good Bread: Still warm, with the crust that crackles.

53. Holding Your Child for the First Time: A moment that is hard to compare with anything else.

54. Completing a Milestone: Going to college, graduating, getting married, landing the job: the days you looked forward to, arriving.

55. Dreams That Make You Smile: Waking up from a good one and trying to stay in it a little longer.

56. The Smell Before and After It Rains: The air changing before the first drop, and again after the last.

57. The Sound of Rain Against a Rooftop: Listening to weather from somewhere dry.

58. The Feeling When You're Dancing: The moment you stop thinking about how you look.

59. The People Who Mean the Most to You: The handful of people whose day would be worse without you in it.

60. Trying Out New Recipes: The first time something you cooked turns out the way the picture promised.

61. The Feeling When Your Favorite Song Comes On the Radio: Hearing it by chance, which is better than choosing it.

62. The Rush When You Step Onto a Stage: Nerves turning into something closer to excitement.

63. Sharing Your Voice, Talents, and Knowledge with the World: Things only you can say in quite the way you would say them.

64. Breakfast in Bed: A slow morning with nowhere to be.

65. Getting a Middle Seat in the Movie Theater: The lights going down, and a seat right in the middle.

66. Breakfast for Dinner: Pancakes at eight in the evening, because why not.

67. Prayer (If You Are Religious): For people of faith, a place to put what is too heavy to carry alone.

68. Forgiveness: Putting down a grudge you had been carrying longer than it deserved.

69. Water Balloon Fights: Getting soaked on purpose on a hot day.

70. New Books by Your Favorite Authors: The announcement that the next one is coming, and the wait for it.

71. Fireflies: Small lights in a dark garden, blinking on and off.

72. Birthdays: Another year, and the people who turn up to mark it.

73. Realizing That Someone Loves You: The moment you notice it was there all along.

74. Spending the Day with Someone You Love: A whole day with no plan except each other.

75. Being Wrapped Up in a Warm Bed: Cold outside, warm under the covers, no alarm set.

76. Someone's Skin Against Yours: The closeness of another person you trust.

77. Holding Hands: One of the smallest ways of saying you are not alone.

78. The Kind of Hugs That Lift a Weight Off Your Shoulders: The hug that lasts a few seconds longer than usual, because it needs to.

79. Singing Off Key with Your Best Friends: Nobody in tune, nobody caring.

80. Road Trips: The playlist, the snacks, the long stretches of road and conversation.

81. Spontaneous Adventures: Plans that were never plans at all, and turned out well.

82. The Feeling of Sand Beneath Your Toes: Warm sand, bare feet, no hurry.

83. The Feeling When the First Ocean Wave Rolls Up and Envelops Your Toes, Ankles, and Knees: The first cold shock of the sea, and then getting used to it.

84. Thunderstorms: Watching lightning from a window and counting the seconds to the thunder.

85. Your First (or Hundredth) Trip to Disneyland: Or wherever your own version of that is: a place that makes you feel like a child again.

86. A Message from an Old Friend: Hearing from someone who has not forgotten you, even after years.

87. The Child-Like Feeling on Christmas Morning: Or whatever day in your year carries that feeling of waking up excited.

88. The Day When Everything Finally Goes Your Way: It does not happen often, which is part of why it is so good when it does.

89. Compliments and Praise: Someone noticing the effort, not just the result.

90. To Look Back on This Moment in 10 Years and Realize You Did It: Being the person who remembers this stretch as something they got through.

91. To Become a Humble Servant of God the Almighty: For people of faith, a life given to something larger than themselves.

92. The First Warm Drink of the Morning: Coffee, tea or just hot water, and the quiet minute before the day starts.

93. A Book That Says the Thing You Could Never Say: The particular relief of finding a sentence that describes exactly what you have been feeling and could not name.

94. Being Bad at Something New: The first week of an instrument, a language, a sport. It is humbling and it is also the only part that is genuinely new.

95. An Animal Being Pleased to See You: A dog at the door, a cat deciding you are acceptable today. Uncomplicated and entirely reliable.

96. The Sea: Standing in front of something that does not care what kind of week you have had is its own kind of relief.

97. Finishing Something You Started: Not a great thing — any thing. The drawer you fixed, the email you sent, the walk you finished. Completion is a feeling you can manufacture.

98. A Stranger's Small Kindness: Someone holding a door, giving up a seat, saying it is fine when it is not quite fine.

99. The First Warm Day After Winter: The one where everybody is outside, slightly overdressed, faintly stunned. Wherever there is a winter, it comes back every year.

100. Remember: You Are a Beautiful Person: Not because of anything on this list, but because there is only one of you.

Embrace the Beauty of Life

None of these reasons has to be true for you today. Some of them may become true later, and some you will replace with your own, which is how a list like this is meant to be used. If you want to keep one, write it down somewhere you will see it.

You are not alone, and reaching out is not a last resort.

Life is beautiful, and you are a beautiful part of it.

If you are struggling or thinking about suicide, please reach out now. findahelpline.com lists free, confidential helplines worldwide. In the US and Canada, call or text 988. In the UK and Ireland, call Samaritans on 116 123. In Australia, call Lifeline on 13 11 14. In India, call Tele-MANAS on 14416. In the Philippines, call the National Center for Mental Health crisis line on 1800-1888-1553 (or 1553 from a Luzon landline). If you are in immediate danger, call your local emergency number.

Sources

The notes under each reason are reflections, not claims. The facts on this page are about the helplines, read on the services' own pages in September 2026:

Find A Helpline — free, confidential helplines, with verified helplines listed in more than 175 countries.
988 Suicide & Crisis Lifeline (US) and 9-8-8: Suicide Crisis Helpline (Canada) — call or text 988.
Samaritans (UK) and Samaritans Ireland — 116 123, free to call.
Lifeline Australia — 13 11 14.
Tele-MANAS on Find A Helpline (India) — 14416, toll-free.
MentalHealthPH, Emergency Help (Philippines) — the National Center for Mental Health crisis line (1553 from Luzon landlines); Find A Helpline, Philippines lists 1800-1888-1553.

Checked September 2026.

The Science Behind Affirmations: Harnessing the Power of Words

Hacks Vitae — Wed, 22 Nov 2023 12:45:05 GMT

Words of Affirmation

Affirmations are short positive statements people repeat to themselves: "I am capable", "I am enough". They are one of the most common self-help practices, and they sit next to a real body of psychology research, which is more specific and more mixed than the usual telling.

This page separates three things that often get blended together: the research on self-affirmation, the research on repeating positive statements about yourself, and beliefs such as the law of attraction, which is not a research finding at all.

Understanding Affirmations

The psychology behind affirmations has a name: self-affirmation theory. Claude Steele popularised it in the late 1980s; his main paper, "The psychology of self-affirmation: Sustaining the integrity of the self", appeared in 1988, and it remains a well-studied theory in social psychology.

The theory says people are driven to protect their self-integrity: their sense of themselves as a good, moral person. When information threatens that picture (a health warning, criticism of their work), they tend to get defensive. If they first reflect on values that matter to them, they are less likely to experience distress and react defensively when confronted with that information.

Note what the theory is about: reflecting on values you already hold, usually by writing about them. That is a different thing from repeating a flattering sentence about yourself, and the evidence for the two is different too.

The Science of Affirmations: A Deeper Insight

Neuroplasticity

Neuroplasticity, the brain's ability to reorganise and rewire its connections, is real. It is often invoked to say that reciting affirmations rewires the brain in a chosen direction. The studies on this page measure mood, defensiveness and behaviour, not that kind of brain change. The one brain-imaging study among them, by Cascio and colleagues (2016), looked at values affirmation: people who were affirmed showed more activity in brain regions linked to self-related processing and reward while reflecting on future-oriented core values.

The Role of the Reticular Activating System (RAS)

The reticular activating system is a set of connected nuclei in the brainstem that regulates wakefulness and the transitions between sleep and waking. That is well established.

The popular self-help version, in which the RAS works as a filter that lets in only information matching your goals, goes well beyond that description.

What is well described is selective attention. Once something is on your mind, you notice it far more often, even if it is not happening more often. The linguist Arnold Zwicky called one form of this the frequency illusion in 2005; it is also known as the Baader-Meinhof phenomenon: the car you are thinking of buying suddenly seems to be everywhere. Holding a goal in mind can make relevant opportunities easier to notice; it does not make them more numerous.

Affirmations and the Law of Attraction

The law of attraction is the New Thought belief that positive or negative thoughts bring positive or negative experiences into a person's life. It is a belief, not a finding: there is no empirical scientific evidence supporting it. Affirmations and the law of attraction are often presented together, which can lend the second the credibility of the first. They are not the same kind of claim. There is more on this further down.

What the research found

Self-affirmation has a measurable effect, and it is small. A 2015 meta-analysis by Epton and colleagues pooled 144 experimental tests of self-affirmation alongside health-risk information. It found "small but reliable positive effects" on accepting the message, on intentions to change and on later behaviour. The authors describe the effects as small in size, and comparable to other health-behaviour interventions.

It does not always hold up outside the lab. Values-affirmation writing exercises have also been tried in schools. In a large replication by Hanselman and colleagues (2017), in a setting where an earlier large field experiment had found benefits, the same procedure showed no evidence of effects, and the estimates were precise enough to rule out benefits larger than an effect size of 0.10.

How it is done may matter. In a 2021 study of 1,056 people by Iles and colleagues, affirmations written as essays worked better than questionnaire-style versions for changing intentions to drink less, and the authors concluded that essays with an explicit focus on values were "most potent".

Repeating a positive statement about yourself is a separate question, and the evidence is disputed. In 2009, Wood, Perunovic and Lee (Psychological Science) had people repeat "I'm a lovable person". Participants with low self-esteem who repeated it felt worse than those who did not. Participants with high self-esteem felt better, "but to a limited degree". The authors concluded that repeating positive self-statements may benefit certain people "but backfire for the very people who 'need' them the most".

In 2020, Flynn and Bordieri (Journal of Contextual Behavioral Science) ran two replications, with about 225 and 237 participants. Neither found the effect: there were no differences between conditions on mood or state self-esteem, and in the first study trait self-esteem did not change the result. The second study compared positive affirmations with writing about values, and found no difference between them either. So the backfire finding is not settled. Neither set of studies shows a reliable lift in mood from repeating a positive statement.

The Secret, the law of attraction, and the law of assumption

These come up often alongside affirmations, so here is where they stand.

Where it comes from. The Secret by Rhonda Byrne was released as a film in March 2006 and as a book later the same year. Byrne's official site reports sales of over 30 million copies. Byrne cites Wallace Wattles' The Science of Getting Rich (1910) as her inspiration, and the idea belongs to the New Thought movement, which coalesced in the United States in the 19th century. Neville Goddard (1905–1972) taught a related method he called the law of assumption.

How the two differ.

	Law of attraction	Law of assumption
Core claim	Positive or negative thoughts bring positive or negative experiences	"Assume the feeling of the wish fulfilled"
How it is said to work	Thoughts send out "vibrations" that something in the universe answers; like attracts like	Mentally and emotionally inhabit the state of already possessing what is desired
Associated with	Rhonda Byrne, The Secret	Neville Goddard

What the evidence says. Neither has scientific support. Reviewing The Secret for Live Science in 2009, the writer Benjamin Radford concluded that its law of attraction "has nothing to do with science". When these methods seem to work, there is a simpler explanation, in our reading: deciding clearly what you want changes what you notice, what you try and what you say yes to.

The criticism that matters. Writing for the Committee for Skeptical Inquiry, Mary Carmichael and Ben Radford said the premise of The Secret contains "an ugly flipside: if you have an accident or disease, it's your fault". That conclusion is built into the idea, and any version worth keeping has to drop it.

What is left is planning: decide what you want, picture it clearly enough to recognise a chance when it comes, and then do the work.

Auto-suggestion: where the idea came from

Auto-suggestion is older than Napoleon Hill: Herbert Parkyn taught it at his Chicago School of Psychology from 1896 and in a 1906 book, and Émile Coué popularised it in the 1920s. Napoleon Hill's Think and Grow Rich (1937) made it the book's third "step toward riches". Hill defined it as "self-suggestion", and he was clear that "plain, unemotional words do not influence the subconscious mind": the words had to be "well emotionalized with BELIEF". Earlier in the book he tells readers to read their written statement of desire aloud twice daily.

Two notes on Hill as a source. Newspapers of his day, quoted by the journalist Matt Novak, report warrants for his arrest in Alabama in 1908, when a postal mail-fraud investigation was also under way, and in Illinois in 1918, for breaching the state's Blue Sky Law on selling stock. In October 1919 the Federal Trade Commission charged him with fraudulent advertising. His story that Andrew Carnegie set him the task of studying successful people appeared only after Carnegie died in 1919. And his central promise, that belief plus repetition produces wealth, is the part the evidence above does not support.

Affirmations worth using

If you want to try affirmations, here are some ways to do it. The first is closest to what the studies tested; the rest are our own suggestions:

Write about a value. The self-affirmation research is mostly about reflecting on values that matter to you, often in a short piece of writing, rather than repeating a sentence about yourself.
Choose something you can nearly believe. "I am handling this" is easier to accept than "I am unstoppable".
Prefer process to identity. "I prepare for difficult conversations" is something you can check and make true; "I am confident" is not.
Pair it with an action. An affirmation with nothing behind it is a wish; attached to a habit, it becomes a plan.
Notice how it lands. If a phrase leaves you feeling worse, that is worth paying attention to, whatever the studies say on average.

Some examples in that spirit:

I am doing what today asks of me.
Mistakes are information. I use them and move on.
I am allowed to be imperfect and still be worth something.
Change is normal. I adapt.
I am making steady progress, even on days I cannot see it.

Persistent low mood or low self-worth is worth raising with a doctor or a qualified counsellor. A list of phrases is not treatment.

Positive thinking, in proportion

What holds up. Optimism, in the research sense, means holding generally favourable expectations for the future. A 2010 review by Carver, Scheier and Segerstrom links it to better well-being in hard times, to engagement coping (dealing with a problem) rather than avoidance, to more persistence in education, and to better relationships. The review also notes that there are instances where optimism brings no advantage, or a disadvantage, though it describes them as relatively rare.

Health. A 2019 meta-analysis by Rozanski and colleagues (JAMA Network Open), covering 15 studies and 229,391 participants, found optimism associated with a lower risk of cardiovascular events and of death from any cause. These are observational studies: they show an association, not that thinking positively prevents or treats illness. The authors themselves call for studies of the mechanisms and of whether interventions help.

The version we find workable: accurate about the situation, confident about responding to it. Notice what is going well, take a setback seriously enough to learn from it, and let a bad day be a bad day.

What you can do with this

Tell the three ideas apart. Self-affirmation (reflecting on your values) has research behind it, with small effects that do not always replicate. Repeating positive statements about yourself has mixed evidence. The law of attraction has none.
Try the version closest to the studies. A few minutes writing about something that matters to you is closer to what the self-affirmation studies tested than repeating a slogan, though one 2020 comparison found no difference between the two.
Treat attention as the useful part. Holding a goal in mind helps you notice chances; acting on them is what changes things.
Be wary of any method that blames you for bad luck or illness. That part is built into the law of attraction, and it is not supported by anything.

Sources

Wikipedia, "Self-affirmation" (the theory; Steele, 1988).
Epton, T., Harris, P. R., Kane, R., van Koningsbruggen, G. M. and Sheeran, P. (2015), "The impact of self-affirmation on health-behavior change: a meta-analysis", Health Psychology 34(3), 187–196 (PubMed 25133846).
Hanselman, P., Rozek, C. S., Grigg, J. and Borman, G. D. (2017), "New Evidence on Self-Affirmation Effects and Theorized Sources of Heterogeneity from Large-Scale Replications", Journal of Educational Psychology (PubMed 28450753).
Iles, I. A., Gillman, A. S., Ferrer, R. A. and Klein, W. M. P. (2021), "Self-affirmation inductions to reduce defensive processing of threatening health risk information", Psychology & Health.
Cascio, C. N. and colleagues (2016), "Self-affirmation activates brain systems associated with self-related processing and reward and is reinforced by future orientation", Social Cognitive and Affective Neuroscience 11(4), 621–629 (PubMed 26541373).
Wood, J. V., Perunovic, W. Q. E. and Lee, J. W. (2009), "Positive self-statements: power for some, peril for others", Psychological Science 20(7), 860–866 (PubMed 19493324).
Flynn, M. K. and Bordieri, M. J. (2020), "On the failure to replicate past findings regarding positive affirmations and self-esteem", Journal of Contextual Behavioral Science 16, 49–61.
Carver, C. S., Scheier, M. F. and Segerstrom, S. C. (2010), "Optimism", Clinical Psychology Review 30(7), 879–889 (PubMed 20170998).
Rozanski, A., Bavishi, C., Kubzansky, L. D. and Cohen, R. (2019), "Association of Optimism With Cardiovascular Events and All-Cause Mortality", JAMA Network Open 2(9), e1912200 (doi).
Wikipedia, "Autosuggestion" (Herbert Parkyn, citing his own Auto-Suggestion: What It Is and How to Use It, 1906) and "Émile Coué" (citing The Oxford Companion to the Mind, 1987).
Wikipedia, "Neuroplasticity", "Reticular formation", "Frequency illusion", "Law of attraction (New Thought)", "The Secret (Byrne book)", "The Secret (2006 film)", "New Thought", "Neville Goddard", "Napoleon Hill" and "Think and Grow Rich".
Napoleon Hill, Think and Grow Rich (1937), contents and chapter 4, "Auto-Suggestion", at sacred-texts.com.
Matt Novak, "The Untold Story of Napoleon Hill, the Greatest Self-Help Scammer of All Time", Gizmodo (6 December 2016), quoting the Pensacola Journal (1908) and the Chicago Daily Tribune (1918).
Benjamin Radford, "The Pseudoscience of 'The Secret'", Live Science (3 February 2009); Mary Carmichael and Benjamin Radford, "Secrets and Lies", Skeptical Inquirer (29 March 2007).
Arnold Zwicky, "Just between Dr. Language and I", Language Log (7 August 2005) (the frequency illusion).

Checked September 2026. The research findings rest on the papers' own abstracts (PubMed, and the publisher's page for Flynn and Bordieri); the history of the law of attraction, The Secret and Goddard follows the Wikipedia articles listed, used only where they give a citation for the sentence; the record on Hill follows Novak's reporting; Hill's words are from his own chapter on auto-suggestion. Where studies disagree, as on repeating positive self-statements, the page gives both.

The Great Dictator Speech Analysis: Charlie Chaplin's Powerful Speech Against Dictatorship & Tyranny

Hacks Vitae — Wed, 22 Nov 2023 12:26:24 GMT

The Great Dictator Speech Analysis

The speech is the last scene of Charlie Chaplin's The Great Dictator, which opened in New York on 15 October 1940. It is spoken by a Jewish barber who has been mistaken for the dictator he resembles, and it is not a satire: it is a plain plea for kindness, freedom and unity, addressed to the soldiers in front of him and to everyone listening. The film historian Jeffrey Vance, in an essay for the US Library of Congress, calls it "the five-minute speech that concludes the film".

Below: what happens in the film, why Chaplin made it when he did, what the speech says and how it is built, and how it has been received since.

Contextual Setting

Two roles, one face

Chaplin plays two roles: a little Jewish barber living in the ghetto, and Adenoid Hynkel, the dictator of Tomainia, a parody of Adolf Hitler. The joke the film runs on is that the two men look alike.

How the barber ends up on the platform

The American Film Institute's catalogue sums up the plot. At the end of the First World War the barber saves the life of an officer, Schultz; a plane crash leaves the barber with amnesia, and he spends years in hospital while Hynkel rises to power. Back in his shop in the ghetto, he resists the persecution around him and is arrested with Schultz, who has also spoken out; both are sent to a prison camp. They escape on the eve of Hynkel's invasion of the neighbouring country, Osterlich, and Hynkel himself is mistaken for the escaped barber and arrested. The barber, taken for the dictator, takes his place on the parade platform and gives the speech.

Why Chaplin made it, and what it cost him

The Great Dictator was Chaplin's first film with dialogue. The Library of Congress essay puts it this way: "Over a decade after the rest of the film industry had accepted talking pictures, the greatest star of the silent-film era began his first full-dialogue film."

He made it while the United States was still out of the war. Britain, Chaplin's native country, had declared war at the beginning of September 1939, but the United States, where he lived, had resolved to keep out of the conflict. The film opened in New York on 15 October 1940, at two Broadway theatres at once. The United States entered the Second World War only after the attack on Pearl Harbor on 7 December 1941.
There was pressure against it from the start. According to the Production Code files cited by the AFI catalogue, in October 1938 the German consul, George Gyssling, wrote to Joseph I. Breen objecting to Chaplin's intention to make a film that would "burlesque" Hitler. Vance writes that Chaplin went ahead "despite death threats once his project was announced".
Chaplin and Hitler were born four days apart. Chaplin was born on 16 April 1889, and Hitler within four days of him. Chaplin's official site notes that when he was writing the film in 1939, he was as famous worldwide as Hitler, and his tramp character wore the same moustache.
Chaplin later said he could not have made it. In his 1964 autobiography, as Vance quotes it, he wrote: "Had I known the actual horrors of the German concentration camps, I could not have made 'The Great Dictator'; I could not have made fun of the homicidal insanity of the Nazis." Satire was possible in 1940 in a way it was not once the camps were known.
It was banned widely and still made money. According to Vance, the film was banned throughout occupied Europe, in parts of South America and in the Irish Free State, and was first seen in Germany and Italy only when it was reissued in 1958. It cost $1,403,526 and became "Chaplin's most profitable film up to that time", earning $5 million worldwide in its original release.

Rhetorical Analysis

The quotations below follow the transcript of the speech published on Chaplin's official website.

How the speech is built

The speech moves in four steps.

A refusal. It opens by turning down power: "I'm sorry, but I don't want to be an emperor. That's not my business. I don't want to rule or conquer anyone."
A diagnosis. The world could provide for everyone, the barber says, "but we have lost the way. Greed has poisoned men's souls, has barricaded the world with hate, has goose-stepped us into misery and bloodshed." The failure he describes is moral, not technical: "Machinery that gives abundance has left us in want. … We think too much and feel too little. More than machinery we need humanity."
A turn to hope. The same inventions point the other way: "The aeroplane and the radio have brought us closer together. The very nature of these inventions cries out for the goodness in men - cries out for universal brotherhood - for the unity of us all."
An appeal to the soldiers. He speaks straight to the troops: "Soldiers! don't give yourselves to brutes", "You are not machines! You are not cattle! You are men!", and closes, "Soldiers! in the name of democracy, let us all unite!"

Universal Themes

The speech names its themes plainly: kindness, freedom, the end of "national barriers", and "a world of reason, a world where science and progress will lead to all men's happiness". It cites one text: "In the 17th Chapter of St Luke it is written: 'the Kingdom of God is within man'". The King James Version of Luke 17:21 reads "the kingdom of God is within you"; the speech widens it to "not one man nor a group of men, but in all men!"

Satire and Irony

The irony is in who is speaking. The man the regime persecutes stands where the tyrant should be and uses the tyrant's platform to ask for the opposite of everything the regime stands for. The speech does not mock the regime's ideas from inside; it rejects them outright.

Whether the speaker is still the barber is a matter of reading. Chaplin's official site describes the ending as a speech "made by the barber, speaking Chaplin's own words". Vance goes further: "The final speech, however, is not given by the Barber character but by Chaplin himself," who "drops his comic mask and speaks directly to the world". Both readings describe the same scene; the second explains why the tone changes so completely.

Enduring Impact

A divided reception

The speech has divided opinion. The AFI catalogue records that contemporary reviews "objected to the final speech as too preachy". Chaplin's official site says many people thought it "superfluous to the film" while "others found it uplifting". The New York Times critic Bosley Crowther, as Vance quotes him, found the film too long in places but still called it "a truly superb accomplishment by a truly great artist".

A growing reputation

Vance writes that critical opinion of the film, "particularly of the final speech, has risen greatly" over the decades since. In 1989, the centenary of Chaplin's birth, it opened the Moscow International Film Festival, and in 2002 it closed the Berlin Film Festival. He also records one story about its most important viewer: according to an agent who fled Germany after working in the film division of the Nazi Ministry of Culture, Hitler screened the film alone, two evenings running. Chaplin's reply, as Vance quotes it: "I'd give anything to know what he thought of it."

An Appeal for Peace

The speech has lasted because it names things that do not date: greed, fear, cleverness without kindness, and ordinary people's power to refuse. It is a speech against dictatorship given by a character who, a moment earlier, was one of its victims.

What to take from it

A few habits for reading any famous speech, which this one rewards:

Read the whole text, not the clip. The most shared lines of this speech sit inside a four-step argument; the ending to the soldiers changes what the opening means.
Ask who is speaking, and to whom. Here it is a persecuted man in a dictator's place, speaking to an army. The same words from a ruler would mean something else.
Check the date against what the speaker could know. Chaplin wrote this before the United States was at war and, by his own later account, before he understood the camps. That is part of reading it fairly.
Separate the story from the reception. The speech was called preachy in 1940 and a masterpiece later. Both are facts about audiences, not about the text.

Sources

Charlie Chaplin official website (Roy Export S.A.S.), "The Final Speech from The Great Dictator" (transcript and notes): https://www.charliechaplin.com/en/articles/29-the-final-speech-from-the-great-dictator-
Charlie Chaplin official website, "Overview of His Life": https://www.charliechaplin.com/en/articles/21-overview-of-his-life
Christian Delage, "Filming The Great Dictator", from Chaplin Facing History (2005), on Chaplin's official website: https://www.charliechaplin.com/en/biography/articles/13-The-Great-Dictator+The Great Dictator/
Jeffrey Vance, "The Great Dictator", National Film Registry essay, US Library of Congress (archived copy): https://web.archive.org/web/20190902100006/https://www.loc.gov/static/programs/national-film-preservation-board/documents/great_dictator.pdf
American Film Institute Catalog, "The Great Dictator" (synopsis and notes): https://catalog.afi.com/Catalog/moviedetails/27635
US National Archives, "Joint Address to Congress Leading to a Declaration of War Against Japan (1941)": https://www.archives.gov/milestone-documents/joint-address-to-congress-declaration-of-war-against-japan
The Bible, King James Version, Luke 17:21, Project Gutenberg: https://www.gutenberg.org/ebooks/10

Checked September 2026.

Life in Pink Meaning: Unveiling the Essence of a Colorful Existence

Hacks Vitae — Fri, 17 Nov 2023 01:10:38 GMT

Unveiling the Essence of a Colorful Existence

You may be looking up "life in pink" because you have heard Édith Piaf's song "La Vie en rose", or because you have seen the phrase used to describe a way of living. Both answers are here, starting with the words themselves.

The short answer

"Life in pink" is a word-for-word translation of la vie en rose. In French, rose is two words: a feminine noun for the flower, and a masculine noun and adjective for the colour pink. In la vie en rose it is the colour.

The French idiom voir la vie en rose means seeing things through rose-tinted glasses: looking at them in their best light. So the phrase is less about the colour than about a way of looking.

The song

Item	Detail
Words	Édith Piaf, written for and with her friend Marianne Michel. It nearly had a different title: "Quand je vois les choses en rose"
Music	Credited to Louiguy, Piaf's accompanist. Who wrote the melody is disputed (see below)
First record	1945, according to France's national audiovisual archive (INA), with Piaf adding the song to her own repertoire the following year. Accounts of the first release differ: discographies list Piaf's Columbia single in 1947. The dates in this table are INA's
Piaf's recording	9 October 1946
Recognition	Piaf's single, in its 1950 Columbia release, was inducted into the Grammy Hall of Fame in 1998

Who wrote the music is not settled. The record credits Louiguy, and he gave his own account of composing it during the German occupation. Piaf said he had only lent his name. INA adds that her orchestra leader, Robert Chauvigny, may also have worked on the melody. The credit stayed with Louiguy, and the national library of France catalogues the song with music by Louiguy and words by Piaf.

INA counts more than 350 artists who have performed it since, among them Dalida, Yves Montand, Mireille Mathieu and Grace Jones.

By Louiguy's own account, the melody was composed during the German occupation of France. Read that way, "life in pink" is not a claim that everything is fine. It is a choice about where to put your attention, in years when a great deal was not fine. That is our reading, not something the song's authors said.

Does pink calm people down?

A common claim is that a pink room calms you down or reduces aggression. The version tested in the studies below is "Baker-Miller pink". By 2014, many prisons across Western countries had begun painting detention cells that colour, based on findings from more than 30 years earlier that it reduces physical strength and aggressive behaviour.

The later evidence is weaker than the story.

A 1987 study of 46 students found lower anxiety after five minutes in a pink room than in a red one, but no difference in grip strength or motor precision. The authors called this "minimal support for the hypothesized calming effects of Baker-Miller Pink."
A controlled experiment, published online in 2014, tested the prison claim directly. In the authors' words, "The results do not replicate the original findings".
A 2015 review of colour psychology as a whole found that much of the research "remains underpowered" and that the field "is at a nascent stage of development", and recommended "patience and prudence" about conclusions and real-world use.

So "pink is calming" is not an established finding. Paint a room pink if you like the colour; the evidence does not say the colour will do the calming.

What you can do with this

The useful part of la vie en rose is the idiom, not the paint.

Choose what you look at. Seeing life "in pink" means picking the most hopeful reading of a situation that the facts allow. It works as a habit of attention, not as a denial of what is wrong.
Keep the context in view. In our reading, a hopeful outlook that knows what it is hopeful in spite of is sturdier than one that ignores it.
Be sceptical of easy fixes. If a colour, a slogan or a product promises to change your mood by itself, the Baker-Miller story is a good reason to ask for the evidence first.

Sources

The idiom and the two senses of rose — Larousse French–English dictionary, entry "rose".
Lyricist, working title, first record, Piaf's recording date, the composer dispute and the cover count — INA, "Les secrets de 'La vie en rose'", 10 October 2016.
Credits — Bibliothèque nationale de France, catalogue record for the work "La vie en rose".
Grammy Hall of Fame — Recording Academy, Hall of Fame list.
1987 pink-room study — Profusek and Rainey, "Effects of Baker-Miller pink and red on state anxiety, grip strength, and motor precision", Perceptual and Motor Skills 65 (1987) (PubMed).
The prison experiment — Genschow, Noll, Wänke and Gersbach, "Does Baker-Miller pink reduce aggression in prison detention cells? A critical empirical examination", Psychology, Crime & Law 21 (2015) (doi:10.1080/1068316X.2014.989172).
Colour psychology review — Andrew J. Elliot, "Color and psychological functioning: a review of theoretical and empirical work", Frontiers in Psychology, 2015.

Checked September 2026. The song's credits are disputed, and this page gives the positions rather than picking one.

Napoleon Bonaparte History: Must Know Facts About the French Emperor

Hacks Vitae — Mon, 13 Nov 2023 02:51:56 GMT

Napoleon at a glance

Napoleon Bonaparte was born on 15 August 1769 in Ajaccio, Corsica, and died on 5 May 1821 on Saint Helena, aged 51. He was Emperor of the French from 18 May 1804 until his first abdication in 1814, and briefly again during the Hundred Days in 1815.

He is remembered for two things that pull in different directions. Many historians rank him among the great military commanders. His legal and administrative reforms, above all the civil code of 1804, later called the Napoleonic Code, outlasted his empire. The same code entrenched equality before the law and also reduced the rights of women and children, so what you make of him depends a good deal on which part you look at.

This page covers the facts people most often search for: his origins, how he rose, the coronation, Russia, the two exiles, his height, his death and tomb, a quotation often credited to him, and whether his army fired on the pyramids.

Was he French?

Yes, though Corsica had only just become French. The Republic of Genoa ceded the island to France in the 1768 Treaty of Versailles, the year before he was born. It became a province of France in 1770.

His family was of Italian origin: the Buonapartes descended from a minor Tuscan noble family that had moved to Corsica in the 16th century. His mother tongues were Corsican and Italian. In January 1779, aged 9, he moved to the French mainland and went to a school in Autun to improve his French. He became fluent but kept a Corsican accent, and at the military academy at Brienne-le-Château other boys bullied him for his accent, his birthplace, his small stature and his poor French.

From Corsican Origins to the French Throne

He was commissioned as an officer in the French royal army in 1785 and supported the Revolution from 1789. He rose quickly after the siege of Toulon in 1793, and in 1796 he led a campaign against the Austrians and their Italian allies that made him a national hero. In 1798 he led the invasion of Egypt and Syria.

In November 1799 he organised the coup of 18 Brumaire against the government of the Directory and became First Consul. His civil code came into force in March 1804, and in May 1804 he became Emperor of the French.

The self-coronation

On 2 December 1804, at Notre-Dame in Paris, with Pope Pius VII taking part, Napoleon was anointed by the Pope and then crowned himself. He then crowned Joséphine.

The popular story is that he snatched the crown from the Pope's hands. The Fondation Napoléon describes it differently: he picked up the crown from the altar and put it on his own head, "following the agreed order of service." The Fondation Napoléon reads it as a symbolic gesture that "emphasized the political but also spiritual power that Napoleon aimed to wield over Pope Pius VII."

What went wrong in Russia, 1812

Napoleon invaded Russia on 24 June 1812. The best-known record of what happened to his army is the chart drawn by the retired engineer Charles Minard, dated 20 November 1869.

By Minard's figures:


Crossed the Neman into Russia	422,000 men
Reached Moscow	100,000
Crossed the Neman back out	10,000

So most of the loss came before the retreat from Moscow began. The main body of the army shrank by a third in the first eight weeks, before any major battle, partly because of diseases such as diphtheria, dysentery and typhus.

Which diseases did the most damage is still being studied. In October 2025 a team from the Institut Pasteur published DNA evidence from 13 soldiers of the retreat, buried in Vilnius. They found the bacteria behind paratyphoid fever and relapsing fever, adding to an earlier study that had found the agents of typhus and trench fever. The institute says their precise role in the deaths is not known, and 13 soldiers is a small sample.

The retreat added its own losses: the army could not be fed, it was still in summer clothing when winter arrived, and Cossacks harassed it constantly. But on Minard's numbers the army was already a fraction of its size by the time it reached Moscow.

The two exiles: Elba and Saint Helena

In 1813 Prussia and Austria joined Russia, and Napoleon was decisively defeated at the Battle of Leipzig. The coalition took Paris and forced him to abdicate in April 1814. He was sent to the Mediterranean island of Elba, and the Bourbons returned to the throne under Louis XVIII.

Ten months later he escaped from Elba with around a thousand men and marched on Paris. Louis XVIII fled, and Napoleon ruled again for the period known as the Hundred Days. His return ended in defeat at Waterloo on 18 June 1815.

This time the British held him on Saint Helena, in the South Atlantic, 1,870 km from the west coast of Africa. He and 27 followers arrived in October 1815, and he lived there until his death five and a half years later.

What if he had won at Waterloo?

It is a popular question, and the numbers around it are worth knowing. On 13 March 1815 the powers at the Congress of Vienna declared Napoleon an outlaw. Four days later Great Britain, Russia, Austria and Prussia each pledged to put 150,000 men into the field to end his rule.

At Waterloo he was defeated by two of the coalition's armies: a British-led force under the Duke of Wellington, and a Prussian one under Blücher. A victory there would still have left him facing the rest of a coalition that had declared him an outlaw. How long he could have held out is speculation, and you can weigh it yourself.

The Height Myth and Where It Came From

Napoleon was not unusually short. Three French sources, his valet Constant, General Gourgaud and his doctor Francesco Antommarchi, gave his height as just over 5 pieds 2 pouces. Those are French units: the French inch, the pouce, was about 2.7 cm, longer than the English inch of 2.54 cm.

Read as French measures, that is about 1.67 m, just under 5 ft 6 in. Estimates vary a little (some historians put him at 1.70 m), and sources describe that as around average, or a little above, for a Frenchman of his time. Read as English feet and inches, the same figure becomes 5 ft 2 in, which is where the "short" idea gets its number.

The image did the rest. British cartoonists repeatedly drew him as short, above all James Gillray, whose caricatures of him were enormously popular and influential.

How he died

In March 1821 he was confined to bed, and he died on 5 May. Antommarchi and the British doctors wrote separate autopsy reports, each concluding that he had died of internal bleeding caused by stomach cancer, the disease that had killed his father. The Fondation Napoléon notes that the autopsy findings were signed off by no fewer than seven doctors.

A later theory held that he had been poisoned with arsenic, because high levels were found in samples of his hair. Later studies found similar levels in hair from his childhood, and from his son and Joséphine; arsenic was widely used in 19th-century medicines and products such as hair creams. A 2021 study by gastrointestinal pathologists again concluded that he died of stomach cancer.

Where he is buried

He was buried on Saint Helena. In 1840 the British government gave King Louis Philippe I permission to bring his remains back to France (the retour des cendres), and a state funeral was held at Les Invalides in Paris on 15 December 1840.

His tomb under the dome took two decades to finish: on 2 April 1861, in the presence of Napoleon III, his coffin was moved into the sarcophagus. The sarcophagus is often described as red porphyry, but it is made of quartzite from Shoksha, in Russian Karelia, on the shores of Lake Onega. It rests on a base of green granite from the Vosges.

His brothers Jérôme and Joseph have tombs in the same church, completed in 1862 and 1864, and in December 1940 the coffin of his son, Napoleon II, was brought from Vienna and placed beside him.

The quote often credited to him

"History is a set of lies agreed upon."

Napoleon used a version of this, but the idea is older than him. On Saint Helena, as recorded by Emmanuel de Las Cases in the Mémorial de Sainte-Hélène, he asked what historical truth was, most of the time, and answered: "une fable convenue", a fable agreed upon. The next words are "ainsi qu'on l'a dit fort ingénieusement", as has been very cleverly said. Quote Investigator reads them as Napoleon disclaiming credit; the 1823 English translation attaches them to the sentence that follows instead. Either way, others had said it first.

The trail goes back to Bernard le Bovier de Fontenelle. In his 1724 essay L'Origine des Fables he wrote, about ancient history, that there are no ancient histories other than these fables. The philosopher Helvétius, in 1758, credited Fontenelle with the saying that history is only une fable convenue, and Voltaire also credited it to him. The word "lies" appears by 1881, in a speech by the American orator Wendell Phillips.

So the fair description is: a saying Napoleon helped make famous, not one he coined. Note too that the French word is fable, not "lie".

Did Napoleon fire at the pyramids?

Ridley Scott's 2023 film Napoleon shows his army firing on the pyramids. The historian Joan Tumblety, writing about the film, treats it as an invention: "you don't need to invent ice traps or Napoleon ordering his men to fire at the pyramids, as Ridley Scott's biopic does, to convey his callous disregard for life."

The Battle of the Pyramids was not fought at the pyramids. It took place on 21 July 1798 near the village of Embabeh, about 24 km (15 miles) from the pyramids.

The campaign was heavily documented. Napoleon took a group of 167 scientists with him to Egypt, and their work was published as the Description de l'Égypte, from 1809.

The Sphinx's nose

The same story is told about the Sphinx: that French gunners shot its nose off. Frederic Louis Norden drew the Sphinx in 1737, and his drawings show the nose already missing; they were published in Voyage d'Égypte et de Nubie in 1755, more than forty years before the French army arrived.

Writing in the early 15th century, the historian al-Maqrizi attributed the damage to a Sufi named Muhammad Sa'im al-Dahr in 1378, who objected to peasants making offerings to the statue. Some Arab writers of the 10th century already blamed the damaged nose on iconoclasts, and examination of the face shows evidence of deliberate work with rods or chisels.

What you can do with this

Check the units before the conclusion. Napoleon's "5 ft 2" is a French measurement read as an English one. Old figures in feet, miles or currencies often need converting before they mean anything.
Look at the timing of losses, not just the total. Minard's chart shows most of the army gone before the famous winter retreat. The part of a story that is easiest to picture is not always the part that did the damage.
When a quote sounds too perfect for its speaker, look for an earlier version. The "fable agreed upon" was in print more than half a century before Napoleon used it.
A dramatic film scene is a prompt to look something up, not a source. The Sphinx story falls apart on dates, and a historian writing about the film calls the pyramid scene an invention.

Sources

Wikipedia, "Napoleon" (birth, early life, career, Egypt, the coronation, the exiles, Waterloo, death, the 1840 return, reforms), and the works it cites for them: Frank McLynn, Napoleon: A Biography (1997); Adam Zamoyski, Napoleon: The Man Behind the Myth (2018); Andrew Roberts, Napoleon: A Life (2014); Philip Dwyer, Napoleon: The Path to Power (2008), Citizen Emperor (2013) and Napoleon: Passion, Death and Resurrection, 1815–1840 (2018); Steven Englund, Napoleon: A Political Life (2010); Susan P. Conner, The Age of Napoleon (2004); Michael Broers, Napoleon: The Decline and Fall of an Empire (2022); Harold T. Parker, "The Formation of Napoleon's Personality", French Historical Studies (1971); Christon Archer and others, World History of Warfare (2008); Alessandro Lugli and others, "The gastric disease of Napoleon Bonaparte", Virchows Archiv (2021); J. Thomas Hindmarsh and John Savory, "The Death of Napoleon, Cancer or Arsenic?", Clinical Chemistry (2008).
Wikipedia, "Napoleon complex" (citing Owen Connelly, Blundering to Glory, 2006, and David A. Bell, Napoleon: A Concise Biography, 2015); History.com, "Was Napoleon Short? Origins of the 'Napoleon Complex'" (height and French units).
Fondation Napoléon, "The sacre or consecration of Napoleon I" (the coronation).
Wikipedia, "French invasion of Russia" (citing Edward Tufte, The Visual Display of Quantitative Information, 2001, for Minard's figures; Carl von Clausewitz, Der Feldzug 1812 in Russland, and Richard K. Riehn, 1812: Napoleon's Russian Campaign, 1990; the memoirs of General de Caulaincourt; and two US Army studies of disease and winter in the campaign), and Charles Minard's chart of 1869 itself.
Institut Pasteur, "Study suggests two unsuspected pathogens struck Napoleon's army during the retreat from Russia in 1812" (24 October 2025).
Fondation Napoléon, "Another Napoleonic Mystery: Napoleon's death".
Fondation Napoléon, "Les Invalides : le tombeau de Napoléon" (Karine Huguenot and Marie de Bruchard) (the 1840 funeral, the 1861 transfer and the family tombs); Wikipedia, "Napoleon's tomb" (citing François Lagrange, "Les Invalides", L'Objet d'art, 2006, for the granite base); Institut français de Finlande, "Le sarcophage de Napoléon est en Finlande !".
Quote Investigator, "What Is History But a Fable Agreed Upon?" (5 July 2016).
Joan Tumblety, "Did Napoleon really fire at the pyramids?", The Conversation (17 November 2023).
Wikipedia, "Great Sphinx of Giza" (citing the Brooklyn Museum's copy of Norden's plate of the Sphinx; Christiane Zivie-Coche, Sphinx: History of a Monument; Mark Lehner, The Complete Pyramids, 1997; and, for al-Maqrizi, Arabic Humanities, Islamic Thought, Brill, 2017), and "Frederic Louis Norden" (citing Den Store Danske for the 1755 edition).

Checked September 2026. Wikipedia is used here only where it gives a citation for the sentence, and the work it cites is named above; the Fondation Napoléon is used for the coronation and the tomb. Where sources differ (his exact height, which diseases killed most soldiers in 1812), the page gives the range.

Dark Psychology Unveiled: Understanding the Unseen Manipulation

Hacks Vitae — Sat, 11 Nov 2023 02:05:37 GMT

DARK PSYCHOLOGY UNVEILED

"Dark psychology" is a popular label for the study of manipulation, influence and control. It is not a field of academic psychology. A search of Europe PMC, a database of biomedical and life-science research, on 29 September 2026 found no paper with "dark psychology" in its title or abstract, against more than 700 for "Dark Triad". Open Library, a public book catalogue, lists hundreds of books with "dark psychology" in the title.

What researchers do study is close by, and more useful. The Dark Triad is Machiavellianism, narcissism and psychopathy, named by Paulhus and Williams in 2002; with everyday sadism added, it becomes the Dark Tetrad. These are personality traits measured in ordinary people, not diagnoses. The tactics people describe under the label are real too, and most of them are well documented. This page covers both, and how to spot them.

What exactly is dark psychology?

As the phrase is used in popular books, "dark psychology" means the techniques people use to manipulate and control others for their own gain, and the traits of the people who use them. The research that covers the same ground goes under other names: the Dark Triad and Dark Tetrad for the traits; persuasion, heuristics and biases for the levers; coercive control and gaslighting for manipulation inside relationships.

What are the 4 types of dark psychology?

The four are the Dark Tetrad traits. A 2022 study by Pineda and colleagues sums them up:

Machiavellianism: "the manipulation and lack of morality".
Subclinical narcissism: "a grandiose sense of identity with the necessity of admiration".
Subclinical psychopathy: "a pattern of callousness and impulsivity".
Everyday sadism: "a person who derives pleasure or joy from others' suffering".

"Subclinical" matters. A 2023 review by Duradoni and colleagues describes the Triad as "remarkably abusive but not pathological personality types", which are "highly correlated with one another" yet conceptually distinct. Researchers usually measure them with short questionnaires, such as the 27-item Short Dark Triad, so a score describes how much of a trait someone shows rather than sorting people into types.

The everyday tactics, and how to counter them

Most manipulation is not exotic. It is a handful of well-documented levers, used on everyone, often in marketing rather than by a villain. Robert Cialdini's research lists seven "shortcuts" that guide decisions: reciprocity, scarcity, authority, consistency, liking, social proof and unity. Anchoring comes from Amos Tversky and Daniel Kahneman's 1974 paper on judgement under uncertainty; loss aversion is, in the words of one large review, "one of the most widely used concepts in behavioral economics".

Tactic	How it works	What helps
Scarcity	"Only 3 left", "offer ends tonight": fear of missing out beats judgement	Ask whether you wanted it before you saw the timer
Social proof	"50,000 people have joined": we copy the crowd when unsure	Check who the crowd is, and who is counting them
Reciprocity	A free gift, sample or favour creates a sense of debt	A gift you did not ask for is not a debt
Authority	Titles, uniforms, "experts say"	Ask: expert in what, and paid by whom?
Anchoring	The first number sets your sense of a fair price	Decide your number before you hear theirs
Loss aversion	Losses may weigh more than equal gains; how much is debated (see below)	Restate the choice as a gain and see if it still appeals

On loss aversion, the estimates cover a range. A 2024 meta-analysis in the Journal of Economic Literature pooled 607 estimates from 150 articles and found a mean loss aversion coefficient of 1.955, meaning losses weighed roughly twice as much as equal gains. A 2018 review by David Gal and Derek D. Rucker argued the opposite: that "current evidence does not support that losses, on balance, tend to be any more impactful than gains". A second 2024 meta-analysis, by Lukasz Walasek, Timothy L. Mullett and Neil Stewart in the Journal of Economic Psychology, looked only at studies that fitted prospect theory to people's choices between risky gambles, found that "much of the data are of poor quality", and put the figure at "a small λ of 1.31" (95% CI 1.10 to 1.53). It is also much smaller: its authors used only published papers, and only studies whose raw, person-by-person choices they could refit, while the larger analysis also pooled unpublished estimates. Both put the average above 1, so we lean towards losses weighing somewhat more than equal gains on average; how much more is not settled, with the two estimates at about 1.3 and about 2. Gal and Rucker argue for a more contextual view of when losses matter more than gains, rather than treating it as a general rule.

What are some dark psychological tricks?

Inside relationships, the same levers take more personal forms. A 2026 systematic review by J. Adair of how gaslighting is defined in research on gender-based violence describes tactics that include:

Gaslighting: the American Psychological Association defines it broadly as manipulating someone "into doubting their own perceptions, experiences, or understanding of events". The word comes from Patrick Hamilton's 1938 play Gas Light, filmed in 1944, about a husband who changes his wife's surroundings to make her question her sanity.
Love bombing: "the use of excessive affection and charm by perpetrators to confuse and control the survivor".
Isolation: cutting someone off from their support systems, "making them more dependent on the abuser".
Belittling: attacking someone's confidence, or calling them "crazy".

What is the dark psychology of love?

In relationships, the pattern that matters is coercive control: behaviour that is controlling or coercive, repeated or continuous, rather than any single incident. The law in England and Wales defines the offence in those terms (see below). One author quoted in the gaslighting review above suggests that relationships in which gaslighting occurs begin with a period of "lovebombing".

How can you tell if someone is using dark psychology?

The review above also groups the effects on the person targeted, including "disruptions to perception and memory, emotional distress, social isolation". Signs worth noticing in yourself:

You doubt your own memory of events more than you used to.
You feel controlled or confused, without being able to say why.
You see less of friends and family than you did, because of one person.

The pattern that matters is not any single incident but the direction: your world getting smaller and your confidence in your own judgement getting weaker. If that is happening, it helps to talk to someone outside the situation.

Is dark psychology good or bad?

The label covers both self-defence and harm. Knowing how the levers work helps people resist them; using them on others deliberately, especially a partner, is manipulation.

Is dark psychology legal?

Persuasion is legal. Some manipulation inside a relationship is not, depending on where you are. In England and Wales, for example, section 76 of the Serious Crime Act 2015 makes it an offence to "repeatedly or continuously" engage in controlling or coercive behaviour towards a partner, former partner or family member when it has "a serious effect" on them. Other countries have their own laws; this is a description, not legal advice.

What is the darkest personality type?

There is no agreed "darkest type". The Dark Tetrad traits are measured on scales, they overlap, and a person can score high on several at once.

How to learn dark psychology?

To learn what the research says rather than what the label sells, look for the research terms: the Dark Triad and Dark Tetrad, persuasion (Cialdini), heuristics and biases (Tversky and Kahneman), coercive control and gaslighting. Those lead to studies you can check.

What are the benefits of studying dark psychology?

Recognition. Most of these levers work because they go unnoticed; naming them in the moment ("this is scarcity", "this is an anchor") gives you time to decide.

Why is it called dark psychology?

Researchers call the Triad traits "dark" because of what the Duradoni review describes as "their malevolent orientation and a so-called 'callous' and manipulative social tendency". Popular books use the same word as the name of a subject. Not every researcher likes the word: in 2025 the psychologists David Chester, Donald Lynam and Joshua Miller argued in the Journal of Psychopathology and Clinical Science that the field should stop using "dark", which a news release from Chester's university, Virginia Commonwealth University, summarises as "stigmatizing, sensationalistic and imprecise".

Why am I afraid of dark psychology?

The idea that people can be steered without noticing is unsettling. The reassuring part is that the common levers are few, well described, and easier to resist once named.

How to control others' minds?

You cannot, and trying to means overriding someone's autonomy. Honest persuasion, where the other person can see your reasons and say no, is the alternative that respects them.

How to use dark psychology to make someone obsessed with you?

That is manipulation, and it is the pattern the gaslighting research describes: excessive affection to confuse and control. A relationship worth having rests on honesty, trust and consent.

How to make someone miss you?

Through real connection: time together, attention and reliability. Manufactured absence and mixed signals are manipulation, and they work against the thing you want.

Is dark psychology manipulation?

Yes. As the phrase is used, it describes manipulation: influence that works by hiding its methods or exploiting someone's weak points rather than giving them reasons.

Am I in love or obsessed?

Love tends to rest on affection, trust and respect for the other person's freedom. Obsession tends to involve control, possessiveness and difficulty accepting "no". Looking at the balance in the relationship, and at whether the other person's life is getting bigger or smaller, can help.

How do I know if I'm obsessed with someone?

Signs include thinking about one person to the point of neglecting the rest of your life, strong jealousy or possessiveness, or wanting to control what they do. Talking it through with someone you trust can help put the feelings in perspective.

How do you tell if someone is too obsessed with you?

Warning signs include constant monitoring, possessive behaviour, pressure to cut off friends and family, and an inability to respect your boundaries. Pressure to cut off friends and family is the isolation tactic described above, and all of these are worth taking seriously.

How do you handle a person who is obsessed with you?

Clear, consistent boundaries help, and so does telling people you trust what is happening. If you feel unsafe, local domestic-abuse services and the police exist for exactly this.

What happens if you get too obsessed with someone?

It can crowd out the rest of your life: friends, work and your own sense of balance. Noticing it early, and talking to someone about it, makes it easier to step back.

What to take from it

Treat "dark psychology" as a label, not a science. Look for the research names behind it.
Learn the levers. Scarcity, social proof, reciprocity, authority and anchoring cover most of what is used on you every day.
Watch the direction, not the incident. In a relationship, a world getting smaller and self-doubt growing are the signs that matter.
Name it in the moment. Saying "this is an anchor" or "this is manufactured urgency" to yourself is often enough to buy time to decide.

Sources

Europe PMC search, title and abstract, "dark psychology" and "Dark Triad" (29 September 2026): https://europepmc.org/search?query=TITLE_ABS%3A"dark psychology"
Open Library title search, "dark psychology" (29 September 2026): https://openlibrary.org/search?title="dark+psychology"
D. Pineda et al., "Dark tetrad personality traits also play a role in bullying victimization", Frontiers in Psychology 13 (2022): https://doi.org/10.3389/fpsyg.2022.984744
M. Duradoni et al., "Cognitive Empathy and the Dark Triad: A Literature Review", European Journal of Investigation in Health, Psychology and Education 13 (2023): https://doi.org/10.3390/ejihpe13110184
Influence at Work (Robert Cialdini), "Principles of Persuasion": https://www.influenceatwork.com/7-principles-of-persuasion/
A. Tversky and D. Kahneman, "Judgment under Uncertainty: Heuristics and Biases", Science 185 (1974): https://pubmed.ncbi.nlm.nih.gov/17835457/
A. L. Brown, T. Imai, F. M. Vieider and C. F. Camerer, "Meta-analysis of Empirical Estimates of Loss Aversion", Journal of Economic Literature 62 (2024): https://doi.org/10.1257/jel.20221698
L. Walasek, T. L. Mullett and N. Stewart, "A meta-analysis of loss aversion in risky contexts", Journal of Economic Psychology 103 (2024) 102740: https://wrap.warwick.ac.uk/id/eprint/185745/13/1-s2.0-S0167487024000485-main.pdf
D. Gal and D. D. Rucker, "The Loss of Loss Aversion: Will It Loom Larger Than Its Gain?", Journal of Consumer Psychology (2018): https://doi.org/10.1002/jcpy.1047
J. Adair, "Defining Gaslighting in Gender-Based Violence: A Mixed-Methods Systematic Review", Trauma, Violence, & Abuse (2026): https://doi.org/10.1177/15248380251344316
Virginia Commonwealth University News, "Psychologists urge an end to labeling antisocial traits as 'dark'" (30 June 2025), on D. Chester, D. Lynam and J. Miller, Journal of Psychopathology and Clinical Science (2025): https://news.vcu.edu/article/2025/06/psychologists-urge-an-end-to-labeling-antisocial-traits-as-dark
Serious Crime Act 2015, section 76, legislation.gov.uk: https://www.legislation.gov.uk/ukpga/2015/9/section/76

Checked September 2026.

Finding Meaning in Life: Unraveling 'Media Vita in Morte Sumus'

Hacks Vitae — Thu, 09 Nov 2023 07:36:57 GMT

Finding Meaning in Life

Media vita in morte sumus is Latin for "In the midst of life we are in death." It is the first line of a medieval church chant, an antiphon: a short sung text that goes with a psalm or canticle in the daily round of prayer. English service books of the Middle Ages set it at Compline, the last prayer of the day, for about two weeks in the middle of Lent. English speakers know it because the first Book of Common Prayer, in 1549, put it into the burial service.

It is not a Crusader or Templar battle chant. It was already in an English service book before the First Crusade was preached, and we found no medieval source that links it to the Crusades or to the Templars.

What the words say

Here is the whole antiphon as the Sarum Breviary gives it (Paris edition, 1531, printed in Julian's Dictionary of Hymnology), beside the English of the Book of Common Prayer as revised in 1662:

Latin	Book of Common Prayer (1662)
Media vita in morte sumus:	In the midst of life we are in death:
quem quaerimus adjutorem nisi te, Domine,	of whom may we seek for succour, but of thee, O Lord,
qui pro peccatis nostris juste irasceris?	who for our sins art justly displeased?
Sancte Deus, sancte fortis, sancte et misericors Salvator:	Yet, O Lord God most holy, O Lord most mighty, O holy and most merciful Saviour,
amarae morti ne tradas nos.	deliver us not into the bitter pains of eternal death.

Read to the end and the chant is a question and a plea. It asks who can help, answers that only God can, and asks not to be handed over to death. The last Latin line says only amarae morti, "bitter death"; "eternal" is the Prayer Book's addition.

The refrain Sancte Deus, sancte fortis has a longer history of its own. Julian's Dictionary of Hymnology says it is based on the Trisagion ("thrice holy"), an invocation introduced into the Greek service books in about AD 446.

What the chant is

Where it was sung. The Sarum (Salisbury) Breviary gives Media vita at Compline as the antiphon to the Nunc Dimittis, "for 15 days about the middle of Lent". It is also in the York Breviary printed in 1493.

How old it is. Nobody knows who wrote it or when. The earliest copy Julian's Dictionary cites is in a manuscript now in the British Library, Harley MS 2961, where it is set "for use during Lent". That manuscript is the Leofric Collectar, written at Exeter in the third quarter of the eleventh century. Pope Urban II called for the expedition that became the First Crusade at the Council of Clermont in 1095, so the chant was already being copied in England before the Crusades began.

What people did with it. Julian reports the German hymnologist Rambach's account that by the middle of the thirteenth century it had come into universal use as a prayer "in times of trouble", and that people also used it "as an incantation".

The Cologne ruling of 1310

The clearest medieval record of how it was used comes from a church council. In 1310 a council of the church province of Cologne, under Archbishop Henry of Virneburg, issued a set of statutes. The twenty-first reads:

Prohibemus item, ne in aliqua ecclesiarum nobis subiectarum imprecationes fiant, nec decantetur Media vita contra aliquas personas, nisi de nostra licentia speciali.

In our translation: "We likewise forbid that curses be made in any of the churches subject to us, or that the Media vita be sung against any persons, except by our special licence." The statute gives the reason: it is for the archbishop to decide "when such things are to be done".

Some books date this ruling to 1316; Julian's Dictionary gives "1316 (1310)". The statutes as printed in Hartzheim's Concilia Germaniae are headed 1310.

So by the early fourteenth century, people were singing a prayer for mercy at other people, as a curse, and the church wanted a say in when that happened.

The Notker legend

The best-known origin story says that Notker of St Gall, a monk who died in 912, wrote the antiphon after watching workmen build a bridge over a gorge near the abbey. Julian's Dictionary says this tradition "has not been traced earlier than the Chronicle of J. Metzler, written in 1613", and that no evidence can be found for either the story or the ascription. It adds that the antiphon appears in only three St Gall manuscripts, all comparatively late (one from the fourteenth century, one from 1431, one from 1507), and none of them names Notker.

Where the "Crusader battle chant" label came from

We could not find where the Crusader and Templar labels began. Here is what the sources do say.

MYTH, as a Crusader or Templar chant. The antiphon was being copied into an English service book for Lent before the First Crusade was preached, and we found no medieval source that connects it with the Crusades or with the Templars.

UNVERIFIED, as a battle song. Julian passes on the claim of a German writer, Bässler, in 1858, that "it was used as a war song by the priests accompanying the hosts before and during battle". Julian names no medieval witness for this, and we did not find one.

Where you have heard it

In German. German translations were in use by the fifteenth century. Martin Luther took one of them, altered it, added two stanzas of his own, and published the hymn Mitten wir im Leben sind ("In the midst of life we are") in the Erfurt Enchiridion of 1524. Julian says it took, and in his day still held, "a foremost place among German hymns for the dying".

In English. Miles Coverdale published an English version of Luther's hymn, "In the myddest of our lyvynge", in 1539. Ten years later the line entered the Church of England's services. The first Book of Common Prayer was first used on Pentecost Sunday, 9 June 1549, and was mainly the work of Thomas Cranmer, Archbishop of Canterbury. In its burial service, at the graveside, the priest says, or the priest and clerks sing: "In the myddest of lyfe we be in death", followed by the rest of the antiphon in English. The wording most people know, "In the midst of life we are in death", is the 1662 revision.

How to check a "medieval chant" you find online

The story of Media vita shows how to test any chant that arrives with a dramatic label:

Look for it in a service book. A chant that was really sung in church is in breviaries and antiphoners, and reference works like Julian's Dictionary of Hymnology say which ones and where in the year.
Compare dates. If the earliest copy is older than the event the label names, the label is later than the chant.
Separate a later writer from a medieval witness. "A nineteenth-century scholar said" is not the same as "a chronicle records".
Read to the last line. The last line of Media vita is a plea not to be handed over to death, which tells you more about the chant than the first line does.

The idea the first line carries is simple: death is not only at the end of life but in the middle of it. The antiphon's own answer is not to seize the day but to ask for help. Readers who want the classical versions of the same thought can compare memento mori in Older Than You Think and Horace's line in Carpe diem.

Sources

John Julian (ed.), A Dictionary of Hymnology, revised edition (John Murray, 1907), vol. 1, pp. 720–721, entry "Media vita in morte sumus": https://archive.org/details/dictionaryofhymn01lond
Joseph Hartzheim (ed.), Concilia Germaniae, vol. 4 (Cologne, 1761), Council of Cologne under Archbishop Henry of Virneburg, 1310, statute XXI; translation from the Latin is ours: https://archive.org/details/gri_33125010872436
EM1060: The Production and Use of English Manuscripts 1060–1220, "London, British Library, Harley 2961": https://em1060.stanford.edu/manuscript/london-british-library-harley-2961
Paul Halsall (ed.), Internet Medieval Sourcebook, "Urban II: Speech at Council of Clermont, 1095": https://sourcebooks.fordham.edu/source/urban2-5vers.asp
The Book of Common Prayer (1549), "The Order for the Burial of the Dead", transcribed by the Society of Archbishop Justus (archived copy): https://web.archive.org/web/20260613182833/http://justus.anglican.org/resources/bcp/1549/Burial_1549.htm
Society of Archbishop Justus, "The First Book of Common Prayer" (archived copy): https://web.archive.org/web/2026/http://justus.anglican.org/resources/bio/162.html
The Book of Common Prayer (1662), "The Order for the Burial of the Dead", transcription: http://www.eskimo.com/~lhowell/bcp1662/occasion/burial.html

Checked September 2026.

The Third Crusade: Battle for the Holy Land

Hacks Vitae — Fri, 03 Nov 2023 16:00:00 GMT

Battle for the Holy Land

The Third Crusade, 1189 to 1192, was the attempt by three Western rulers, Philip II of France, Richard I of England and the Emperor Frederick Barbarossa, to retake the Holy Land after Saladin captured Jerusalem in 1187. Because of who led it, it is also known as the Kings' Crusade.

It was partly successful. The crusaders took back Acre and Jaffa and much of the coast, but did not retake Jerusalem. This page follows what happened, using Saladin's biographer Baha ad-Din, who served in Saladin's army as its judge and saw the siege of Acre and the battle of Arsuf, and the standard modern accounts. Where the sources give different numbers, it gives the range.

I. Prelude to the Third Crusade

A. The Fall of Jerusalem

On 3–4 July 1187, Saladin destroyed the army of the Kingdom of Jerusalem at the Horns of Hattin in Galilee. Jerusalem was besieged from 20 September, and on 2 October 1187 Balian of Ibelin surrendered the city to him.

B. The Call to Crusade

News of the defeat reached the West in the early autumn of 1187, and the new pope, Gregory VIII, was soon proclaiming a new crusade. Henry II of England and Philip II of France stopped fighting each other to take part, and both raised the Saladin tithe, a ten per cent tax on income and movable goods, to pay for it. Henry died on 6 July 1189, so the English army went east under his son, Richard I.

II. The Leading Figures

A. Richard the Lionheart

Richard I of England was regarded, for about four centuries after his death, as a model king and a paragon of chivalry. His own time was more mixed: clergy chroniclers criticised him for taxing the Church for both the crusade and his ransom.

B. Philip II of France — who left early

Philip reached Acre on 20 April 1191, while it was already under siege, and began building siege equipment before Richard arrived on 8 June. By the time the city surrendered on 12 July 1191, he was seriously ill with dysentery. His relations with Richard had worsened, and the death at Acre of Philip, Count of Flanders, opened a question of succession back home. He left the Holy Land in August 1191, leaving 7,000 French crusaders and 5,000 silver marks to pay them.

C. Emperor Frederick I — who never arrived

Frederick I, Barbarossa, set out overland in May 1189. The size of his army is uncertain: a source written in the 1220s gives 100,000 men, which modern scholars think inaccurate; their estimates run from 12,000 to about 26,000 men. He won battles in Anatolia at Philomelium and Iconium.

He never reached the Holy Land. On 10 June 1190 he drowned in the Saleph river, in Cilician Armenia. The accounts of how differ. One says he chose to swim across and was swept away; the chronicler Ibn al-Athir says he went down to the river to wash and drowned where the water was not even up to his waist; and a churchman with the army wrote that he went to bathe to cool down after weeks of hard marching.

Most of his army then turned back. His son, Frederick of Swabia, led the rest on, and about 5,000 men reached Acre. An attempt to preserve the emperor's body in vinegar failed, and his remains were buried in three places: his flesh at Antioch, his bones at Tyre and his heart and inner organs at Tarsus. From then on the crusade was, in practice, the story of two kings.

III. The Battles

A. Siege of Acre

The siege of Acre ran from late August 1189 to 12 July 1191, almost two years. The crusaders took the city after Philip and then Richard arrived with their armies.

B. Battle of Arsuf

After Acre, Richard marched south along the coast towards Jaffa, which he needed before any attack on Jerusalem. On 7 September 1191 Saladin attacked his army at Arsuf, trying to make it break formation. Richard kept his army together until the Hospitallers broke ranks and charged; he then ordered a general counter-attack, which won the battle. Baha ad-Din, on Saladin's side, called it "a blow to all Moslem hearts".

Arsuf let Richard take and hold Jaffa. It did not destroy Saladin's army, and it did not decide the war: Jerusalem stayed in Saladin's hands.

C. Battle of Jaffa

In late July 1192 Saladin's army besieged Jaffa and stormed its walls; only the citadel held out. Richard, who had been about to leave for England, came back by sea with a small force of little more than 2,000 men, drove Saladin's troops out of the town and then beat off their counter-attack. The battle ended in failure for Saladin, and the treaty that ended the crusade followed within weeks.

IV. The Role of Saladin

A. Saladin's Leadership

Saladin had brought the forces of Egypt and Syria under his own control before 1187. His hold on his army had limits: in December 1191 pressure from his emirs forced him to disband most of it, and at Jaffa in 1192, by Baha ad-Din's account, he told the defenders' envoys that he could not stop his men from fighting on.

B. Negotiations

Richard and Saladin never met. Saladin refused, saying it was customary for kings to meet only after a peace treaty had been agreed. They exchanged gifts, and Richard had a number of meetings with Saladin's brother, al-Adil. One proposal discussed was a marriage between Richard's sister Joan, or his niece Eleanor, and al-Adil; it came to nothing.

V. Acre, Jerusalem and the road home

Richard is remembered as chivalrous; this is the rest of the record.

The prisoners at Acre

After Acre surrendered, Saladin agreed to pay the ransom for its garrison in three instalments, a month apart. Baha ad-Din says the first instalment (the relic of the True Cross, 100,000 dinars and 1,600 prisoners) was ready when it fell due on 11 August, but the crusaders' commissioners said not all the prisoners they had named had been brought together, and neither side would hand over first. On 20 August 1191 Richard had the Muslim prisoners killed in full view of Saladin's army.

The numbers differ. The Itinerarium Peregrinorum, a Latin chronicle of the early 1220s, gives 2,700; Wikipedia's article on the massacre, citing Steven Runciman's A History of the Crusades, gives about 2,700 to 3,000. Baha ad-Din says the prisoners numbered "more than three thousand, all tied together with ropes", and that only those of note and those strong enough to work were spared. Saladin responded by killing the Christian prisoners he had captured.

Richard did not besiege Jerusalem

He came close twice. In the winter of 1191 to 1192 his army reached Beit Nuba, about 12 miles from Jerusalem, then turned back in cold, heavy rain, for fear that an army besieging the city could be trapped by a relieving force. In June 1192 it came within sight of the city and retreated again, this time because its leaders could not agree: Richard and most of the army council wanted to attack Egypt, while the Duke of Burgundy insisted on a direct attack on Jerusalem.

He was captured on the way home

Richard left for home on 9 October 1192. After his ship was wrecked near Aquileia, he travelled overland and was captured near Vienna shortly before Christmas 1192 by Duke Leopold V of Austria, who accused him of arranging the murder of Conrad of Montferrat, and whom he had also offended at Acre by throwing down his banner. In March 1193 he was handed over to the Emperor Henry VI, who demanded 150,000 marks for his release. Over his whole reign Richard spent very little time in England, perhaps as little as six months.

VI. The Legacy of the Third Crusade

A. The Truce

The crusade ended with the Treaty of Jaffa, signed on 1 or 2 September 1192. It set a three-year truce. Jerusalem stayed under Muslim rule and was open to Christian pilgrims; the crusaders held the coast from Tyre to Jaffa; and Ascalon's fortifications were to be demolished. Balian of Ibelin sat on the crusader council that heard and accepted its terms. Saladin died in Damascus on 4 March 1193, six months later.

B. What changed

The Kingdom of Jerusalem, which had lost almost all its territory after Hattin, was restored as a coastal strip between Tyre and Jaffa, with its capital at Acre. Its crown had also changed hands during the crusade: in April 1192 the nobles elected Conrad of Montferrat king, and he was murdered before he could be crowned; Richard's nephew Henry of Champagne then married Queen Isabella.

VII. The Journey to the Holy Land

The three armies took different routes. Frederick went overland through Hungary and Byzantine territory, crossing into Anatolia in the spring of 1190. Philip and Richard left together from Vézelay on 4 July 1190 and split at Lyon: Philip crossed the Alps to Genoa, and Richard went by sea from Marseille. The French and English armies spent the winter together at Messina, in Sicily, and Philip sailed for the Holy Land on 30 March 1191.

VIII. The Impact Beyond the Battlefield

The crusade's financing reached far beyond the battlefield. The Saladin tithe taxed income and movable goods in England and France, and Richard's ransom of 150,000 marks was, according to the crusades historian Thomas F. Madden, the same amount the Saladin tithe had raised a few years earlier, and two to three times the English crown's annual income under Richard.

IX. Lessons and Reflections

Winning battles is not the same as winning the war. Richard won at Arsuf and Jaffa, and Jerusalem still stayed with Saladin.
Check who actually took part. Frederick never arrived and Philip left in August 1191; for most of the fighting, this was Richard's crusade.
Treat medieval numbers as ranges. Frederick's army is 100,000 in one later source and 12,000 to about 26,000 in modern estimates; the prisoners at Acre are 2,700 in one count and more than 3,000 in another.
A reputation is not the whole record. Richard's chivalry and the killings at Acre belong to the same man and the same summer.

Sources

Baha ad-Din ibn Shaddad, The Life of Saladin, translated by C. W. Wilson (Palestine Pilgrims' Text Society, 1897), Internet Archive (Jerusalem in 1187, the siege of Acre, the ransom talks, Arsuf, the killing of the prisoners at Acre, Jaffa in 1192, the peace of 1192, Saladin's death).
Itinerarium Peregrinorum et Gesta Regis Ricardi, as cited in Wikipedia's "Massacre at Ayyadieh" (Helen J. Nicholson's translation, 1997, p. 231) (the 2,700 figure).
Wikipedia, "Third Crusade", "Frederick Barbarossa", "Siege of Acre (1189–1191)", "Massacre at Ayyadieh", "Treaty of Jaffa (1192)", "Battle of Jaffa (1192)", "Richard I of England", "Philip II of France", "Battle of Hattin" and "Siege of Jerusalem (1187)", and the works they cite for the facts used here: John Gillingham, Richard the Lionheart (1979) and Richard I (1999); Christopher Tyerman, God's War (2006); John Freed, Frederick Barbarossa: The Prince and the Myth (2016); G. A. Loud, The Crusade of Frederick Barbarossa (2010); Jim Bradbury, Philip Augustus (1997); John W. Baldwin, The Government of Philip Augustus (1991); Hans Eberhard Mayer and Jonathan Riley-Smith in The New Cambridge Medieval History, vol. IV (2008); Anne-Marie Eddé, Saladin (2011); Frank McLynn, Richard and John (2007); Charles Oman, A History of the Art of War in the Middle Ages; Thomas F. Madden, Crusades: The Illustrated History (2005); Steven Runciman, A History of the Crusades, vol. III.

Checked September 2026. Dates and events follow Baha ad-Din's account and the Wikipedia articles listed, used only where they give a citation for the sentence (the works cited are named above); where they give different figures (the size of Frederick's army, the number of prisoners killed at Acre, the date of the treaty), the page gives the range.

The June Rebellion and French Revolution: A Look at French History and Uprisings

Hacks Vitae — Mon, 30 Oct 2023 16:00:00 GMT

A Look at French History and Uprisings

The June Rebellion, also called the Paris Uprising of 1832, was an anti-monarchist rising by Parisian republicans on 5 and 6 June 1832. Many people first meet it in Les Misérables, which is often assumed to be set during the Revolution of 1789. This page sets out how the two are connected, what happened in 1832, and why it matters that the king the rebels faced was Louis Philippe, not Charles X.

The French Revolution: A Revolution of Ideals

The Seeds of Revolution

The French Revolution was a period of political and social change that began with the Estates General of 1789 and ended with Napoleon's coup of 18 Brumaire on 9 November 1799. It was caused by a combination of social, political and economic factors, and historians still weigh them differently.

Key Events of the Revolution

The storming of the Bastille in Paris on 14 July 1789 led to a series of radical measures, including the abolition of feudalism and the Declaration of the Rights of Man and of the Citizen. The monarchy was replaced by the First Republic in September 1792. The Reign of Terror followed in 1793 and 1794. Napoleon, who rose to prominence as a general during the Revolution, became Emperor of the French on 18 May 1804.

Why you have probably heard of it: Les Misérables

The barricade in Victor Hugo's Les Misérables belongs to this rising. The novel begins in 1815 and climaxes with the battles of the June Rebellion of 1832. Its secret society, the Friends of the ABC, made up mostly of students, is led by Enjolras, and Marius fights on the barricade. The stage musical and the films are based on the novel.

Hugo was in Paris on the day. On 5 June 1832 he was writing a play in the Tuileries gardens when he heard gunfire from the direction of Les Halles. He followed the sound, turned into the Passage du Saumon, and found himself shut in between barricades; for a quarter of an hour, bullets flew both ways while he sheltered between some columns. He published Les Misérables thirty years later, in 1862.

So the rising is a real event that many people know through a musical, which is a good reason to keep track of which parts are history and which are Hugo's.

The June Rebellion: A Lesser-Known Chapter

Origins of the June Rebellion

In the July Revolution of 1830 the elected Chamber of Deputies had set up a constitutional monarchy and replaced Charles X with his more liberal cousin, Louis Philippe. That angered republicans, who saw one king replaced by another, and by 1832 there was a feeling that their revolution, for which many had died, had been stolen. The June Rebellion was an attempt to reverse the July Monarchy of Louis Philippe. It was not aimed at Charles X, who was no longer king.

Triggers and Catalysts

France had been through hard years from 1827 to 1832: harvest failures, food shortages and a rising cost of living. Then, in the spring of 1832, Paris suffered a widespread outbreak of cholera, which killed 18,402 people in the city and 100,000 across France, by the figures Wikipedia gives from the historian Jill Harsin. The poor neighbourhoods were devastated, and suspicion spread that the government had poisoned the wells.

The epidemic took two well-known victims: the prime minister, Casimir Perier, who died on 16 May, and General Jean Maximilien Lamarque, who died on 1 June. Lamarque was a hero of the Napoleonic Wars and, from 1828, a deputy on the left in the Chamber of Deputies, which made his public funeral an opportunity for the opposition to show its strength.

What happened


When	5–6 June 1832
Where	Paris: the insurgents held much of the eastern and central districts for one night
Against	King Louis Philippe and the July Monarchy
Rebels	Roughly 3,000
Government	20,000 National Guard and about 40,000 regular troops (Wikipedia's summary box gives 30,000 for the whole government side)
Casualties	About 800 killed and wounded on both sides
Outcome	Suppressed. The last fighting, at the Cloître Saint-Merry, went on until the early evening of 6 June

Republican conspirators made their move at Lamarque's funeral on 5 June. Demonstrators took charge of the procession and redirected it to the Place de la Bastille. When a red flag reading La liberté ou la mort ("liberty or death") was raised, the crowd broke into disorder and shots were exchanged with government troops. Lafayette, who had spoken in praise of Lamarque, called for calm, but the outbreak spread.

The roughly 3,000 insurgents held much of eastern and central Paris for one night. During that night the 20,000 part-time militiamen of the Paris National Guard were reinforced by about 40,000 regular army troops. The insurgents made their stronghold in the Faubourg Saint-Martin, building barricades in the narrow streets around the rue Saint-Martin, and the final struggle came at the Cloître Saint-Merry. The army and National Guard lost 73 killed and 344 wounded; the insurgents 93 killed and 291 wounded.

Those identified afterwards were mostly working people: 66% working-class, a high proportion of them construction workers, and 34% shopkeepers or clerks.

Was Lamarque's death the cause? Not on its own. Historians treat it as the occasion rather than the cause. In the words of the historian Mark Traugott, there were "simmering discontents, especially strong among republicans, who felt that they had spilled their blood on the 1830 barricades", only to see their revolution "stolen".

Contrasting the French Revolution and the June Rebellion

Scope and Scale

The Revolution of 1789 ran for ten years and changed the whole country. The June Rebellion lasted two days, was fought in the streets of Paris, and failed to spread further.

A Difference in Aims

The Revolution set out to remake French society and government. The June Rebellion had a narrower aim: to undo what republicans saw as the betrayal of 1830 and replace Louis Philippe's monarchy with a republic.

Legacy and Influence

The French Revolution's Enduring Legacy

Many of the Revolution's ideas are considered fundamental principles of liberal democracy, and its values remain central to French political discourse today.

The Impact of the June Rebellion

In the short term the rising changed little. The government portrayed the rebels as an extremist minority. Delacroix's Liberty Leading the People, a painting of the 1830 revolution, had been bought by the state in August 1831. The art historian Albert Boime wrote that after the rising "it was never again openly displayed for fear of setting a bad example". The Louvre's own record is a little different: it lists the painting at the Musée du Luxembourg in 1832 and 1833, then in storage from about 1833–34. The state reclaimed it from Delacroix in March 1848, showed it at the Exposition Universelle of 1855, and moved it to the Louvre in November 1874. Louis Philippe's reign lasted until the revolution of 1848, when he abdicated and the Second Republic was founded.

What you can do with this

Check which king. "Against the monarchy" in 1832 means Louis Philippe. A date and a name are the quickest way to tell one French revolution from another.
Separate the spark from the fuel. Lamarque's funeral set the day; cholera, hunger and the sense of a stolen revolution were already there.
Enjoy Les Misérables as a novel. Hugo was near the fighting, but the book is fiction; the history is the frame, not the story.
Read numbers with their source. The count of government troops differs between sources, so a single figure is worth checking before you repeat it.

Sources

Wikipedia, "June Rebellion" (dates, causes, the funeral, the numbers on each side, casualties, aftermath, Hugo on 5 June), and the works it cites for them: Mark Traugott, The Insurgent Barricade (2010); Jill Harsin, Barricades: The War of the Streets in Revolutionary Paris, 1830–1848 (2002); Philip Mansel, Paris Between Empires (2003); William Duckett (ed.), Dictionnaire de la conversation et de la lecture, vol. 11 (casualties); Graham Robb, Victor Hugo: A Biography (1998); Albert Boime, Art in an Age of Civil Struggle, 1848–1871 (2008).
Wikipedia, "Jean Maximilien Lamarque" (citing Traugott, and Charles Mullié's Biographie des célébrités militaires, 1852).
Wikipedia, "French Revolution" (dates and key events, citing among others Simon Schama, Citizens, 1989, and Frank McLynn, Napoleon, 1997), "Reign of Terror" (citing Will Durant, The Age of Napoleon, 1975) and "Napoleon".
Wikipedia, "Louis Philippe I" (his abdication in 1848, citing the Library of Congress research guide on the revolutions of 1789, 1830 and 1848).
Victor Hugo, Les Misérables, translated by Isabel F. Hapgood (1887), Project Gutenberg (the Friends of the ABC; Marius at the barricade).
Tim Masters, "Bon anniversaire! 25 facts about Les Mis", BBC News (1 October 2010).
Musée du Louvre, collections record for Le 28 juillet 1830. La Liberté guidant le peuple (RF 129), object history.

Checked September 2026. Wikipedia is used here only where it gives a citation for the sentence, and the work it cites is named above; the Louvre's record is used for the painting, and the novel itself for its characters. Where sources give different numbers (the government's troops) or accounts (the painting after 1832), the page gives both.

3 Mind Blowing Quotes by Alan Watts

Hacks Vitae — Sun, 29 Oct 2023 16:00:00 GMT

Alan Watts Quotes

Three lines from Alan Watts that get shared constantly, with the book each one comes from, what the passage around it is saying, and a short note on who he was. All three can be traced to his own work: two to books he wrote, one to a lecture recorded near the end of his life and published after his death.

Who Alan Watts was

Alan Watts (1915–1973) was born on 6 January 1915 in Chislehurst, England, and moved to the United States in 1938. He trained at Seabury-Western Theological Seminary, was ordained an Episcopal priest in 1944, and served as chaplain at Northwestern University until 1950, when he left the church. From the 1950s he became what one reference work calls a "pioneer popularizer of Eastern philosophy in the United States", through his books, his lectures and radio broadcasts on Berkeley's KPFA from 1953. He died on 16 November 1973.

He was careful not to present himself as a religious authority. One reference work notes that he did not wish to identify with any religious group, "on the ground that partisanship in religion closes the mind", and that he once called himself a spiritual "entertainer". The same entry describes his thinking as "more an amalgamation of ideas than traditional Zen", drawing on Taoism and Advaita Vedanta as well. Monica Furlong's 1986 biography, Zen Effects, as summarised in its Publishers Weekly review, describes heavy drinking and a man who privately insisted he was "a rogue, a fake and entertainer".

None of that makes the ideas wrong. It does suggest a useful way to read him: as a doorway into these traditions rather than an authority on them. If one of his lines lands, the traditions he was interpreting are the next place to look.

Embracing Oneness

"We do not 'come into' this world; we come out of it, as leaves from a tree."

— Alan Watts, The Book: On the Taboo Against Knowing Who You Are (1966), chapter 1, "Inside Information"

The line comes from the first chapter of The Book. Just before it, Watts describes the feeling most of us have of being a separate "I" living inside a body and confronting an "external" world, and calls it "a hallucination". The leaves-and-tree line is his alternative picture: in the sentences that follow, each person is an expression of the whole of nature rather than a visitor to it.

Read that way, the line is less about mysticism than about a habit of thought. It asks you to notice how often everyday language ("I came into this world", "the conquest of nature") assumes a split between you and everything else, and to consider whether the split is as sharp as it feels.

Dancing with Change

"The only way to make sense out of change is to plunge into it, move with it, and join the dance."

— Alan Watts, The Wisdom of Insecurity (1951), chapter III, "The Great Stream"

A central claim of the book is that "the desire for security and the feeling of insecurity are the same thing". In his image, "to hold your breath is to lose your breath." Just before this line Watts writes that "fixing" will never make sense out of change; just after it, he turns to religion, which he says has tried to make sense of life by fixing it in place.

So the line is a claim about method rather than a pep talk. Holding tighter to what is changing does not help you understand it; taking part in it might.

Engaging in the Present

"This is the real secret of life — to be completely engaged with what you are doing in the here and now. And instead of calling it work, realize it is play."

— Alan Watts, "Work as Play", in The Essential Alan Watts (1977)

This one is not from a book Watts wrote as a book. It comes from a chapter of The Essential Alan Watts, published by Celestial Arts in 1977, four years after his death. The collection's foreword says the "Work as Play" chapter was derived from The Essential Lectures of Alan Watts, video programmes he recorded in 1971. The words are his, edited from a spoken talk.

In the talk the line follows examples of doing something for its own sake, such as chanting or shooting an arrow into the sky, and it leads into the Hindu idea of creation as lila, which he glosses as "dance or play". The point is about attention: an activity done wholeheartedly for itself feels different from one done only to get to the end of it.

Using These Lines Well

Keep the source with the quote. A line with its book and year attached can be checked in minutes; a line without them cannot. All three here can be.
Read the paragraph around it. Each of these makes more sense, and says something more specific, in context than on a poster.
Treat them as prompts, not proofs. Watts offered ways of looking. Whether a way of looking helps is something you can test in your own life, one week at a time.

Sources

Alan Watts, The Book: On the Taboo Against Knowing Who You Are (New York: Pantheon, 1966), chapter 1, "Inside Information"
Alan Watts, The Wisdom of Insecurity (New York: Pantheon, 1951), chapter III, "The Great Stream"
Alan Watts, The Essential Alan Watts (Berkeley: Celestial Arts, 1977), "Work as Play" and the foreword
Alan Watts Organization, biography
"Watts, Alan (Wilson) (1915-1973)", Encyclopedia.com
Review of Monica Furlong, Zen Effects: The Life of Alan Watts (Houghton Mifflin, 1986), Publishers Weekly

Checked September 2026.

The Ride of the Rohirrim: Theoden's Inspiring Speech and the Greatest Cavalry Charge Scene

Hacks Vitae — Sun, 29 Oct 2023 16:00:00 GMT

The Ride of the Rohirrim

"The Ride of the Rohirrim" is the charge of the Riders of Rohan onto the Pelennor Fields, outside the besieged city of Minas Tirith. It is fiction. It comes from J.R.R. Tolkien's novel The Return of the King, where it is the title of a chapter, and the version most people picture is from Peter Jackson's film The Lord of the Rings: The Return of the King (2003), with Bernard Hill as King Théoden.

The film and the novel give Théoden different words. Both are below, with where each line comes from.

The Context: A World on the Brink

In the story, Minas Tirith in Gondor is under siege. In the novel the city's gate has just been broken ("As if stricken by some blasting spell it burst asunder") when horns are heard from the north: "Rohan had come at last." Tolkien's verse puts their host at "six thousand spears".

Théoden gives his orders first. In the novel he tells his captains who leads on which side, admits "Other plans we cannot make, for we know not yet how things stand upon the field", and ends: "Forth now, and fear no darkness!" Soon after, "rising in his stirrups", he gives the cry.

Theoden's Inspiring Speech: A Call to Arms

The film (Bernard Hill as Théoden; wording as in the tk421.net transcript):

"Arise! Arise! Riders of Théoden! Spears shall be shaken, shields shall be splintered! A sword-day! A red day, ere the sun rises!"

"Ride now, ride now, ride! Ride for ruin and the world's ending!"

"Death!" … "Forth Eorlingas!"

Tolkien's text (The Return of the King, Book V, chapter 5, "The Ride of the Rohirrim"):

Arise, arise, Riders of Théoden! Fell deeds awake: fire and slaughter! spear shall be shaken, shield be splintered, a sword-day, a red day, ere the sun rises! Ride now, ride now! Ride to Gondor!

Then, in the novel, Théoden "seized a great horn from Guthláf his banner-bearer, and he blew such a blast upon it that it burst asunder."

What the film changed:

Singular to plural. Tolkien has "spear shall be shaken, shield be splintered"; the film has "Spears shall be shaken, shields shall be splintered".
A line cut. "Fell deeds awake: fire and slaughter!" is not in the film's speech.
A new ending. Tolkien's Théoden ends "Ride to Gondor!" The film's "Ride for ruin and the world's ending!" echoes a later moment in the novel, in the chapter "The Battle of the Pelennor Fields": when Éomer finds his sister Éowyn among the fallen, he spurs to the front and cries "Death! Ride, ride to ruin and the world's ending!"
A cry from the previous book. "Forth Eorlingas!" is the cry of Rohan's riders in The Two Towers.
The orders. The film keeps "Forth, and fear no darkness!" but reshuffles the captains: on screen Éomer takes the left, Gamling follows the king's banner and Grimbold takes the right; in the novel Éomer leads the first éored behind the king's banner in the centre, Elfhelm takes the right and Grimbold the left.

The Cavalry Charge: A Cinematic Triumph

Whether it is the greatest cavalry charge put on film is a matter of taste. What can be checked is how the film was received. At the 2004 Academy Awards, The Return of the King won Best Picture, Best Director for Peter Jackson, Best Original Score for Howard Shore and Best Visual Effects, among the 11 Oscars it won.

In the story, Théoden does not survive the battle. In the novel, as he rallies his riders ("Up Eorlingas! Fear no darkness!"), his horse Snowmane is struck by a black dart and falls, and the king falls beneath him.

Is any of it history?

Rohan, the Rohirrim, Minas Tirith and the Pelennor Fields exist only in Tolkien's fiction, and the charge on screen is a film sequence. A quotation of the speech is clearest when it says whether it is Tolkien's or the film's.

What the scene shows about a rallying speech

Read the novel's version in order and it has a shape worth noticing, whatever you think of the prose:

Orders before rhetoric. Théoden first says who leads where and admits what he does not know ("Other plans we cannot make"). Only then does he give the cry.
Short, concrete lines. The cry is five short lines built on things the riders can picture: spear, shield, sword-day, sunrise.
A single destination. Tolkien's version ends on one instruction: "Ride to Gondor!"

That is our reading of the scene, not a rule from Tolkien. The practical habit it suggests is a small one: with a famous speech, it helps to know which version you are quoting. The film's words and Tolkien's are both memorable, and they are not the same.

Sources

J.R.R. Tolkien, The Return of the King (Quality Paperback Book Club edition, 2001), Book V, chapter 5, "The Ride of the Rohirrim", pp. 111–112, and chapter 6, "The Battle of the Pelennor Fields", pp. 115 and 119; also pp. 77, 102–103 and 196. Read via the Internet Archive's search-inside text.
J.R.R. Tolkien, The Two Towers (Houghton Mifflin, 1993), pp. 123 and 146. Read via the Internet Archive's search-inside text.
The Lord of the Rings: The Return of the King (film, 2003), transcript at tk421.net: https://www.tk421.net/lotr/film/rotk/22.html
Academy of Motion Picture Arts and Sciences, The 76th Academy Awards (2004): https://www.oscars.org/oscars/ceremonies/2004
Variety, "Bernard Hill, 'Lord of the Rings' and 'Titanic' Actor, Dies at 79" (5 May 2024): https://variety.com/2024/film/news/bernard-hill-dead-lord-of-the-rings-1235992436/

Checked September 2026.

J. Robert Oppenheimer: A Man of Science and Conscience ∙ Bhagavad Gita Speech "Destroyer of Worlds"

Hacks Vitae — Mon, 23 Oct 2023 16:00:00 GMT

Bhagavad Gita Speech "Destroyer of Worlds"

J. Robert Oppenheimer did say "Now I am become Death, the destroyer of worlds", and he said it on film. The clip everyone knows was filmed for NBC's documentary The Decision to Drop the Bomb, broadcast on 5 January 1965, nearly twenty years after the first atomic bomb test. In it he remembers the test, on 16 July 1945, like this:

"We knew the world would not be the same. A few people laughed, a few people cried, most people were silent. I remembered the line from the Hindu scripture, the Bhagavad-Gita."

He goes on to say that Vishnu, trying to persuade a prince to do his duty, "takes on his multi-armed form and says, 'Now, I am become Death, the destroyer of worlds.' I suppose we all thought that one way or another."

So the line is real, it is his, and it is a memory: he describes it as something that came to mind while he watched. This page covers where the line comes from, what the Sanskrit says, and what Oppenheimer did and said about nuclear weapons afterwards.

The Influence of the Bhagavad Gita

Oppenheimer was born in New York City on 22 April 1904. At Harvard he read Hindu texts in English translation. In 1933, as a young physics professor at Berkeley, he began studying Sanskrit with Arthur W. Ryder, and he went on to read the Bhagavad Gita in the original.

On 7 October 1933 he wrote to his brother Frank that he had been reading the Gita with two other Sanskritists. He called it "the most beautiful philosophical song existing in any known tongue", and he gave copies to friends. Wikipedia's biography notes that he later cited it as one of the books that most shaped his philosophy of life.

He was using it long before 1965. At the memorial service for President Franklin D. Roosevelt at Los Alamos in April 1945, he quoted his own translation of Gita 17.3: "Man is a creature whose substance is faith. What his faith is, he is."

The Gita is a conversation on a battlefield. The prince Arjuna does not want to fight in a war against his own cousins, friends and teachers, and his charioteer, the god Krishna, argues with him about duty. Oppenheimer's "Vishnu" is Krishna: Krishna is an avatar of Vishnu.

The "Destroyer of Worlds"

Oppenheimer was the director of the Manhattan Project's Los Alamos laboratory, and he was present at the Trinity test in New Mexico. The famous line has a longer history in print than on film:

1948. The earliest printed version the historian James Hijiya could find appeared in Time magazine on 8 November 1948, with "shatterer" rather than "destroyer".
1958. The "shatterer" version appeared again in Robert Jungk's Brighter than a Thousand Suns, whose author had interviewed Oppenheimer; the quotation may come from that interview.
1965. The filmed version, with "destroyer", made for NBC.

The New York Times reporter William Laurence, who interviewed Oppenheimer a few hours after the test, later claimed that he would never forget the "shattering impact" of Oppenheimer's words, but Hijiya found no printing of the line by Laurence before 1959.

What he said out loud at the moment is a different question. His brother Frank recalled his first words as "I guess it worked."

Alex Wellerstein, who writes the nuclear history blog Restricted Data, describes the famous account as one in which Oppenheimer "says (or at least thinks)" the lines. Read that way, it is a verse he later said came to mind, told on film in his own words nearly twenty years on.

What the Sanskrit says

The verse is Bhagavad Gita 11.32, from the passage where Krishna shows Arjuna his universal form. It opens kālo 'smi loka-kṣaya-kṛt pravṛddho.

The key word is kāla. In a word-by-word gloss of the verse it is given as "time". The scholar of religion Vasudha Narayanan notes that kāla also means "death". Both readings have support: "time" is the word-for-word gloss, and "death" is a real second meaning, not a mistake.

Oppenheimer was not the first to choose "Death". His teacher Ryder's translation, published in 1929, renders the verse "Death am I, and my present task Destruction." Wellerstein, summarising James A. Hijiya of the University of Massachusetts Dartmouth, calls Ryder's translation "a somewhat idiosyncratic but defensible one", and notes that others have translated the word as "time". Hijiya puts it more strongly: since the first English translation in 1785, "most experts have translated this word not as Death but as Time."

Two modern translations, for comparison:

"Time I am, the great destroyer of the worlds" (Bhagavad-gītā As It Is).
"I am mighty Time, the source of destruction that comes forth to annihilate the worlds" (Swami Mukundananda).

The rest of the verse matters too. In Mukundananda's translation it goes on: "even without your participation, the warriors arrayed in the opposing army shall cease to exist." Wellerstein, drawing on Hijiya, argues that in this picture Oppenheimer is not Krishna but Arjuna, the prince who did not want to fight: "It is the bomb that is here for destruction. Oppenheimer is merely the man who is witnessing it." That is one reading, not the only one.

Oppenheimer's Complex Relationship with Nuclear Weapons

"Now we are all sons of bitches" was not his line. It was said to him, immediately after the Trinity test, by the physicist Kenneth Bainbridge, who directed the test.

In August 1945 atomic bombs were used on Hiroshima and Nagasaki. On the evening of the Hiroshima bombing, at an assembly at Los Alamos, the crowd cheered Oppenheimer, and he expressed regret that the weapon was ready too late for use against Nazi Germany. On 17 August he travelled to Washington to hand-deliver a letter to the Secretary of War, Henry L. Stimson.

On 25 October 1945 he met President Harry S. Truman in the Oval Office. Truman did not record his Oval Office conversations, and accounts of what was said differ. In Dean Acheson's account, printed in The New York Times in 1969, Oppenheimer said "I have blood on my hands", and Truman afterwards told Acheson not to bring him in again. In a memo written months later, Truman called him a "cry baby" scientist.

After the war Oppenheimer lobbied for international control of nuclear power and weapons, to avert an arms race with the Soviet Union. In 1947 he became director of the Institute for Advanced Study in Princeton and chairman of the General Advisory Committee of the new Atomic Energy Commission. He later opposed the development of the hydrogen bomb, partly on ethical grounds.

The 1954 security hearing

In 1954 the Atomic Energy Commission held a security hearing on Oppenheimer. Its Personnel Security Board voted 2 to 1 to revoke his clearance: it unanimously cleared him of disloyalty, but a majority found that he would represent a security risk. On 29 June 1954 the Commission upheld the board's findings by a 4 to 1 decision.

In 1963 the same Commission selected him for the Enrico Fermi Award. He was diagnosed with throat cancer in late 1965 and died in his sleep at his home in Princeton on 18 February 1967, aged 62.

On 16 December 2022 the US Secretary of Energy, Jennifer Granholm, vacated the 1954 decision. Her statement said that the Commission had revoked his clearance "through a flawed process that violated the Commission's own regulations."

What you can do with this

When a famous line has a date on it, look for the first time it appeared. Here the earliest print version found is from 1948 and the film from 1965, both years after the event. That does not make the line false; it tells you it is a recollection.
With a translated quotation, ask whose translation it is. "Death" and "time" are both defensible readings of kāla; the choice changes how the line sounds.
Keep the famous words separate from the people around them. "Now we are all sons of bitches" was said to Oppenheimer, not by him.
Read the verse around the line. Gita 11.32 goes on to say the warriors will fall even without Arjuna, which changes how the line reads.

Sources

Atomic Archive, "J. Robert Oppenheimer 'Now I am become death...'" (transcript of the 1965 clip).
James A. Hijiya, "The Gita of J. Robert Oppenheimer", Proceedings of the American Philosophical Society 144, no. 2 (June 2000), 123–167.
Alex Wellerstein, "Oppenheimer and the Gita", Restricted Data: A Nuclear History Blog (23 May 2014; archived copy).
The Paley Center for Media, catalogue record for NBC White Paper: The Decision to Drop the Bomb (NBC, 5 January 1965).
Vasudha Narayanan, "Oppenheimer often used Sanskrit verses, and the Bhagavad Gita was special for him", The Conversation (16 August 2023).
Bhagavad Gita 11.32: Bhagavad-gītā As It Is (Vedabase), and Swami Mukundananda's translation (holy-bhagavad-gita.org).
Wikipedia, "J. Robert Oppenheimer", "Kenneth Bainbridge" and "Arthur W. Ryder".
Contingent Magazine, "Blood on Our Hands" (7 December 2023), on the accounts of the Truman meeting.
US Department of Energy, "Secretary Granholm Statement on DOE Order Vacating 1954 Atomic Energy Commission Decision in the Matter of J. Robert Oppenheimer" (16 December 2022).

Checked September 2026. The 1965 wording is taken from the Atomic Archive transcript of the clip; the 1948 and 1958 printings from James Hijiya's article and Wikipedia's biography; the hearing details from Wikipedia's biography, which cites the published histories; the Truman meeting from the accounts gathered in Contingent Magazine.

Napoleon: "There is nothing we can do." Meaning and A Historical Perspective on the Famous Meme

Hacks Vitae — Sun, 22 Oct 2023 16:00:00 GMT

A Historical Perspective on the Famous Meme

Napoleon never said "There is nothing we can do." The line was written for a meme in 2023, about two hundred years after he died.

It comes from a TikTok photo slideshow posted on 15 August 2023. Two rage-comic figures, drawn as Napoleon and one of his men exiled on St Helena, trade two lines: "Napoleon, what will we do?" and "There's nothing we can do." The slideshow then cuts to paintings of Napoleon on the island, set to a slowed remix of the song "Amour plastique" by Videoclub.

We found no letter, speech or memoir in which Napoleon says it, and the meme never claimed one: it started as dialogue written for two cartoon figures. So the answer to "did Napoleon say it?" is no. The more interesting question is why it stuck to him, and that one has a real answer.

"There is nothing we can do" Meaning

In the meme, the line means resignation. Something is lost, there is no move left, and the only thing to do is stare out to sea. Creators paired it with all kinds of prompts, and much of the joke is the gap between the problem and the gloom.

The pictures matter as much as the words. Most of the images are paintings of Napoleon on St Helena, the island where he spent his last years as a prisoner. That is why the line lands: it borrows the one stretch of his life when he really had run out of moves.

The man himself tells a different story. When a general wrote to tell him something could not be done, Napoleon's answer went the other way. That letter is below.

"There is nothing we can do." Meme

The first slideshow drew roughly 1.2 million plays in a month. It was copied by another account on 17 August 2023, but the format only became common in late September 2023. By 21 September 2023 the slowed "Amour plastique" sound behind it had been used in more than 22,600 posts.

Most versions follow the same three steps: a prompt, the line, then a painting of Napoleon looking sad and thoughtful on St Helena.

What Napoleon Wrote About "Impossible"

On 9 July 1813, at Dresden, Napoleon answered a letter from General Lemarois, the governor of Magdeburg. Lemarois had told him that something he wanted done was not possible. The reply, as printed in his collected correspondence:

Vous avez 240,000 boisseaux d'avoine à Magdeburg. « Ce n'est pas possible », m'écrivez-vous : cela n'est pas français. Je suis donc mécontent de votre lettre.

In our translation: "You have 240,000 bushels of oats at Magdeburg. 'It is not possible,' you write to me: that is not French. I am therefore displeased with your letter." He then ordered two boatloads of oats sent at once for the horses of the Guard, "which are dying".

The saying Impossible n'est pas français ("Impossible is not French") is commonly linked to this letter. The letter's own words are longer, and the subject was oats. Still, it is the opposite of the meme: a man refusing to accept "there is nothing we can do" from someone else.

The Rise of Napoleon: Not From Nothing

Napoleon was born in Ajaccio, Corsica, on 15 August 1769. He did not come from humble beginnings. His family's noble status was officially recognised in September 1771, and that recognition is what let him enter the royal military school at Brienne on 15 May 1779 as a king's scholar. The school was meant for nobles short of money, and four degrees of nobility were required.

His father, Charles (Carlo) Bonaparte, was elected to the Corsican Estates as a representative of the nobility of the province of Ajaccio in 1771, 1772 and 1777.

From Brienne he went to the École militaire in Paris on 22 October 1784. In the exams of September 1785, 58 of 137 candidates passed and Napoleon came 42nd, commissioned as a second lieutenant in the artillery.

The rest of the rise is well known. He took power in the coup of 18 Brumaire (9 November 1799) and was crowned Emperor of the French in Notre-Dame on 2 December 1804, placing the crown on his own head.

The Fall of Napoleon: A Series of Unfortunate Events

Russia, 1812. In June 1812 nearly 440,000 men were lined up along the Niemen, ready to invade Russia. Counts of what came back differ: the Fondation Napoléon's preface to his correspondence gives about sixty thousand troops recrossing the Niemen, while World History Encyclopedia says fewer than 100,000 of the 615,000 men committed to the whole campaign returned. The crossing of the Berezina in late November alone cost between 25,000 and 40,000 men. On 5 December Napoleon left the army at Smorgon to return to Paris.

Elba, 1814. He abdicated at Fontainebleau, unconditionally by 6 April 1814, and was sent into exile on the island of Elba.

Waterloo, 1815. On 1 March 1815 he landed at Golfe-Juan, and by 20 March he was back in the Tuileries palace. On 18 June 1815 he was defeated at Waterloo, in Belgium, by the British and Prussian allied armies. He abdicated on 22 June and gave himself up to the British navy.

St Helena: The Real Dead End

This is the part of his life the meme borrows. Napoleon reached St Helena on 15 October 1815. He had already escaped from Elba once, and this time the guard was heavy: the Fondation Napoléon describes the island as "well guarded by three thousand men, and four vessels constantly surveyed its coastline to prevent any landing."

It was a genuine dead end, and what he did with it is the part the meme leaves out. His days were built around meals, riding and dictating his memoirs to his companions. He had brought more than 600 books with him. He died on St Helena on 5 May 1821, at the age of fifty-one.

Legacy and Impact

The Civil Code. Promulgated on 21 March 1804, the Code civil ran to 2,281 articles. Its property laws swept away feudal traditions, and in France and across continental Europe it fixed ideas such as equality, citizenship, freedom of conscience and the protection of property. It also restored the authority of the husband: a married woman could not perform any legal act or manage her own property without her husband's agreement.

Slavery. A law of 20 May 1802 provided that slavery would be maintained in the colonies returned to France under the Treaty of Amiens. The Fondation Napoléon calls the re-establishment of slavery, and the repression that followed, a stain on the posterity of the consular regime.

Both belong in the story. The meme is about a lonely man on an island; the record is about a ruler whose laws spread across Europe, for better and for worse.

What to Do With a Quote Like This

A line that goes viral under a famous name can usually be checked in a few minutes. For this one, four questions do it:

Is there a date and a document? A real quotation comes with a letter, a speech, a book or a recorded conversation. This one comes with a TikTok date and nothing earlier.
Is there an original-language version? His collected correspondence is in French. An English line with no French behind it is a caption, not a quotation.
Does finding the words prove the quote? No. His letters contain ordinary phrases such as il ne faut rien faire à présent ("nothing must be done for now", about a fortified post at the mouth of the Havel, 16 May 1813). Matching words in a different sense is not the same line.
What did he say on the subject? Here the record gives the opposite: "that is not French".

The meme is still worth enjoying as a meme. It is a good joke about resignation, and St Helena was a real place where the options ran out. It helps to keep the two apart: the mood belongs to 2023, and the man belongs to the record. For a longer method, see how to spot a fake quote or an invented source, and for the rest of his life, Napoleon Bonaparte: the facts.

Checked September 2026.

Sources

Know Your Meme, Napoleon "There Is Nothing We Can Do" (origin and spread of the meme, dated entries for August and September 2023).
Correspondance de Napoléon Ier, vol. 25 (Paris, 1868), letters no. 20014 (to Berthier, Dresden, 16 May 1813) and no. 20256 (to General Lemarois, Dresden, 9 July 1813). Scanned copy, Internet Archive.
Le Journal du Dimanche, D'où vient l'expression « Impossible n'est pas français » ?.
World History Encyclopedia, Napoleon Bonaparte and Napoleon's Invasion of Russia.
Fondation Napoléon, 1769–1793: Napoleon Bonaparte's early years.
Fondation Napoléon, Timeline: Consulate / First French Empire.
Fondation Napoléon, Preface to volume 12 of the General Correspondence of Napoleon Bonaparte.
Fondation Napoléon, Napoleon's Russian campaign: the retreat.
Fondation Napoléon, From Waterloo to the island of St Helena.
Fondation Napoléon, The French Civil Code or Code Civil, 21 March 1804: an overview.
Fondation Napoléon, "Napoleon, the dark side": Napoleon's re-establishment of slavery.

Salahuddin Ayyubi: Righteousness of the Faith

Hacks Vitae — Sat, 21 Oct 2023 04:33:45 GMT

Salahuddin Ayyubi

Salahuddin Ayyubi, known in Europe as Saladin, was the sultan of Kurdish descent who united Egypt and Syria, founded the Ayyubid dynasty, defeated the Crusaders at Hattin in 1187 and took Jerusalem the same year. "Salah ad-Din" means "righteousness of the faith". It was a title of honour. His own name was Yusuf ibn Ayyub.

He was born in 1137 or 1138 and died in Damascus on 4 March 1193. We know him well because Baha ad-Din ibn Shaddad, the judge of his army who went with him on his later campaigns, wrote his life. Where it matters, this page sets that account beside the Frankish chronicle attributed to Ernoul, a squire of Balian of Ibelin, the man who negotiated Jerusalem's surrender on the other side.

Meaning of Salahuddin Ayyubi

Salah

Ṣalāḥ (صلاح) is Arabic for "righteousness".

Uddin

Al-Dīn (الدين) means "the faith" or "the religion". "Uddin", "ad-Din" and "al-Din" are different English spellings of the same words. Together, Salah ad-Din means "righteousness of the faith".

Titles built this way, a word followed by al-Din, were a standard kind of honorific in classical Arabic names. Library cataloguers list them alongside examples such as Nur al-Din, the name of Saladin's own overlord. So Salah ad-Din was a title, not a given name. His personal name was Yusuf, the Arabic form of Joseph.

Ayyubi

His father was Ayyub, the Arabic form of Job, so he was Yusuf ibn Ayyub, "Yusuf son of Ayyub". Ayyubi is a nisba, the part of an Arabic name formed with the ending -i that shows where or whom someone comes from. The dynasty he founded is called the Ayyubids for the same reason.

So "Salahuddin Ayyubi" can be read as "Righteousness of the Faith, of the family of Ayyub".

Early Life and Faith

Baha ad-Din says he was born in the year 532 of the Islamic calendar (AD 1137–1138) in the citadel of Tikrit, where his father was governor. The family was of Kurdish descent. His father and his uncle, Shirkuh, served the Zengid ruler of Mosul, and later moved to Syria, where the boy began his own service under his father.

Baha ad-Din's portrait of his faith is detailed rather than general: he kept the daily prayers, travelling or not, and his charity was so constant that when he died he left, in Baha ad-Din's words, "but forty-seven Nasri dirhems, and one Tyrian gold piece" in his treasury. That is a friend's account, written after his death, and it reads like one. It is also the closest witness we have.

How he rose

His power was built long before Jerusalem.

Egypt. He went to Egypt with his uncle Shirkuh, who became vizier to the Fatimid caliph. When Shirkuh died in March 1169, Saladin took his place at the head of the government.
The end of the Fatimid caliphate. Shortly before the last Fatimid caliph, al-Adid, died in September 1171, Saladin had the Friday sermon read in the name of the Abbasid caliph in Baghdad, ending the Shiʿi Fatimid caliphate in Egypt.
Syria. His overlord Nur ad-Din died in Damascus on 15 May 1174. Saladin moved into Syria and took Damascus, and over the following years he fought Muslim rivals as well as the Crusaders.

That unification of Egypt and Syria is what made 1187 possible: the Crusader states now faced one ruler on two fronts.

Hattin, 4 July 1187

The victory that opened Jerusalem was won near Hattin, west of Tiberias. The Crusader army under King Guy of Lusignan marched to relieve Tiberias, which Saladin was besieging, across ground with little water. Baha ad-Din describes the Crusaders on the hill of Hattin "tortured by thirst" as the Muslims set fires around them. Guy was captured, and the relic of the True Cross was lost.

What happened next is told in detail by Baha ad-Din. Saladin gave the thirsty king "a bowl of sherbet made with iced rose-water". When Guy passed it to Reynald of Châtillon, the lord of Kerak who had attacked a caravan during a truce, Saladin said it was the king, not he, who was giving Reynald a drink: by custom, a captive who ate or drank with his captor was spared. Later that day Saladin struck Reynald himself, and his men finished him. He told the frightened king: "It is not the wont of kings to kill kings; but that man had transgressed all bounds."

Baha ad-Din also records that Saladin resolved to execute the captured Hospitallers and Templars, "and he spared not a single one."

Jerusalem, 1187

Saladin reached Jerusalem on 20 September 1187, and the city surrendered on 2 October 1187. There was no massacre. That is a real contrast with the Crusader capture of 1099, when, in the Crusaders' own account, the killing at the Temple was so great that "our men waded in blood up to their ankles."

But leaving the city was not free. Three early accounts, from both sides, give the same ransom, and a fourth differs on one figure:

10 dinars for a man, 5 for a woman and 1 for a child, according to Baha ad-Din; 10, 5 and 1 bezants in the chronicle attributed to Ernoul; and 10, 5 and 1 in the Latin De expugnatione Terrae Sanctae, which counts boys of seven and under as children.
The historian Ibn al-Athir, in an English translation quoted by the Muslim Heritage site, gives two dinars for a child. With three accounts from both sides agreeing on one, the evidence leans to one.
In Ernoul's account Saladin first asked 20, 10 and 5, adding that anyone who could not pay would remain a slave, and brought the price down after negotiation with Balian. Baha ad-Din puts it the other way round: everyone who paid "was to receive his freedom".

Many of the poor were freed anyway. Ernoul's chronicle says Saladin's brother, whom the Franks called Saphadin, asked for a thousand of them as slaves and freed them; the Patriarch and Balian were each given more to free; Saladin then had the poor let out through the postern of St Lazarus from sunrise to sunset, without counting, as his own alms. The same passage says that anyone found carrying the means to pay was to be imprisoned, and that the young men and women were put "between two walls" while the old were put outside the city. Balian and the Patriarch also bought the freedom of 7,000 poor people for 30,000 bezants, taken from the king of England's treasure held by the Hospitallers. When the counting was done, the same chronicle says, about eleven thousand were still unransomed, and Saladin refused an offer from the Patriarch and Balian to stay as hostages in their place. Saladin's own secretary, Imad ad-Din, counts differently: in a passage quoted by the historian Andrew Holt, he writes that the women and children alone "came to 8,000 and were quickly divided up among us". Under the terms, those who could not pay remained slaves; the Latin account's version of the terms says that those who did not have ten besants "were to become booty, to be slain by the army's swords".

Baha ad-Din adds that everyone who paid was escorted to Tyre, and that the Muslim prisoners freed in the city numbered more than three thousand. Both accounts, from opposite sides, describe the same thing: restraint far beyond a sack, together with a ransom that many people could not pay.

The Third Crusade and his death

The loss of Jerusalem brought a new crusade from Europe. Richard I of England defeated Saladin at Arsuf on 7 September 1191, a battle Baha ad-Din calls "a blow to all Moslem hearts", but Richard never took Jerusalem. The war ended with a three-year peace dated 2 September 1192: Jaffa, Caesarea, Arsuf, Haifa and Acre went to the Crusaders, Ascalon was to be demolished, and Christian pilgrims went up to Jerusalem under Saladin's escort.

Saladin died in Damascus on 4 March 1193, six months after the peace. Baha ad-Din was there. "We were obliged to borrow money to purchase everything necessary for the funeral," he writes, "even down to things that cost but a halfpenny."

Quick answers

Who was Saladin? Yusuf ibn Ayyub (born 1137–1138, died 1193), founder of the Ayyubid dynasty and sultan of Egypt and Syria. "Salah ad-Din" is a title meaning "righteousness of the faith".
How did he take Jerusalem? By destroying the Crusader field army at Hattin on 4 July 1187. Jerusalem surrendered on 2 October 1187 after a siege and a negotiation with Balian of Ibelin.
Did he spare everyone in Jerusalem? There was no massacre, but safe passage had to be bought. Thousands of the poor were freed without paying; the Frankish chronicle says about 11,000 were still unransomed at the end and, under the terms, remained slaves.
Who defeated Saladin? Richard I beat him at Arsuf in 1191, but Jerusalem stayed in Saladin's hands and the war ended in a negotiated peace in 1192.
Why is he remembered as chivalrous? Partly for real restraint in 1187, and partly because European writers took him up: Dante places him among the virtuous non-Christians in the first circle of the Inferno, "alone, apart", and Walter Scott made him a central figure of The Talisman.

Legacy of Salahuddin Ayyubi

Saladin was admired by writers on both sides of the war he fought. His biographer wrote as a devoted servant. In Europe, Dante's line places him, alone, among the virtuous non-Christians of Limbo, and Scott's The Talisman made him a romantic hero.

The sources give a fuller picture than either legend. The same biographer who records the empty treasury also records the execution of every captured Templar and Hospitaller after Hattin. The Frankish chronicle that praises his courtesy to the widows and daughters of knights also counts the thousands left unransomed. Both are part of the record.

Righteousness of The Faith

What can a reader take from his story?

Read both sides. On the ransom, Saladin's biographer and the chronicle attributed to a Frankish squire agree almost exactly, and a Latin account agrees with them; Ibn al-Athir differs only on the child's rate. When opposing witnesses agree, that is about as solid as medieval evidence gets.
Keep the numbers with the praise. "There was no massacre" and "about eleven thousand were still unransomed" are both in the sources. A story that keeps only one of them is a legend, not history.
Notice what his own side admired. Baha ad-Din opens his book not with victories but with chapters on Saladin's religion, his love of justice and his generosity. Whatever you make of the man, it is a record of what his contemporaries thought righteousness looked like.

Checked September 2026.

Sources

Baha ad-Din ibn Shaddad, The Life of Saladin (What Befell Sultan Yusuf), translated by C. W. Wilson (Palestine Pilgrims' Text Society, 1897). Internet Archive.
Chronique d'Ernoul et de Bernard le Trésorier, ed. L. de Mas Latrie (Paris, 1871). Internet Archive.
Gesta Francorum, on the capture of Jerusalem in 1099, in the Internet Medieval Sourcebook, The Siege and Capture of Jerusalem: Collected Accounts (Fordham University).
IFLA, Names of Persons: Arabic Names (2018).
De expugnatione Terrae Sanctae per Saladinum, tr. James Brundage, in the Internet Medieval Sourcebook, De expugnatione Terrae Sanctae, 1187 (Saladin takes Jerusalem) (Fordham University).
Ibn al-Athir, al-Kamil, as quoted in English in Muslim Heritage, Jerusalem: thriving trade, scholars and magnificent architectural works.
Andrew Holt, Saladin's legacy: some thoughts (25 July 2017), quoting Imad ad-Din.
Behind the Name: Salah ad-Din, Yusuf, Ayyub.
Dante, Inferno, canto IV, line 129, in La Divina Commedia (Project Gutenberg).
Walter Scott, The Talisman (Project Gutenberg).

Battle of Kerak: A Historic Saga of Valor and Strategy

Hacks Vitae — Fri, 20 Oct 2023 05:22:20 GMT

The short answer

The "Battle of Kerak" was two sieges of a Crusader castle, and Saladin withdrew from both:

1183: Saladin was besieging Kerak in November. A relief army led out by King Baldwin IV reached the castle in early December, and Saladin withdrew.
1184: he came back in the summer, had the castle surrounded by mid-August, and again drew off when a Frankish relief army arrived in early September.

Kerak passed into Saladin's hands only in late 1188, by surrender, after the Frankish defeat at Hattin. There was no battle at Kerak in 1177: that year's battle between Baldwin IV and Saladin was Montgisard, on 25 November 1177, which William of Tyre, the kingdom's chancellor and historian, records as a Christian victory.

The details below come from three writers of the time: William of Tyre, writing inside the kingdom; the Old French chronicle that goes under the name of Ernoul; and Baha ad-Din Ibn Shaddad, who served Saladin as judge of the army. Where they disagree, this page says so.

The castle

William of Tyre says the site was "formerly called Petra of the Desert". He describes Kerak as a citadel built during the reign of King Fulk by a lord called Paganus the Butler, on a very high mountain "surrounded by deep valleys", and later strengthened by his successors with "a moat and towers". Below the citadel lay a village. At only two points, William says, could anyone reach the top of the mountain.

Kerak mattered because of the roads it watched. Baha ad-Din says the castle "so effectually commanded the road to Egypt that caravans could not travel without a strong military escort", and the notes to his translation add that Reynald's castles commanded the pilgrim road to Mecca. Today the town of Karak, with its Crusader castle, is in Jordan.

Key Players

Reynald of Châtillon

Reynald held Kerak as lord of Oultrejordain, the land beyond the Jordan. Ernoul says he was given "the lady of Kerak and Montreal" as his wife; her name was Stephanie. The notes to the 1897 English translation of Baha ad-Din call him "a bitter enemy of the Moslems" and record his march toward Medina in 1182–83. William of Tyre calls his decision to defend the village below the castle in 1183 rash, and William's modern translators note that his information probably came from nobles opposed to Reynald.

Salahuddin Ayyubi (Saladin)

Saladin, Salah ad-Din Yusuf, son of Ayyub, was of Kurdish descent. His biographer Baha ad-Din entered his service, became judge of the army and went with him on his later campaigns. He dates both campaigns against Kerak in the Islamic calendar.

Baldwin IV and the wedding couple

Baldwin IV was the leper king of Jerusalem. In 1183, William of Tyre writes, his leprosy grew worse: "His sight failed and his extremities became completely deadened". The couple whose wedding was under way in Kerak when the siege began were Stephanie's son, Humphrey of Toron, and Baldwin's younger half-sister, Isabella, daughter of King Amalric and Queen Maria.

The Battle Unfolds

The first siege, 1183

The start. Saladin appeared before Kerak, William says, when the wedding celebrations were barely over, "in fact, on that very day, it is said". Reynald tried to hold the village outside the citadel. His men were driven back, and William credits a single knight, Iven, with stopping the attackers at the bridge to the citadel gate. The defenders then threw the bridge down themselves. The citadel was crowded with townspeople and with "actors and performers on the flute and psaltery" who had come for the wedding.

The bombardment. Saladin set up eight siege engines, six on the site of the old town and two outside, and kept up the attack "by night as well as by day".

The wedding courtesy. Ernoul tells the story this siege is remembered for. The groom's mother, the wife of Prince Reynald, sent Saladin bread, wine, oxen and sheep from her son's wedding. Saladin thanked her warmly, asked which tower the bride and groom were in, and had it cried through his army that no one should shoot at or attack that tower. Ernoul adds that her message reminded Saladin that he had often carried her in his arms when he was a slave in the castle and she was a child. Neither William of Tyre nor Baha ad-Din mentions the gift, the spared tower or that claim.

The relief. Baldwin IV summoned the kingdom's army, "took the Life-giving Cross and marched thither himself". Neither William nor Ernoul says how he travelled; the historian Bernard Hamilton judges that he was "almost certainly too ill to ride, and would have had to be carried in a litter". At the Dead Sea he handed command of the whole army to Raymond, count of Tripoli. Hearing that the Christian army was near, Saladin abandoned his engines and withdrew. Baha ad-Din's dates put the Franks' arrival at Kerak on 16 Sha'ban 579, which works out to about 4 December 1183, and Saladin back in Damascus on 24 Sha'ban, about 12 December.

How long it lasted depends on the source. William says Saladin kept up the siege "for an entire month". Ernoul's text says eight weeks; some manuscripts say six.

The second siege, 1184

Baha ad-Din gives the second siege in detail. Saladin halted near Kerak for several days to wait for his nephew, who joined him on 19 Rabi II (30 July 1184). He gathered troops from Egypt, Syria and the Jazira, and by 4 Jumada I (13 August 1184) Kerak "was completely invested", with mangonels battering it. A Frankish army came out to relieve the castle and camped at a place he calls el-Waleh. On 26 Jumada I (4 September 1184) the Franks moved their camp nearer Kerak. Saladin then sent his army instead into the lands toward the coast, which had been left unprotected: his troops stormed and pillaged the town of Nablus, though not its castle, and took Jenin, before Saladin returned to Damascus. Kerak held again.

How Kerak finally fell

In 1187 Saladin defeated the Frankish army at Hattin. Baha ad-Din lists Reynald, "Lord of el-Kerak", and "the son of el-Honferi (Honfroi de Toron)" among the leaders taken prisoner. Kerak itself held out into the next year. Baha ad-Din says that during the month of Ramadan in the year 584 (late October to late November 1188) Saladin "obtained possession of el-Kerak, the officers in command surrendering the fortress in order that their lord might be set at liberty". Ernoul, for his part, says Saladin sent Humphrey, freed, to his mother, the wife of Prince Reynald.

Legacy of the Battle

Kerak is one of the places where Saladin's campaigns did not succeed by force. A castle on a mountain, and a relief army on its way, twice made him give up. He took it in the end by surrender, after his victory at Hattin.

How to read a medieval battle story

The sieges of Kerak also show how to read any medieval battle story:

Ask whose account it is. William of Tyre was the kingdom's chancellor and drew on nobles hostile to Reynald; Baha ad-Din served Saladin; Ernoul tells the best stories, but his account of the 1183 siege runs straight on to Saladin going to Nablus, staying two days and leaving the town unharmed apart from the food his army took; Baha ad-Din has Saladin's troops storming and pillaging Nablus in the second campaign, in 1184. Each is useful, and none is neutral.
Check the calendar. Baha ad-Din dates events by the Islamic calendar. A Christian-calendar date for them, like 4 December 1183, is a conversion to the Julian calendar used at the time, not something he wrote.
Expect the numbers to differ. One source says the first siege lasted a month; another says eight weeks, or six. Where the sources disagree, a range is the fairest answer.
Treat the best anecdote with care. Of these three writers, only Ernoul tells the wedding courtesy. That does not make it false, but it is one witness, not three.

Sources

William of Tyre, A History of Deeds Done Beyond the Sea, trans. Emily Atwater Babcock and A. C. Krey (Columbia University Press, 1943), with the translators' introduction and notes; Book 21, ch. 22–23 (Montgisard) and Book 22, ch. 25 and 28–30 (Kerak, 1183): https://archive.org/details/williamoftyrehistory
Chronique d'Ernoul et de Bernard le Trésorier, ed. Louis de Mas Latrie (Société de l'histoire de France, 1871), pp. 30–31, 102–106 and 253; translations from the Old French are ours: https://archive.org/details/chroniquedernoul00ernouoft
Baha ad-Din Ibn Shaddad, The Life of Saladin, trans. C. R. Conder (Palestine Pilgrims' Text Society, 1897), ch. 25, 28, 35 and 47, with the translator's notes: https://archive.org/details/lifesaladin00condgoog
Bernard Hamilton, The Leper King and His Heirs (Cambridge University Press, 2000), p. 196.
Jordan Tourism Board, Karak: https://international.visitjordan.com/wheretogo/karak/

Checked September 2026.

Different Crusader Knights and Crusader Military Orders: A Comprehensive Overview

Hacks Vitae — Sat, 23 Sep 2023 19:53:44 GMT

Different Crusader Knights and Crusader Military Orders

The crusades are usually dated from Pope Urban II's call at the Council of Clermont in 1095 to the capture of Acre in 1291, which ended the kingdom of Jerusalem. Out of them came a new kind of institution, the military order: men who took religious vows like monks and fought like knights.

This page covers six of them: the Templars, the Hospitallers and the Teutonic Knights, then the Knights of the Holy Sepulchre, the Order of Saint Lazarus and the Order of Calatrava. After that, what knights did on crusade, and what they did at home when there was no war. The sources are listed at the end.

The Crusader Knights: Noble Warriors of the Holy Land

Pope Urban II's appeal, in the version recorded by Robert the Monk, asked each man who made the vow to "wear the sign of the cross of the Lord on his forehead or on his breast"; those who joined were said to have "taken the cross". The three main military orders of the Holy Land each began with a different job.

Knights Templar

Founded to guard pilgrims. William of Tyre, writing about fifty years later, says a group of knights "bound themselves to Christ's service in the hands of the Lord Patriarch", and that their first duty was "protecting the roads and routes against the attacks of robbers and brigands", "especially in order to safeguard pilgrims". The Catholic Encyclopedia names their leader as Hugues de Payens, "a knight of Champagne", with eight companions. William gives no year; the Fordham translation dates his "same year" to 1118. The 1911 Encyclopaedia Britannica gives 1119, and modern reference works give about 1119 or about 1120.
Named after their headquarters. King Baldwin II gave them part of his palace in Jerusalem. Because it stood "next to the Temple of the Lord", William explains, "they are called the Brothers of the Militia of the Temple".
White mantle, red cross. A church council at Troyes, in France, gave them a rule and a white habit. The red cross came later: William says the knights began to add "crosses made of red cloth to their mantles" in the time of Pope Eugenius III.
Soldiers who also kept money. Their houses in Paris and London held deposits: the Catholic Encyclopedia notes that "numerous princes and private individuals had banked there" and that "in Paris the royal treasure was kept in the Temple".
The end of the order. The Templars in France were arrested on 13 October 1307 on the orders of King Philip IV. Pope Clement V dissolved the order, without condemning it, by a bull of 22 March 1312, and its property went to the Hospitallers, except in the Iberian kingdoms; in Portugal it passed to the new Order of Christ in 1319. Jacques de Molay, the last grand master, withdrew his confession in public in March 1314 and was burned in Paris with another leader of the order.

Knights Hospitaller

A hospital first. The order grew out of a hospice for pilgrims in Jerusalem. The order's own history traces it to merchants from Amalfi, who built a hospital there in the second half of the eleventh century; the Catholic Encyclopedia rejects that link. Both name its founder as a man called Gerard, and a bull of Pope Paschal II dated 1113 is addressed to him as founder and head of the Jerusalem hospice.
Soldiers later. Under his successor Raymond, the Catholic Encyclopedia says, the hospital paid for "an armed escort" for pilgrims, "which in time became a veritable army, comprising knights".
White cross on black. The Hospitallers "had the black mantle with a white cross", and in war "a red surcoat with the white cross". The Templars, by contrast, wore a white habit with a red cross.
A network across Europe. The order's estates were grouped into "tongues, priories, and commanderies"; a commander collected the revenues of his lands and sent a fixed share to the houses in the Holy Land.
Rhodes, then Malta. The knights ruled Rhodes from 1309 to 1522, when they surrendered it after a six-month siege by the Ottoman sultan Suleiman, and Malta from 1530, when Emperor Charles V granted it to them, until they surrendered it to Napoleon Bonaparte on 12 June 1798. On both islands they kept a fleet.
Today. The order continues as the Sovereign Order of Malta, which describes itself as having "medical, social and humanitarian projects in most countries in the world".

Teutonic Knights

A field hospital at Acre. During the Third Crusade, in 1190, merchants from Bremen and Lübeck set up a hospital outside the besieged city of Acre; the Catholic Encyclopedia describes "a large tent, constructed from the sails of their ships". The brothers cared for sick pilgrims and wounded crusaders.
Knights by the end of the decade. In March 1198 the brothers were raised to an order of knights, and in 1199 Pope Innocent III gave them the Templars' rule for their fighting members. In 1205 he granted them "the white habit with a black cross".
From the Holy Land to the Baltic. From 1229 the order fought the pagan Prussians and built a state of its own on the Baltic coast. Its grand master moved from Venice to Marienburg, now Malbork in Poland, in 1309.
Malbork Castle. UNESCO, which listed it as a World Heritage Site in 1997, calls it "the most complete and elaborate example of a Gothic brick-built castle complex in the characteristic and unique style of the Teutonic Order".

Crusader Military Orders: Guardians of the Holy Land

Other orders grew up beside the big three, in the Holy Land and in Spain. Three are covered here, and each has a different story about its own origins.

Knights of the Holy Sepulchre

Where it began is uncertain. The Latin Patriarchate of Jerusalem traces the order to the canons of the Church of the Holy Sepulchre set up after the First Crusade, with armed knights among its members. The order's own history is more careful: it calls its origins "somewhat obscure", says the link to the First Crusade rests on "an undocumented tradition", and dates the first documentary evidence of knights "of the Holy Sepulchre" to 1336.
Restored in the nineteenth century. Pope Pius IX restored the Latin Patriarchate of Jerusalem in 1847 and issued apostolic letters restoring the order in 1868.
Today. It is the Equestrian Order of the Holy Sepulchre of Jerusalem, and the Patriarchate describes its members' contributions as its institutions' "main source of funding".

Order of Saint Lazarus

A leper hospital. The order "originated in a leper hospital founded in the twelfth century by the crusaders of the Latin Kingdom", in the words of the Catholic Encyclopedia, and it began purely as an order of hospitallers.
An older origin story that does not hold up. The knights later claimed to descend from earlier leper hospitals in the East, "to pass as the oldest of all orders"; the same encyclopedia calls that claim "apocryphal".
What became of it. Once the Jerusalem hospital was lost, the order's European houses stopped caring for the sick. Pope Innocent VIII suppressed it in 1490 and gave its property to the Hospitallers, a transfer that in practice took effect only in Germany.

Order of Calatrava

Founded in Castile in 1158. King Sancho III gave the frontier castle of Calatrava, which the Templars had given up, to Raymond, abbot of the Cistercian monastery of Fitero. His lay brothers became its garrison, and the order took the castle's name.
Monks under arms. The Cistercians and Pope Alexander III approved the arrangement in 1164. The knights slept in their armour and wore "the Cistercian white mantle with the scarlet cross fleurdelisée", a red cross whose arms end in lilies.
Lost and won back. After the defeat at Alarcos in 1195 the knights lost Calatrava; it was retaken in 1212, the same year as the Christian victory at Las Navas de Tolosa.

What did knights do on crusade?

Between the famous battles, the daily work looks different from the legend.

They took a vow, and it counted as a pilgrimage. In Fulcher of Chartres's version of Urban II's speech, the Pope promises that "all who die by the way" will "have immediate remission of sins". Robert the Monk's version calls the journey a "holy pilgrimage".
They largely paid their own way. The historian Andrew Holt, summarising Jonathan Riley-Smith's study of the First Crusaders' charters, writes that the crusader and his family usually met the costs, and "often, as the charters show, they had to sell property or take out mortgages to do so".
They held castles for years. The Hospitallers built Krak des Chevaliers, in what is now Syria, from 1142 to 1271, and UNESCO, noting further building by the Mamluks in the late thirteenth century, ranks it "among the best-preserved examples of the crusader castles".

The code of chivalry in practice

Chivalry was a real ideal and an unevenly kept one. When Jerusalem fell in 1099, the anonymous Gesta Francorum describes the crusaders "killing and slaying even to the Temple of Solomon, where the slaughter was so great that our men waded in blood up to their ankles". And the Fourth Crusade, aimed at Egypt, was diverted against the Christian city of Zara and then Constantinople, which it sacked in 1204; the Byzantine historian Nicetas Choniates describes the altar of Hagia Sophia "broken into bits and distributed among the soldiers". Both events are part of the record, and an account of the knights includes them.

What knights did in peacetime

What a knight was

"Knight" was a rank, and a costly one. In England, the 1911 Encyclopaedia Britannica notes, it became necessary in the early thirteenth century "to compel landowners to take up the knighthood"; more generally, it adds that "knighthood was avoided by poor nobles". So a knight in peacetime was, above all, a landholder, and one who owed what the 1911 Britannica calls "personal service in war", though from the twelfth century that duty could be commuted for a payment known as scutage.

The everyday work: land, rents and courts

A knight's estate brought him rents and services from his tenants, and a manor court, which in the thirteenth century dealt with criminal, civil and manorial cases as one court. In England, men of his class also served the crown in the counties. The first Statute of Westminster ordered that "none but lawful and discreet knights should be chosen as coroners", and Simon de Montfort's parliament of 1265, as the History of Parliament Trust describes it, included "two knights elected in every county". Knights had come to parliament before; 1265 was the first time such men were summoned with no tax to consent to.

Tournaments

Tournaments were dangerous. In 1139 the Second Lateran Council banned "those abominable jousts and tournaments", which "often result in human deaths and danger to souls", and denied church burial to a knight who died in one. The ban did not stop them: the 1911 Encyclopaedia Britannica notes that "many a tournament had its tale of killed and wounded in the chronicle books".

Castles and fortifications

Some great castles had professional designers. Edward I's castles in North Wales, for example, were "designed and directed by James of St George", whom UNESCO calls the king's "chief architect".

Courts, poetry and patronage

Troubadour poetry grew up at noble courts, and some troubadours were knights or lords themselves: Arnaut Daniel was "a knight of Riberac", Bertran de Born was besieged in his own castle of Hautefort by Richard the Lionheart, and the earliest poet of the school, Guilhem, Count of Poitiers, "reigned from 1087 to 1127".

Who chivalry protected

The chivalric code bound together what the 1911 Encyclopaedia Britannica calls "the whole warrior caste of Christendom". It did much less for the people who worked the land. The same encyclopedia notes that in thirteenth-century England "more than half the population were serfs", and that "disputes between a serf and his lord were decided in the latter's court", although "the king's courts attempted to protect the serf's life and limb and necessary implements of work". Chivalry is real, and it is narrower than the word now suggests.

Reading about the orders: a short checklist

Popular accounts of the military orders mix record, legend and later reinvention. A few questions help sort them out:

Whose date is it? A founding date can come from a chronicler writing decades later, a modern encyclopedia or the order's own history, and they do not always agree: the Templars' start is 1118 in the Fordham translation of William of Tyre, 1119 in the 1911 Britannica, and about 1120 in some modern reference works.
Is it the order's own story? Orders that still exist tell their history in their own way. The Holy Sepulchre's own history is more cautious about its origins than some other accounts.
Which cross? A red cross on white is the Templars, a white cross on black the Hospitallers, a black cross on white the Teutonic Knights.
Which century? The Hospitallers of 1113, of Rhodes and of today are one order with very different jobs.

Sources

William of Tyre, "The Foundation of the Order of Knights Templar", trans. in the Internet Medieval Sourcebook, Fordham University.
Urban II, speech at the Council of Clermont, 1095: five versions (Fulcher of Chartres, Robert the Monk and others), Internet Medieval Sourcebook, Fordham University.
"The Siege and Capture of Jerusalem: Collected Accounts" (Gesta Francorum and others), Internet Medieval Sourcebook, Fordham University.
Nicetas Choniates, "The Sack of Constantinople (1204)", Internet Medieval Sourcebook, Fordham University.
Second Lateran Council (1139), canons, Papal Encyclicals Online.
Catholic Encyclopedia (1907–1913): "Knights Templars", "Hospitallers of St. John of Jerusalem", "Teutonic Order", "Order of St. Lazarus of Jerusalem", "Military Order of Calatrava" (New Advent).
Encyclopaedia Britannica (1911): "Templars", "Teutonic Order", "Crusades", "Knighthood and Chivalry", "Coroner", "Manor", "Tournament", "Troubadour" (Wikisource).
Historisches Lexikon Bayerns, "Deutscher Orden: Politische Geschichte (Mittelalter)", Dieter J. Weiß.
Sovereign Order of Malta, "History" and home page.
Equestrian Order of the Holy Sepulchre of Jerusalem, "History"; Latin Patriarchate of Jerusalem, "Order of the Holy Sepulchre".
UNESCO World Heritage Centre: "Castle of the Teutonic Order in Malbork"; "Crac des Chevaliers and Qal'at Salah El-Din"; "Castles and Town Walls of King Edward in Gwynedd".
Andrew Holt, "Jonathan Riley-Smith on the Motivations of the First Crusaders" (2016).
Mark Cartwright, "Knights Templar", World History Encyclopedia (2018); EBSCO Research Starters, "Order of the Knights Templar".
Sovereign Order of Malta, "1113 – The bull 'Pie postulatio voluntatis'".
History of Parliament Trust, "Simon de Montfort's 1265 Parliament and Magna Carta" (2015).
Almagro Noticias, on the 1158 charter of Sancho III held in Spain's Archivo Histórico Nacional (September 2026).

Checked September 2026. Dates and descriptions come from the sources above; where they differ (the Templars' founding year, the origins of the Hospitallers and of the Holy Sepulchre order), each version is given.

A Motivational Poem About Resilience and Success: "INVICTUS" by William Ernest Henley

Hacks Vitae — Sat, 26 Aug 2023 05:12:03 GMT

A Motivational Poem About Resilience: What is "Invictus" All About?

"Invictus" is a sixteen-line poem by the English poet William Ernest Henley (1849–1903). The Latin word invictus means "unconquered" or "unsubdued". The speaker is in the dark, knocked about by chance, facing whatever waits after death, and says he is still unbowed. It ends with the two lines most people know:

I am the master of my fate: I am the captain of my soul.

Two things change how the poem reads: Henley wrote it out of a long stay in hospital, and he never gave it the title it is known by. Both are below, after the poem itself.

The poem in full

Henley dated the poem 1875. This is the text as printed in his A Book of Verses; the poem has long been out of copyright.

Out of the night that covers me, Black as the pit from pole to pole, I thank whatever gods may be For my unconquerable soul.

In the fell clutch of circumstance I have not winced nor cried aloud. Under the bludgeonings of chance My head is bloody, but unbowed.

Beyond this place of wrath and tears Looms but the Horror of the shade, And yet the menace of the years Finds and shall find me unafraid.

It matters not how strait the gate, How charged with punishments the scroll, I am the master of my fate: I am the captain of my soul.

The man who wrote it, and why

Henley was born in Gloucester. From the age of twelve he had tuberculosis of the bone, and his left leg was amputated below the knee a few years later. Accounts of the exact year differ; one medical history says he was sixteen, and that the operation was at St Bartholomew's Hospital in London.

In 1873 surgeons said his right foot should be amputated too. He would not accept that and looked for help elsewhere. He went to Edinburgh, to the surgeon Joseph Lister, whose antiseptic method of surgery was then new, and Lister operated and saved the foot. Henley stayed almost two years in the Royal Infirmary of Edinburgh. The sequence of poems he wrote there, In Hospital, is signed with the place and dates "The Old Infirmary, Edinburgh, 1873–75".

"Invictus" is dated 1875 in his collections, and the Poetry Foundation says he began writing poems, including this one, as he healed in the infirmary. "My head is bloody, but unbowed" reads differently once you know it came from a young man on a surgical ward who did not know whether he would keep his foot.

In January 1875 the critic Leslie Stephen visited him in hospital and brought along a young Scottish writer, Robert Louis Stevenson, who became his closest friend for the next decade. In a letter of May 1883, while waiting for the proofs of Treasure Island, Stevenson wrote to him: "It was the sight of your maimed strength and masterfulness that begot John Silver in Treasure Island." He added that Silver was not like Henley "in any other quality or feature".

How it got its title

Henley printed the poem without a title.

In A Book of Verses, his first collection (first edition printed in 1888), it is number IV in a section called "Life and Death (Echoes)", with no title. The printing we read carries only a dedication, "To R. T. H. B.", for R. T. Hamilton Bruce.
The 1907 edition of his collected Poems, published after his death, still gives it no title; the dedication has become "I. M. R. T. Hamilton Bruce (1846–1899)", in memory of Bruce.
It appears as "Invictus", poem number 842, in Arthur Quiller-Couch's anthology The Oxford Book of English Verse, first published in 1900.

None of Henley's own collections that we read call it "Invictus". The title most readers know is the anthology's, not one that Henley's own books give.

Who has carried it

The poem has travelled a long way from that ward.

Winston Churchill, in the House of Commons on 9 September 1941, borrowed its closing lines: "We still are master of our fate. We still are captain of our souls."
Nelson Mandela is reported to have memorised it on Robben Island and recited it to fellow prisoners. At his memorial on 10 December 2013, Barack Obama quoted the last stanza as "the words that brought him comfort within the four walls of his cell".
The 2009 film Invictus, directed by Clint Eastwood, is set around the 1995 Rugby World Cup; in it, Mandela (played by Morgan Freeman) tells the Springbok captain that the poem inspired him in prison.
The Invictus Games, an international sporting event for wounded, injured and sick service personnel created by Prince Harry, were first held in London from 10 to 14 September 2014. The foundation behind them says they celebrate "the unconquered human spirit".

And one that is not inspiring. In 2001 the man executed for the 1995 Oklahoma City bombing left a handwritten copy of the poem as his final statement. The poem is about endurance, not about whether a cause is right, and anyone can pick it up. That is worth knowing about a text so often handed to people as a creed.

What the poem says, stanza by stanza

A plain reading, stanza by stanza:

The night. The speaker is in total darkness, "black as the pit", and thanks "whatever gods may be" — he does not name a god — for the one thing he still has: an "unconquerable soul".
The blows. "The fell clutch of circumstance" and "the bludgeonings of chance" hurt him. He does not pretend otherwise: his head is "bloody". His claim is only that he has not winced or cried out, and is "unbowed".
What comes next. Beyond "this place of wrath and tears" there is only "the Horror of the shade", which reads as death. The years threaten him, and they find him "unafraid".
The judgement. "Strait the gate" means a narrow gate, and it echoes the King James Bible, Matthew 7:14: "strait is the gate, and narrow is the way". However narrow the gate and however long the list of punishments on the "scroll", he says, "I am the master of my fate".

One way to read the ending, given where it was written: the speaker cannot control what is being done to him, so he claims the one thing left to him, how he meets it.

What you can do with it

Read it with its date and place. The poem is a young man's answer to repeated surgery, not a general promise that everything can be controlled. Read that way, "master of my fate" is about how you face what happens, not a claim that nothing can happen to you.
Keep the last stanza for bad days. It is four lines, easy to learn by heart, which is how Mandela is reported to have carried it.
Keep endurance and judgement apart. The poem tells you how to hold yourself, not whether what you are doing is right. Its use in 2001 shows why that matters.
Mind the spelling. It is "strait" (narrow), not "straight".

Sources

William Ernest Henley, A Book of Verses, second edition or later (New York: Scribner & Welford; first edition printed 1888), "Life and Death (Echoes)" IV. Internet Archive text
William Ernest Henley, Poems (David Nutt, 1907). Project Gutenberg #1568
Arthur Quiller-Couch (ed.), The Oxford Book of English Verse, 1250–1900 (first edition 1900), no. 842. Project Gutenberg #66619
Lewis & Short, A Latin Dictionary, "invictus". Perseus
Poetry Foundation, "William Ernest Henley". poetryfoundation.org
Andrzej Diniejko, "William Ernest Henley: A Biographical Sketch", The Victorian Web. victorianweb.org
George Venters, "The Royal Infirmary of Edinburgh and the legacy of Long John Silver", Hektoen International (2020). hekint.org
The Letters of Robert Louis Stevenson, vol. 1, to W. E. Henley, Hyères, May 1883. Project Gutenberg #622
Hansard, House of Commons, "War Situation", 9 September 1941. api.parliament.uk
The White House, "Remarks by President Obama at Memorial Service for Former South African President Nelson Mandela", 10 December 2013. obamawhitehouse.archives.gov
J. Dana Stuster, "Mandela's Favorite Poem Is Also Beloved by White Supremacists", Foreign Policy, 7 December 2013. foreignpolicy.com
African Studies Centre Leiden, "Invictus, a film by Clint Eastwood (2009)". ascleiden.nl
Invictus Games Foundation, "Aims and Origins" and "London 2014". invictusgamesfoundation.org
The Bible, King James Version, Matthew 7:14. Project Gutenberg #10

Checked September 2026. If you can show any of this wrong, with a source, write to hacksvitae@gmail.com.

The French Revolution's Lessons in Greed and Corruption: A Reflection on Today's World

Hacks Vitae — Thu, 18 May 2023 02:20:35 GMT

The French Revolution

The French Revolution began in 1789 and reshaped France within a few years: the abolition of feudal privileges, a declaration of rights, the end of the monarchy, and then the Terror. It had several causes, long-term and short-term, and historians still disagree about how to weigh them. Privilege, unfair taxation and a loss of trust in the people running the country were part of it, and those are the parts this page is about. They are also the parts that travel best to other times and places.

Unearthing the roots of greed and corruption

Old-regime France was divided into three estates: the clergy, the nobility and the commoners, known as the Third Estate. The first two enjoyed a variety of tax exemptions, so the tax burden fell mainly on the lower classes. The system was also opaque: tax rates varied widely from region to region, were often different from the official amounts, and were collected inconsistently. That complexity, and the lack of accountability, caused resentment among all taxpayers.

The Church was wealthy. It owned nearly 10% of the land in France and collected tithes, in effect a 10% tax on income paid by peasant farmers in crops, though many of the parish priests elected to the Estates General earned less than unskilled labourers. Over 25% of French farmland was subject to feudal dues, which provided the nobility with most of their income.

Money was the trigger. France had borrowed heavily to fight Britain in the war of 1778 to 1783, kept borrowing afterwards, and by 1788 half of state revenue went on servicing its debt. The debt itself was not high compared with Britain's; the problem was a tax system that could not be reformed, because the regional courts called parlements could block new tax laws. Bad harvests in 1787 and 1788 pushed up food prices and unemployment.

Scandal did the rest. In the Affair of the Diamond Necklace, which broke in August 1785, Queen Marie Antoinette was accused of taking part in a fraud against the Paris jewellers Boehmer and Bassenge. The accusation was false, but it still damaged her reputation, and scandals like it fuelled widespread anger at the court, the nobility and church officials.

Why historians disagree. Historians still argue about the causes. From the 1930s, social and economic explanations led French studies of the Revolution: Georges Lefebvre developed a Marxist analysis, and Albert Soboul wrote in the same tradition. Alfred Cobban challenged that reading, arguing that the Revolution was primarily a political conflict, and François Furet and Denis Richet also argued for the primacy of political decisions. Since the 1990s Western scholars have largely set aside the Marxist class-struggle reading, but no new explanation has gained widespread support. Greed and corruption are one thread, not the whole cloth.

The dangers of unchecked greed and corruption

What the old regime shows is how a system loses the benefit of the doubt. When taxes fall hardest on those least able to pay, when the rules are too complex to check, and when the people at the top are the subject of scandal, a false story can be believed as easily as a true one. The Diamond Necklace affair is the clearest case: the accusation against the queen was false, and it hurt her anyway.

The Revolution did not end corruption either. The government that followed the Terror, the Directory (1795 to 1799), had its own: corruption was a serious problem, particularly among the businessmen who supplied the army and the government, and charges of corruption surrounded the director Paul Barras and his allies.

The imperative of accountability and transparency

The revolutionaries wrote accountability into their founding document. The Declaration of the Rights of Man and of the Citizen, approved by the National Assembly and published on 26 August 1789, says in Article 14 that "All the citizens have a right to decide, either personally or by their representatives, as to the necessity of the public contribution; to grant this freely; to know to what uses it is put; and to fix the proportion, the mode of assessment and of collection and the duration of the taxes." Article 15 is shorter: "Society has the right to require of every public agent an account of his administration."

In the same summer the Assembly passed the August decrees, which abolished feudalism: feudal dues and church tithes were cancelled, and other decrees brought in equality before the law and opened public office to all.

Comparisons in today's world

This section is our reading, not a finding. The pattern in the sources above is not specific to eighteenth-century France: rules too complicated for the people paying to check, privileges that some enjoy and others fund, and a public that has stopped extending the benefit of the doubt. Where those three meet, trust wears thin, and a rumour can do as much damage as a fact. The eighteenth-century tools for this were pamphlets and newspapers; ours are faster.

Who was who, and what the Terror was

The other half of the story is what happens when a revolution starts turning on its own.

The people at the centre of it

Maximilien Robespierre: a lawyer and radical Jacobin, a member of the Committee of Public Safety from July 1793. In a speech on 5 February 1794 he said: "Terror is nothing other than justice, prompt, severe, inflexible; it is therefore an emanation of virtue; it is not so much a special principle as it is a consequence of the general principle of democracy applied to our country's most urgent needs." He was executed on 28 July 1794. Historians still argue about him. One view, found in the work of Marisa Linton and Pierre Serna, is that the other members of the Committee, with those of the Committee of General Security, were as much responsible for running the Terror as he was, and may have exaggerated his role to play down their own, using him as a scapegoat after his death. The opposite view, long the common one and still held by some historians, sees him as demagogic, dictatorial and fanatical; the historian Albert Mathiez argued against it in the 1920s.
Jean-Jacques Rousseau: the philosopher whose political thought influenced aspects of the Revolution. He died on 2 July 1778, eleven years before it began.
Lafayette: a French officer who served as an unpaid volunteer in the American army during the War of Independence, then on 11 July 1789 presented a declaration of rights to the National Assembly and was made commander of the new National Guard of Paris. In August 1792 the Assembly declared him a traitor; he fled to neutral territory, was taken prisoner, and was held for five years in Prussian and then Austrian prisons until his release in 1797.
Madame Roland: a political hostess close to the Girondins, a faction less extreme than the Mountain, whose salon became a meeting place for leaders of the popular movement, among them Brissot and Robespierre. She was guillotined on 8 November 1793.
Paul Barras: sided with the men who overthrew Robespierre's faction in 1794, then became one of the five directors who ran the Directory.
Napoleon Bonaparte: trained at the École Militaire as an artillery officer, rose rapidly after the siege of Toulon in 1793, and overthrew the Directory in the coup of 18 Brumaire on 9 November 1799.

The Reign of Terror (1793 to 1794). Encyclopaedia Britannica sums it up: "During the Reign of Terror, at least 300,000 suspects were arrested; 17,000 were officially executed, and perhaps 10,000 died in prison or without trial." Elsewhere on the same page it puts the deaths without trial or in prison at as many as 23,000. Another count gives 16,594 official death sentences from June 1793 to the fall of Robespierre, 2,639 of them in Paris, and estimates ten to twenty thousand more people executed without trial or dead in prison awaiting it. Totals cited to Greer and to the historian Jean-Clément Martin put the deaths at 35,000 to 45,000 at least.

Most of the dead were not aristocrats. The historian Donald Greer estimated that 8.5% of the Terror's victims were nobles, 6.5% clergy and 85% from the Third Estate, the commoners.

The Terror also consumed its own makers. The radical Hébert was executed on 24 March 1794 with 19 of his colleagues, and Danton on 5 April. Fabre d'Églantine, a member of the Jacobin club and Danton's private secretary, was arrested on 12 January 1794 on a charge of malversation and forgery connected with the affairs of the French East India Company, and guillotined on 5 April, the same day as Danton. Robespierre followed in July. Historians differ on why: François Furet argued that the revolutionaries' ideological commitment required the extermination of any opposition, while a middle position holds that the violence was not inevitable but the product of a series of complex internal events, made worse by war.

Pamphlets and newspapers

Before 1789 France had a small number of heavily censored newspapers that needed a royal licence. From 25 September 1788, the public debate produced an average of 25 new political pamphlets a week; one of the most influential, the abbé Sieyès's What Is the Third Estate?, denounced the privileges of the clergy and nobility. Then came the newspapers, among them Marat's L'Ami du peuple, first published as Le Publiciste parisien on 12 September 1789. Estimates of how many were founded over the next decade differ: more than 2,000 in one, over 1,300 in another. Newspapers were read aloud in taverns and clubs and passed from hand to hand.

Life hacks and lessons for today's world

Some ways to use this, as a reader rather than a revolutionary:

Ask the Article 15 question. Of any organisation that takes your money or makes rules for you, ask whether it gives an account of what it did with them. You can ask it of a club, a company or a council.
Treat complexity as something to look into. In old-regime France, a tax system too tangled to check was part of what people resented.
Notice when trust has gone. Once people stop believing an institution, a false story can land as hard as a true one, as the Diamond Necklace affair shows.
Be wary of single-cause stories. Historians still argue over what caused the Revolution. A neat explanation, including "greed", is a starting point.
Watch what happens to the machinery of punishment. In 1794 the Terror reached Hébert, Danton and finally Robespierre in a little over four months.

Sources

Wikipedia, "French Revolution" (taxation, the Church, feudal dues, debt, the August decrees, the Declaration, the historians' debate, the Terror, the press), and the works it cites for them: Thomas J. Sargent and François R. Velde, "Macroeconomic Features of the French Revolution", Journal of Political Economy (1995); Gerri Chanel, "Taxation as a Cause of the French Revolution: Setting the Record Straight" (2015); David Weir, "Tontines, Public Finance, and Revolution in France and England, 1688–1789", Journal of Economic History (1989); Jean-Pierre Jessene, "The Social and Economic Crisis in France at the End of the Ancien Régime" (2013); Jack Censer and Lynn Hunt, Liberty, Equality, Fraternity (2001); Robert Forster, "The Survival of the Nobility during the French Revolution", Past & Present (1967); Georges Lefebvre, The French Revolution (1962); Keith Michael Baker, "The Idea of a Declaration of Rights" (1994); Niek Pas, De geschiedenis van Frankrijk in een notendop (2008); Jay M. Smith, "Nobility" (2015); William Doyle, The Oxford History of the French Revolution (1990); Simon Schama, Citizens (1989); George Rudé, The French Revolution: Its Causes, Its History and Its Legacy After 200 Years (1988); Pascal Dupuy, "The Revolution in History, Commemoration, and Memory" (2013); Rebecca Spang, "Paradigms and Paranoia", American Historical Review (2003); David A. Bell, "Class, Consciousness, and the Fall of the Bourgeois Revolution", Critical Review (2004); François Furet, "A Deep-rooted Ideology as Well as Circumstance" (1989); Paul Hanson, Contesting the French Revolution (2009); Alfred Cobban, A History of Modern France, vol. I (1963); Harvey Chisick, "The Pamphlet Literature of the French Revolution", History of European Ideas (1993); Jane Chapman, "Republican Citizenship, Ethics and the French Revolutionary Press" (2005).
Wikipedia, "Reign of Terror" (the counts), citing Donald Greer, The Incidence of the Terror during the French Revolution (1935), and Jean-Clément Martin, La Terreur: vérités et légendes (2017).
Encyclopaedia Britannica, "Reign of Terror".
Claire Rioult and Romain Fathi, "The French Revolution executed royals and nobles, yes – but most people killed were commoners", The Conversation (12 July 2023), for Donald Greer's figures.
"Declaration of the Rights of Man, 1789", The Avalon Project, Yale Law School (Articles 14 and 15).
Maximilien Robespierre, "On the Moral and Political Principles of Domestic Policy" (5 February 1794), Internet Modern History Sourcebook, Fordham University.
Wikipedia, "Maximilien Robespierre", citing Raphaël Matta Duvignau, "Le Comité de salut public", La Révolution française (2012); Marisa Linton, Choosing Terror (2013); Pierre Serna, La République des girouettes (2005); Albert Mathiez, "Robespierre: l'histoire et la légende"; and Jonathan Israel, Revolutionary Ideas (2014).
Wikipedia, "Affair of the Diamond Necklace", citing Antonia Fraser, Marie Antoinette: The Journey (2001).
Wikipedia, "French Directory", citing Georges Lefebvre, La France sous le Directoire (1795–1799) (1977).
Wikipedia, "Jean-Jacques Rousseau", citing the Historical Dictionary of Switzerland.
Wikipedia, "Napoleon", citing Adam Zamoyski, Napoleon: The Man Behind the Myth (2018), and Philip Dwyer, Napoleon: The Path to Power (2008); and "French Directory", citing Jean Tulard, Jean-François Fayard and Alfred Fierro, Histoire et dictionnaire de la Révolution française (1998), for the date of the coup.
Wikipedia, "History of French journalism", citing Philip M. Taylor, Munitions of the Mind (2013), and Hugh Gough, The Newspaper Press in the French Revolution (1988).
Wikipedia, "L'Ami du peuple", citing Louis R. Gottschalk, Jean-Paul Marat: A Study in Radicalism (1927, reissued 1966).
Encyclopædia Britannica, 11th edition (1911), on Wikisource: "Robespierre", "La Fayette", "Roland, Jean Marie" (for Madame Roland), "Barras", "Fabre d'Églantine" and "Diamond Necklace, The Affair of the".

Checked September 2026. Wikipedia is used here only where it gives a citation for the sentence, and the work it cites is named above. Lafayette, Madame Roland, Barras, Fabre d'Églantine and the dates of Robespierre's last months are checked against the 1911 Encyclopædia Britannica, the Terror's counts against Britannica and Greer, and the Declaration and Robespierre's speech against their own texts. Where counts differ (deaths in the Terror, newspapers founded), the page gives each.

A Leadership Masterclass: Lessons from Band of Brothers, Churchill and Reagan

Hacks Vitae — Wed, 17 May 2023 16:00:00 GMT

HBO'S THE BAND OF BROTHERS

Band of Brothers is a ten-part HBO miniseries released in 2001, created by Tom Hanks and Steven Spielberg and based on Stephen Ambrose's 1992 book. It follows Easy Company, 2nd Battalion, 506th Parachute Infantry Regiment, 101st Airborne Division, from training through D-Day and the fighting in Holland, Belgium and Germany. It won the Emmy for Outstanding Miniseries in 2002.

The series is a dramatisation of real men, and its dialogue was written for the screen. So this page does not quote the men or the script. The lessons below come from the record of what they did, and the same goes for the two other leaders compiled here: Winston Churchill in the Second World War and Ronald Reagan in the Cold War.

LEADERSHIP LESSONS FROM THE BAND OF BROTHERS

Lead where the danger is

On D-Day, 6 June 1944, Winters, then a lieutenant, led a small group of paratroopers near Brécourt Manor against a battery of four German guns. The two main accounts differ in the details. A 2014 US Army article says he led about a dozen paratroopers against four 105 mm howitzers firing on a Utah Beach causeway, had them hit the guns from the flanks, using a trench to take them one at a time, and that all the guns were destroyed. His Distinguished Service Cross citation, issued in July 1944, tells it in two stages: Winters "with seven enlisted men" put two of the guns out of action, withdrew for reinforcements, and returned with tank support before the remaining two were put out of action. The citation calls them 88 mm guns; the Army article and the National WWII Museum say 105s. Both agree the guns were taken out and that the action reduced the threat to the troops landing on the beach.

Use the ground you have

Both accounts describe a small force taking the guns in stages rather than all at once: in the Army's telling, one at a time from a trench; in the citation's, two guns first, then a return with reinforcements and tanks for the other two. It is a reminder to build the plan around the ground in front of you, and to go back for help when the job is not finished.

Earn the next job

Winters joined Easy Company as a platoon leader. By the end of the war he was a major commanding the regiment's 2nd Battalion, which included his old company. The National WWII Museum says the attrition of war played a part in that rapid rise, and that it was "his superb leadership and ability to complete tough assignments" that endeared him to the men under his command.

Promote the people who are already leading

Carwood Lipton joined the army as a private in 1942, became Easy Company's first sergeant in the field, and was given a battlefield commission as a second lieutenant at Haguenau, France, on 16 February 1945. He later called it "the greatest honor ever awarded" to him.

Read the source, not the poster

Winters's own account is his memoir Beyond Band of Brothers, written with Cole C. Kingseed and based on his wartime diary. It is the better place to go for what he thought about leadership than any line attributed to him online.

Winston Churchill in the Second World War

Before the war: the making of him

Churchill was 65 when he became prime minister in May 1940, and he had already had several careers.

Born 30 November 1874 at Blenheim Palace, the son of Lord Randolph Churchill and his American wife, Jennie Jerome. Educated at Harrow and Sandhurst.
Cuba, 1895. A young officer in the 4th Hussars, he spent his leave watching the war in Cuba. By his own account he first came under fire on his twenty-first birthday; one historian has questioned whether it was that day or a few days later.
The Boer War, 1899. Working as a war correspondent, he was captured and held in Pretoria, and escaped on the night of 12 to 13 December.
Elected MP for Oldham in 1900 as a Conservative. He moved to the Liberals in 1904 and back to the Conservatives in the 1920s, becoming Chancellor of the Exchequer in 1924.
Gallipoli. As First Lord of the Admiralty he was behind the disastrous Gallipoli campaign; heavily criticised, he gave up the post in 1915 and went to serve on the Western Front. The Churchill Archives Centre says he took up painting as an antidote to the anguish of that disaster.

That record matters. The man who led Britain in 1940 was not an untested hero: he was a politician who had already been blamed for a disaster and had come back from it.

Dunkirk, and saying what it was

Between 26 May and 4 June 1940, more than 338,000 British and French soldiers were evacuated from Dunkirk in Operation Dynamo, directed by Admiral Bertram Ramsay from a room deep in the Dover cliffs.

Churchill's speech to the House of Commons on 4 June 1940 is remembered for "we shall fight on the beaches". Earlier in the same speech he told the House: "We must be very careful not to assign to this deliverance the attributes of a victory. Wars are not won by evacuations." It is a useful model: celebrate the rescue, and be plain about the defeat that made it necessary.

After the war

He lost the 1945 general election to Labour's Clement Attlee, the same year Britain celebrated victory in Europe on 8 May.
He returned as prime minister in 1951 and served until 1955.
He was a writer and a painter. He wrote a multi-volume history of the Second World War, was awarded the Nobel Prize in Literature in 1953 "for his mastery of historical and biographical description as well as for brilliant oratory in defending exalted human values", and painted over 500 works.

A line he did not say

"Success is not final, failure is not fatal: it is the courage to continue that counts" is often credited to Churchill. The International Churchill Society says it can find no attribution for it.

Ronald Reagan and the Cold War

Before the presidency

Ronald Reagan was born on 6 February 1911 in Tampico, Illinois. He studied economics and sociology at Eureka College, graduating in 1932, then worked as a radio sports announcer before becoming a film actor. The Reagan Library calls Knute Rockne, All American his most acclaimed film and Kings Row his most artistically acclaimed role. He was first elected president of the Screen Actors Guild in 1947.

His televised speech for Barry Goldwater's presidential campaign, "A Time for Choosing", aired on 27 October 1964 and launched his political career. Two lines from it:

"There is a price we will not pay." "There is a point beyond which they must not advance."
"You and I have a rendezvous with destiny. We'll preserve for our children this, the last best hope of man on earth, or we'll sentence them to take the last step into a thousand years of darkness."

In the same speech Reagan calls "peace through strength" a phrase of Barry Goldwater's.

He was elected governor of California in 1966, was inaugurated in January 1967 and served eight years.

What "he balanced the budget" leaves out

In his autobiography Reagan said he inherited a $200 million deficit. In the summer of 1967, his first year as governor, he signed what PolitiFact describes as a record tax increase for California: an 18 percent rise of roughly $1 billion. According to his biographer Lou Cannon, it raised the sales tax from three to five cents on the dollar, lifted the top income tax rate from 7 to 10 percent, and increased bank, corporation, inheritance, liquor and cigarette taxes.

It went against his campaign promises. Whether it was right depends on your politics; it is a clear case of a leader changing course when the numbers changed.

The 40th President

Reagan was inaugurated as the 40th President on 20 January 1981 and left office on 20 January 1989.

The recession came first

The recession of July 1981 to November 1982 was, in the Federal Reserve's own history, "triggered by tight monetary policy in an effort to fight mounting inflation", and until 2007–09 it was the worst US downturn since the Great Depression. Unemployment reached 10.8% in November and December 1982, the highest in the monthly series that begins in 1948 until 2020. The fall in inflation that followed is credited in the Fed's history to that tighter policy under its chairman, Paul Volcker.

Federal debt rose from about $908 billion at the end of fiscal 1980 to about $2.86 trillion at the end of fiscal 1989, roughly a threefold increase.

Iran–Contra

Americans had been taken hostage in Lebanon by groups with links to Iran. In the hope of freeing them, the administration sold arms to Iran, and in April 1986 a plan was approved to divert more than $12 million of the proceeds to the Contra rebels opposing Nicaragua's government. The arms sales became public in November 1986. Two officials, John Poindexter and Oliver North, were convicted on felony charges; both sets of convictions were later set aside on appeal.

In a televised address on 4 March 1987 Reagan said:

"A few months ago I told the American people I did not trade arms for hostages. My heart and my best intentions still tell me that's true, but the facts and the evidence tell me it is not."

A leader saying in public that the facts contradict him is worth more than most speeches.

How much of the Cold War's end was his

Reagan and Gorbachev signed the Intermediate-Range Nuclear Forces (INF) Treaty on 8 December 1987. It required both sides to destroy their ground-launched ballistic and cruise missiles with ranges between 500 and 5,500 kilometres.

On 12 June 1987, at the Brandenburg Gate in West Berlin, Reagan said: "Mr. Gorbachev, tear down this wall!" The wall came down on 9 November 1989. The US State Department's historians give weight to several things: the Reagan administration's "strong rhetoric in support of the political aspirations of Eastern European and Soviet citizens", Gorbachev's reforms, which "further legitimized popular calls for reform from within", and his decision that the Soviet Union would no longer use military force to keep communist governments in power. How much credit Reagan deserves is still argued over; the INF Treaty needed both men.

What the Three Have in Common

These are three very different leaders, and the record of each includes failures as well as successes. A few patterns are worth taking from them:

Say what happened. Churchill called Dunkirk a deliverance, not a victory. Reagan told the country the facts contradicted him.
Change course when the facts change. Reagan's 1967 tax rise went against his campaign promises; he said he had inherited a deficit.
Come back from a failure. Churchill lost the Admiralty over Gallipoli in 1915 and was prime minister twenty-five years later.
Plan around the ground in front of you. Winters's small force took the guns in stages: one at a time from a trench in the Army's account, and with reinforcements for the last two in his 1944 citation.
Check a quote before you lean on it. The line on persistence credited to Churchill above is not his.

Sources

HBO, Band of Brothers; Television Academy, Band of Brothers awards; Open Library record for Stephen E. Ambrose, Band of Brothers (1992)
US Army, Normandy monument dedicated to Maj. Richard Winters (2014)
The National WWII Museum, Easy Company: England to the Eagle's Nest (tour brochure, 2017); Beyond Band of Brothers
C. Carwood Lipton: biography and obituary
GOV.UK, Past Prime Ministers: Sir Winston Churchill; Nobel Prize, Winston Churchill, biographical and prize summary; Churchill Archives Centre, biography
International Churchill Society: Churchill in Cuba; Escaped a Boer POW camp; "We Shall Fight on the Beaches"; Quotes falsely attributed
Imperial War Museums, What you need to know about the Dunkirk evacuations; National Trust, Churchill the artist
Ronald Reagan Presidential Library: pre-presidential timeline; "A Time for Choosing"; Address on Iran–Contra, 4 March 1987; Brandenburg Gate remarks, 12 June 1987
Miller Center, University of Virginia: Ronald Reagan and Foreign Affairs
PolitiFact, "Even the Gipper" (11 January 2008)
Federal Reserve History, Recession of 1981–82 and The Great Inflation; FRED, unemployment rate (UNRATE); US Treasury Fiscal Data, Historical Debt Outstanding
US Department of State: INF Treaty; Office of the Historian, Fall of Communism in Eastern Europe, 1989

Checked September 2026.

Veni Vidi Vici Method: How to Conquer Life and Achieve Success ∙ Life Hacks for Success

Hacks Vitae — Thu, 23 Mar 2023 16:00:00 GMT

Where the three words come from

Veni, vidi, vici means "I came, I saw, I conquered". Julius Caesar used it about a short war in 47 BC against Pharnaces, son of Mithridates the Great, who was trying to win back his father's kingdom of Pontus. Caesar beat him in a single battle near the city of Zela, in north-eastern Anatolia.

Plutarch and Suetonius report the words in two different settings:

Plutarch says that in announcing the swiftness and fierceness of this battle to a friend at Rome, Amantius, "Caesar wrote three words". He adds that in Latin the three words share the same ending, which gives them their punch.
Suetonius says the words were carried as an inscription in Caesar's triumph for the war in Pontus, "not indicating the events of the war, as the others did, but the speed with which it was finished."

Speed is the point. Suetonius says Caesar beat Pharnaces "within five days after his arrival and four hours after getting sight of him". Encyclopaedia Britannica calls the campaign "a brief local war"; the civil war was decided elsewhere, as the fighting went on in Africa, where Caesar won at Thapsus in 46 BC, and in Spain, where he won at Munda on 17 March 45 BC.

Caesar was never a Roman emperor. Rome was still a republic. He ruled as dictator, and accepted the dictatorship for life, one of the honours Suetonius lists when he explains why Caesar was thought to have abused his power. He was killed on the Ides of March, 44 BC. The first Roman emperor was his great-nephew and adopted son, Augustus, who took that name in January 27 BC.

For how the phrase is reported in other Latin sayings, see famous Latin phrases, checked.

The Veni Vidi Vici Method

The "Veni Vidi Vici method" is this site's own framework, named after Caesar's line. Caesar did not teach it, and none of the people later in this article used it. It is a simple way to remember three stages of getting something done, and it works as a loop: after each goal, look back at what worked and what didn't, and carry that into the next one.

Veni (I Came)

Decide what you are going after, and start. Name the goal clearly enough that you would know if you had reached it, then take a first step, however small. Learning a skill, starting a project and changing jobs all begin the same way.

Vidi (I Saw)

Look hard at the situation before you commit. What will this take? What is in the way? What do you already have, and what are you missing? This is the step that does most of the work, and the easiest to skip. Caesar's line is about speed, but speed came after he had seen the enemy.

Vici (I Conquered)

Act, and follow through. Work through the plan, adjust it when it meets reality, and keep going until the goal is done or you have good reason to drop it.

The Veni Vidi Vici Method and Life

The same three steps can be run through different parts of life. A few examples:

Personal Development

Veni: pick one area, such as a new skill or a fear you want to face.
Vidi: find out what it takes: a course, a teacher, practice time, feedback.
Vici: do the practice, and ask for feedback along the way.

Career Advancement

Veni: name the move you want, such as a promotion or a change of field.
Vidi: read what the role asks for and compare it with what you have; talk to people who do the work.
Vici: close the gap you found, whether that is new responsibilities, a course or a side project.

Entrepreneurship

Veni: choose the idea, product or expansion you want to pursue.
Vidi: check whether people want it, and what money, skills or partners it needs.
Vici: build it, launch it, and learn from how people respond.

The Veni Vidi Vici Method in Real Life

None of the people below used this framework. Their stories are read through it here because each shows one of the three steps clearly, and each part is on the record.

Julius Caesar: speed after sight

Caesar's own line is about the gap between seeing and winning: four hours, by Suetonius's account. The lesson is not to rush everything. It is that once the situation was clear, he did not wait.

Gaius Julius Caesar: summary and life lessons from his rise and fall

Warren Buffett: patience after study

Warren Buffett was chief executive of Berkshire Hathaway until Greg Abel took over on 1 January 2026. On 18 September 2026 Buffett became chairman emeritus and stayed on the board; his son Howard G. Buffett was elected chairman. Abel's first letter to shareholders, dated February 2026, sets out the company's approach: "We will assess value carefully, act patiently, and hold for the long term – preferably forever." That is a Vidi-heavy way of working: long looking, then a decision held for years. (This is a description of a company's stated approach, not investment advice.)

Nelson Mandela: holding to the goal

Nelson Mandela was arrested on 5 August 1962 and released on 11 February 1990, more than 27 years later. According to the Nelson Mandela Foundation, he rejected at least three conditional offers of release while in prison. In 1993 he shared the Nobel Peace Prize with F. W. de Klerk, and on 10 May 1994 he became South Africa's first democratically elected president.

Walt Disney: starting again

Laugh-O-Gram Films, Walt Disney's animation studio in Kansas City, went bankrupt in 1923. The US National Archives, which holds the bankruptcy file, notes that as the case wrapped up, Disney "moved west to California for a new start, where he founded a new, more successful company."

Marcus Aurelius: doing, not talking

Marcus Aurelius, Roman emperor and Stoic, wrote a note to himself that could stand for the Vici step: "No longer talk at all about the kind of man that a good man ought to be, but be such." (Meditations 10.16, George Long's translation.)

Traits and Values to Conquer Life Using the Veni Vidi Vici Method

These are the qualities each step leans on. They are ordinary virtues, not special techniques:

Veni leans on courage (to start) and passion (to care enough to begin).
Vidi leans on critical thinking (to see the situation as it is) and adaptability (to change the plan when the facts change).
Vici leans on discipline, perseverance and resilience (to keep going after setbacks).
All three lean on integrity: being honest with yourself about what you saw, and with others about what you did.

For more on facing fear, see lessons on becoming fearless from Cus D'Amato and Mike Tyson; for discipline, lessons from King Leonidas and the 300 Spartans.

The Veni Vidi Vici Method and Other Methods

Three well-known methods fit neatly inside the three steps:

SMART Goals

The SMART acronym is usually traced to George T. Doran, writing in Management Review in 1981. As later researchers cite his version, the letters stood for specific, measurable, assignable, realistic and time-related; many later versions swap in other words, such as "achievable". Either way, SMART is a tool for the Veni step: it turns a wish into a goal you can check.

Pomodoro Technique

The Pomodoro Technique was created by Francesco Cirillo in the late 1980s, when he was a university student, using a tomato-shaped kitchen timer. Its official site describes working in 25-minute intervals with breaks between them, and calls it "a complete productivity system — not just a timer". It suits the Vici step: steady work in short, protected blocks.

Getting Things Done (GTD)

Getting Things Done is David Allen's method. Its official site sets out five steps: capture, clarify, organize, reflect and engage. The first four are a disciplined version of Vidi (seeing everything you have taken on); the fifth is Vici.

The Veni Vidi Vici Method and Life Hacks for Success

Small habits that support each step:

Keep a daily journal

Write the day's goal and the first step in the morning; note what got in the way during the day; look back in the evening. It is the whole loop in a notebook.

Use tools that fit the job

A task list, a calendar and a timer cover most needs. The tool matters less than the habit of looking at it.

Picture the obstacle, not just the finish

This is the Vidi step, and research on goal-setting points the same way. In four studies published in 2002, the psychologist Gabriele Oettingen and a colleague found that positive expectations (judging a desired future likely) predicted high effort and success, "but the reverse was true for positive fantasies" (enjoying images of the desired future). A technique built on this work, mental contrasting with implementation intentions, has people define the future they want, identify the obstacles in reality, and plan in advance what they will do when the obstacle comes up. A 2021 meta-analysis of 21 studies found it effective for goal attainment "with a small to medium effect size", and noted that because of some publication bias, "the actual effect sizes may be smaller".

Tell someone

Sharing a goal with a friend, a colleague or a group gives you someone to report back to.

Break big goals into smaller ones

A large goal is easier to start when the first piece is small enough to do today.

Mark progress, and learn from misses

Notice what went well, and when something fails, look at why before trying again. That is the loop closing: Vici feeds the next Vidi.

Finding Meaning in Life's Suffering: Life Lessons and Life Hacks from The Last of Us

Conquer Life Using The Veni Vidi Vici Method

Caesar's three words are a boast about speed. The useful part for the rest of us is the middle word. Before acting, see clearly: what you want, what is in the way, and what you will do about it. Then act, and look again when it is done. Arrive, look honestly at what is in front of you, then act and follow through.

Sources

Plutarch, Life of Caesar 50, trans. Bernadotte Perrin (Loeb, 1919), LacusCurtius: https://penelope.uchicago.edu/Thayer/E/Roman/Texts/Plutarch/Lives/Caesar*.html
Suetonius, Life of Julius Caesar 35, 37 and 76, trans. J. C. Rolfe (Loeb, 1914), LacusCurtius: https://penelope.uchicago.edu/Thayer/E/Roman/Texts/Suetonius/12Caesars/Julius*.html
Encyclopaedia Britannica, "Julius Caesar" (section "Civil war") and "Augustus" (sections "Life" and "Military successes"): https://www.britannica.com/biography/Julius-Caesar-Roman-ruler, https://www.britannica.com/biography/Augustus-Roman-emperor
Berkshire Hathaway, news releases of 5 May 2025 and 18 September 2026, and 2025 letter to shareholders (Greg Abel, 28 February 2026): https://www.berkshirehathaway.com/news/may0525.pdf, https://www.berkshirehathaway.com/news/sep1826.pdf, https://www.berkshirehathaway.com/letters/2025ltr.pdf
Nelson Mandela Foundation, "Biography of Nelson Mandela": https://www.nelsonmandela.org/biography
US National Archives at Kansas City, "Walt Disney Bankruptcy": https://www.archives.gov/kansas-city/highlights/disney-bankruptcy
Marcus Aurelius, Meditations 10.16, trans. George Long, Project Gutenberg: https://www.gutenberg.org/ebooks/15877
G. T. Doran, "There's a S.M.A.R.T. way to write management's goals and objectives", Management Review 70 (1981), as cited in Queiruga-Dios et al., Frontiers in Psychology (2023) and Hew et al., Journal of Computing in Higher Education (2023): https://europepmc.org/article/PMC/PMC10050353, https://europepmc.org/article/PMC/PMC9458299
Francesco Cirillo, the Pomodoro Technique official site and book page: https://www.pomodorotechnique.com/, https://www.pomodorotechnique.com/pomodoro-book/
David Allen Company, "What is GTD?": https://gettingthingsdone.com/what-is-gtd/
G. Oettingen and D. Mayer, "The motivating function of thinking about the future: expectations versus fantasies", Journal of Personality and Social Psychology 83 (2002): https://pubmed.ncbi.nlm.nih.gov/12416922/
G. Wang, Y. Wang and X. Gai, "A Meta-Analysis of the Effects of Mental Contrasting With Implementation Intentions on Goal Attainment", Frontiers in Psychology 12 (2021): https://doi.org/10.3389/fpsyg.2021.565202

Checked September 2026.

Lessons from Hacksaw Ridge and Desmond Doss: Personal Development Hacks ∙ Life Lessons ∙ Life Hacks

Hacks Vitae — Tue, 14 Mar 2023 21:09:46 GMT

Life Lessons from Hacksaw Ridge and Desmond Doss

Hacksaw Ridge is a 2016 film directed by Mel Gibson, with Andrew Garfield as Desmond Doss. Doss was real. He was a US Army medic who would not carry a weapon, and for what he did on Okinawa he received the Medal of Honor. He was the first conscientious objector to receive it.

The film is a dramatisation, and some well-known lines from it are film dialogue, not his words. This page sets out what the record says he did, which words are his and which are the screenplay's, and then what a reader can take from the record rather than from the film.

Who Desmond Doss was

Doss was born in Lynchburg, Virginia, on 7 February 1919, and was raised a strict Seventh-day Adventist. He kept Saturday as the Sabbath, and his faith forbade him to bear arms.

When the United States entered the war he was working at the Newport News naval shipyard. Accounts differ on how he came into the Army: the Dictionary of Virginia Biography says he was offered a deferment and chose to join the Army on 1 April 1942; the National WWII Museum says he was drafted in the spring of 1942. Both agree he did not use his objection to stay out. Recalling it in 2001, he said: "I felt like it was an honor to serve my country according to the dictates of my conscience."

He did not like the label conscientious objector. He preferred "conscientious cooperator", and he asked for medical duty. In October 1945 he explained: "While I believe in the commandment 'Thou shall not kill,' and that bearing arms is a sin against God, my belief in freedom is as great as that of anyone else, and I had to help those boys who were fighting for it."

A law passed in 1940 allowed conscientious objectors to serve in non-combatant roles, and he was not alone in doing so. The National WWII Museum puts the number of men designated conscientious objectors during the war at over 70,000, of whom 25,000 joined the armed forces in non-combat roles such as medics and chaplains.

Training was hard on him. Boots and other items were thrown at him as he prayed at night, and his commanding officers tried to have him discharged for mental illness, a discharge he refused to accept. He became a company aid man in the 307th Infantry Regiment, 77th Infantry Division.

Okinawa was not his first battle. His division took part in the liberation of Guam in July 1944 and then fought on Leyte in the Philippines. The Dictionary of Virginia Biography records a Bronze Star for his work on Leyte; the National WWII Museum says he received one on Guam as well.

What he did on the escarpment

On Okinawa, his regiment moved into the line on 29 April 1945 at the Maeda Escarpment, the cliff known as Hacksaw Ridge. He ministered to the wounded there between 29 April and 21 May.

His Medal of Honor citation describes "a jagged escarpment 400 feet high". When the battalion reached the summit, artillery, mortar and machine-gun fire crashed into them, "inflicting approximately 75 casualties and driving the others back". Doss "refused to seek cover and remained in the fire-swept area with the many stricken, carrying them one by one to the edge of the escarpment and there lowering them on a rope-supported litter down the face of a cliff to friendly hands."

He described it himself in 1945: "They had no way of getting back and I could not leave them up there. I was the only medical corpsman with them, so I just went ahead and continued to pick up the wounded still lying in front of the lines and then began the job of getting them off the cliff."

Decades later, in the documentary The Conscientious Objector, he remembered how he kept going: "I was praying the whole time. I just kept praying, Lord, please help me get one more."

The citation goes on. On 2 May he rescued a wounded man 200 yards forward of the lines. Two days later he treated four men cut down while attacking a defended cave, advancing to within eight yards of the enemy, and made four separate trips under fire to bring them back. On 5 May he gave plasma to a wounded artillery officer while shells fell close by, and later that day carried another man 100 yards to safety.

Where "75 men" comes from

Doss is widely credited with saving 75 men on the escarpment. It is an agreed figure, not a count. He later said that his commanding officer wanted to credit him with saving a hundred lives, but he estimated the number at fifty, and they compromised on seventy-five.

The citation itself, as reprinted in full by the National WWII Museum, gives no count of the men he saved. Its "approximately 75" refers to casualties from the barrage, and near its end it says he "saved the lives of many soldiers".

So the record gives a range: fifty by his own estimate, a hundred by his commander's, seventy-five by agreement. Accounts also differ on the day: the Dictionary of Virginia Biography puts the rescue on the first day, 29 April; the National WWII Museum places it on 5 May.

What the citation adds

The citation's last episode is easy to skip past. On 21 May, in a night attack on high ground near Shuri, he stayed in the open while the rest of his company took cover, treating the wounded until a grenade seriously wounded him in the legs. Rather than call another aid man out of cover, he treated his own wounds and waited five hours for litter bearers.

While they were carrying him back, he saw a more critically wounded man nearby, crawled off the litter and told the bearers to take the other man first. Waiting for them to return, he was hit again, suffering a compound fracture of one arm. He bound a rifle stock to the arm as a splint and crawled 300 yards over rough terrain to the aid station.

The Dictionary of Virginia Biography dates the grenade wound to 12 May; the citation, the official document, says 21 May, and this page follows the citation.

President Harry S. Truman presented him with the Medal of Honor at the White House on 12 October 1945. The cost stayed with him. He spent about six years in military and veterans' hospitals recovering from his wounds and was never able to work full time again. He had contracted tuberculosis and lost a lung and five ribs, and in 1976 he suddenly lost his hearing. He died on 23 March 2006.

Hacksaw Ridge Movie and Desmond Doss

Hacksaw Ridge tells this story as a film, with dialogue written for it by its screenwriters, Andrew Knight and Robert Schenkkan. Because the actor is playing Doss, it is easy for that dialogue to be mistaken for Doss's own words.

In the film, Garfield's Doss says he wants to serve as a medic "right in the middle with the other guys", that he is "going to be saving it" while everybody else is taking life, and that "with the world so set on tearing itself apart" he wants to put "a little bit of it back together". Those are lines from the film. We have not found them in any of Doss's recorded words we could read: his 1945 newspaper account and his 2001 oral history as quoted in the Dictionary of Virginia Biography, and the documentary excerpts broadcast by NPR.

His own words, with where they come from:

His statement on the commandment and on freedom, quoted in full earlier on this page — Doss, October 1945, quoted in the Dictionary of Virginia Biography from the Washington Post of 12 October 1945
"They had no way of getting back and I could not leave them up there." — the same 1945 account
"I felt like it was an honor to serve my country according to the dictates of my conscience." — his 2001 oral history for the Library of Congress Veterans History Project, as quoted in the Dictionary of Virginia Biography
"I just kept praying, Lord, please help me get one more." — from Terry Benedict's documentary The Conscientious Objector (2004), as broadcast by NPR

These are the lines to use when quoting Doss; lines from the film read best credited to the film.

Personal Development Hacks ∙ Life Lessons ∙ Life Hacks

What follows is our reading of the record, not a programme. Each point is tied to something Doss did.

Say where your line is, early and plainly. Doss asked for medical duty and said what he would not do, long before he reached a battlefield. NPR notes that the Army could not force him to use a weapon.
You can hold a conviction and still serve. He refused the weapon, and he refused the discharge his officers tried to arrange. Much of his story is the refusal to treat those as the same choice.
Count honestly, even when the bigger number flatters you. Offered credit for a hundred lives, he said fifty. A figure argued down by the person it flatters is easier to trust.
Work in units of one. "Lord, please help me get one more." The citation describes him carrying men "one by one". When a task is too large to face whole, the next one may still be possible.
Choose the name that describes you. He preferred "conscientious cooperator" to a label that described only what he would not do.
Check a quotation before you pass it on. Lines spoken by Doss in the film were written for it. Our guide to spotting a fake quote or an invented source shows how to check one in a few minutes.

Sources

Marc Leepson, "Desmond Thomas Doss (7 February 1919–23 March 2006)", Dictionary of Virginia Biography, Library of Virginia, published 2015. Its own sources include Doss's 1945 Washington Post account and his 2001 oral history in the Library of Congress Veterans History Project.
The National WWII Museum, "Private First Class Desmond Thomas Doss: Medal of Honor", including the full text of his Medal of Honor citation.
Elizabeth Blair, "The Real 'Hacksaw Ridge' Soldier Saved 75 Souls Without Ever Carrying A Gun", NPR Morning Edition, 4 November 2016, transcript, with excerpts from Terry Benedict's documentary The Conscientious Objector.
Hacksaw Ridge, directed by Mel Gibson, written by Andrew Knight and Robert Schenkkan. Film dialogue as transcribed at Wikiquote.

Checked September 2026.

Conquer Your Fears: Face Your Fears ∙ Lessons on Becoming Fearless from Cus D'Amato and Mike Tyson

Hacks Vitae — Mon, 13 Mar 2023 16:00:00 GMT

Conquer Your Fears

Fear is a normal reaction to danger, and it can push you forward or hold you back. This page looks at what fear does in the body, at what is credited to the boxing trainer Cus D'Amato, at how his most famous pupil, Mike Tyson, fits the story, and at what psychology has found about facing fears and recovering from setbacks. At the end are a few ways to use it.

The Science and Psychology of Fear

Understanding Fear

The body's alarm starts in the brain. Harvard Health describes the sequence: the eyes or ears send information to the amygdala, "an area of the brain that contributes to emotional processing", which, when it perceives danger, sends a distress signal to the hypothalamus. The hypothalamus works through the autonomic nervous system, and the adrenal glands release epinephrine (adrenaline): the heart beats faster, breathing quickens, and muscles tense. A second wave of hormones ends with the release of cortisol, which keeps the body on high alert until the threat passes. This combination is what is known as the "fight-or-flight" response.

It evolved to help us react fast to real danger. The trouble starts when it fires at things that are not dangerous, or keeps us away from things we want to do.

The Science of Facing Fear

The clearest evidence on facing fears comes from the treatment of anxiety. In a 2015 review of cognitive-behavioral therapy, the psychologists Antonia Kaczkurkin and Edna Foa describe the reasoning behind exposure therapy: a fear "is maintained by avoidance behaviors which do not allow for new learning to occur". Exposure works by meeting the feared thing and learning that the expected disaster does not happen. For specific phobias, they write, it can be done all at once or as "systematic exposure of gradually increasing intensity", and in vivo exposure is "highly effective for specific phobias" compared with no treatment or placebo.

That is the closest research parallel to the advice credited to D'Amato: face fear rather than run from it. Therapy is not the same as training in a boxing gym, but the logic is similar: avoidance keeps a fear in place, and repeated, manageable contact can change it.

The Philosophy of Fear According to Cus D'Amato

Constantine "Cus" D'Amato (17 January 1908 – 4 November 1985) was a boxing manager and trainer who opened his home to young fighters. He began training Floyd Patterson in 1952, and in 1956, at 21, Patterson won the heavyweight championship, the youngest boxer to do so until Mike Tyson. In 1962 D'Amato moved his gym to Catskill, in upstate New York.

The lines credited to D'Amato treat fear as something to use rather than something to be ashamed of. The one most often credited to him puts it this way:

"Fear is the greatest obstacle to learning. But fear is your best friend. Fear is like fire. If you learn to control it, you let it work for you. If you don't learn to control it, it'll destroy you and everything around you." — Cus D'Amato, as widely quoted (not matched to a primary source)

We have not found it in a source we could check, so treat it as a line attributed to him, not as his confirmed words.

Influence of Cus D'Amato on Mike Tyson

Tyson was born in Brooklyn in 1966 and was in trouble with the law as a boy. During one of his stays in the juvenile system, a counselor there, Bobby Stewart, spotted his talent and introduced him to D'Amato. In 1984 D'Amato became his legal guardian.

Tyson's first professional fight was on 6 March 1985. In just over a year he won 19 consecutive fights by knockout. D'Amato did not see what came next: he died of pneumonia on 4 November 1985. On 22 November 1986, Tyson beat Trevor Berbick to win the WBC heavyweight title and, at 20 years and 145 days, became the youngest heavyweight champion in boxing history. ESPN notes that the record still stands.

So the story gives his method the credit for a success D'Amato never saw. That is worth stating plainly, rather than leaving it as an impression.

Lessons on Becoming Fearless from Cus D'Amato and Mike Tyson

The teaching credited to D'Amato and the exposure research point the same way. Put together, the approach looks like this. It is a description of how the method works, not a treatment plan; a fear that is running your life is worth taking to a professional.

Naming the fear. Being specific about what is feared, and what is being avoided because of it.
Breaking it down. Splitting it into smaller steps, from the least frightening to the most.
Starting where it is manageable. The research describes two forms: flooding, "exposure to the most intense feared stimulus", and gradual exposure, "of gradually increasing intensity", where the steps start at the less intense end.
Staying with it. Avoidance is what keeps a fear in place, so the learning happens by staying long enough to see what actually happens.
Building up. Moving to the next step once the current one has become easier.
Having a corner. D'Amato's young fighters lived in a household that backed them. Having people on your side matters; the research on this is in the resilience section below.

What courage is

Fear is the feeling; courage is what you do with it. The two are not opposites: without fear there is nothing to be brave about, only indifference, which looks similar from outside and behaves very differently under pressure.

A usable definition: courage is acting on what matters while afraid. That rules out recklessness (acting without weighing the cost) and it rules out waiting to stop being afraid, which is a wait with no end.

Where it shows up

Most courage is small and unphotographed:

Everyday. Saying the unpopular thing in a meeting. Backing the person nobody is backing. Telling the truth when a vaguer answer would be easier.
At work. Taking the project you are not yet qualified for; raising the problem that implicates you; giving the feedback that will not be thanked.

It works like a skill

Our reading of the exposure research is that courage behaves more like a trained capacity than a fixed temperament: facing a fear in manageable steps tends to shrink it, and avoiding it keeps it in place.

A famous line puts courage first among the virtues: "Courage is rightly esteemed the first of human qualities, because, as has been said, it is the quality which guarantees all others." It is often credited to Aristotle, but it is Winston Churchill's, from a 1931 essay on King Alfonso XIII of Spain, and Churchill's "as has been said" probably points back to Samuel Johnson, who told James Boswell that "courage is reckoned the greatest of all virtues; because, unless a man has that virtue, he has no security for preserving any other."

Other Real-Life Examples of Conquering Fear

None of these people were unafraid. They were afraid and did it anyway, which is the only version available to the rest of us.

Rosa Parks. On 1 December 1955, on a bus in Montgomery, Alabama, she refused when the driver told her to move back. Her arrest became a rallying point for the Montgomery bus boycott.
Nelson Mandela. Arrested on 5 August 1962 and released on 11 February 1990, he rejected at least three conditional offers of release while in prison.
Malala Yousafzai. In 2012 she was shot in the head on a school bus by a Taliban gunman. She survived, and in 2014 she shared the Nobel Peace Prize with Kailash Satyarthi, "for their struggle against the suppression of children and young people and for the right of all children to education".

A line often credited to Bruce Lee, "Do not pray for an easy life, pray for the strength to endure a difficult one", is close to a much older one by the Boston clergyman Phillips Brooks, which President John F. Kennedy quoted at a prayer breakfast on 7 February 1963: "Do not pray for easy lives. Pray to be stronger men! Do not pray for tasks equal to your powers. Pray for powers equal to your tasks."

Bouncing back: resilience after a fall

What resilience means

Sometimes, despite real effort, things fall apart: a business fails, a relationship ends, a plan collapses. Psychologists call the ability to recover resilience. The American Psychological Association has defined it as "the process of adapting well in the face of adversity, trauma, tragedy, threats or even significant sources of stress".

In a 2014 panel paper, the researchers Steven Southwick, George Bonanno, Ann Masten, Catherine Panter-Brick and Rachel Yehuda stress that resilience "more likely exists on a continuum" rather than being simply present or absent, and that clinicians and researchers "can learn to simultaneously evaluate and teach methods to enhance resilience". In other words, it is not a fixed trait you either have or lack.

What seems to help

Other people. A 2007 review by F. Ozbay and colleagues on social support and stress concludes that "social support may moderate genetic and environmental vulnerabilities and confer resilience to stress".
Treating yourself kindly. In five studies, M. R. Leary and colleagues at Duke University found that self-compassion "attenuates people's reactions to negative events", in some cases more helpfully than self-esteem.
A careful word on growth mindset. It is often recommended here, but the research is split. A 2023 meta-analysis by B. N. Macnamara and A. P. Burgoyne found a small overall effect of mindset interventions that was "nonsignificant after correcting for potential publication bias"; a second 2023 meta-analysis, by J. L. Burnette and colleagues, found small positive effects in targeted groups. Our mindsets article sets out both.

One popular list of this kind is the psychotherapist Amy Morin's 13 Things Mentally Strong People Don't Do, published by William Morrow in December 2014. Its framing is a list of habits to drop rather than habits to add.

What difficult times can teach

Hard periods are not automatically instructive. People mostly learn from them afterwards, and only if they look. These are some of the lessons people commonly take from them; they are reflections, not findings:

Failure is part of the route, not the opposite of it. Every useful skill has a stretch of being bad at it.
Bad seasons are normal, not a verdict. Expecting that some years will be worse takes away some of their power to shock.
A problem is sometimes a redirection. The job that fell through and the relationship that ended often look different at a distance.
Your energy is finite, so choose your battles. Not every provocation deserves a response.
Permanent decisions can usually wait for better weather. Quitting, sending the message or burning the bridge can almost always wait a couple of days.
One honest conversation helps. Telling one person the truth about how you are doing is the opposite of isolation, and support is one of the things the research above keeps finding.

If it goes deeper than discouragement. If you are thinking about suicide or feel unable to keep going, please reach out now. findahelpline.com lists free, confidential helplines worldwide. In the US and Canada, call or text 988. In the UK and Ireland, call Samaritans on 116 123. In Australia, call Lifeline on 13 11 14. In India, call Tele-MANAS on 14416. In the Philippines, call the National Center for Mental Health crisis line on 1800-1888-1553 (or 1553 from a Luzon landline). If you are in immediate danger, call your local emergency number.

Athletes who came back

Mike Tyson

Tyson's career after D'Amato was far from smooth. On 26 March 1992 he was sentenced to six years in prison, and the judge refused to let him go free during his appeal; he was paroled on 25 March 1995, a day short of three years later. He returned to the ring with two fights in the second half of 1995, starting on 19 August, and in 1996 won back two heavyweight titles: the WBC title against Frank Bruno on 16 March and the WBA title against Bruce Seldon on 7 September.

Manny Pacquiao

On 8 December 2012, in their fourth fight, Juan Manuel Márquez knocked Manny Pacquiao out in the sixth round. Three and a half years later, on 19 May 2016, Pacquiao was proclaimed one of the winners of that year's election to the Philippine Senate.

Face Your Fears

What you can take from all this, as ideas to try rather than rules:

Fear is information about what your body thinks is dangerous, not proof that it is.
Avoidance keeps a fear in place; facing it, all at once or in small, repeated steps, is how it tends to shrink.
Courage is a choice made while afraid, not the absence of fear, and in our reading of the exposure research it tends to get easier with practice.
Recovery is helped by other people and by treating yourself kindly, more than by any single slogan.

In the line credited to D'Amato, fear is fire: useful if you control it, destructive if you do not.

Sources

Harvard Health Publishing, "Understanding the stress response".
Antonia N. Kaczkurkin and Edna B. Foa, "Cognitive-behavioral therapy for anxiety disorders: an update on the empirical evidence", Dialogues in Clinical Neuroscience 17(3), 2015.
"D'Amato, Constantine ('Cus')", Scribner Encyclopedia of American Lives, via Encyclopedia.com.
Timothy Fiorvanti, "The long, winding road that brought Mike Tyson back to boxing", ESPN.
Scott Horner, "Judge: Tyson two different people", UPI, 26 March 1992.
Quote Investigator, "Courage Is Rightly Esteemed the First of Human Qualities" (2019, updated 2025); James Boswell, The Life of Samuel Johnson, Project Gutenberg.
National Archives (US), "An Act of Courage, The Arrest Records of Rosa Parks"; Nelson Mandela Foundation, "Biography of Nelson Mandela"; the Nobel Prize, "Malala Yousafzai: Facts".
John F. Kennedy, "Remarks at the 11th Annual Presidential Prayer Breakfast", 7 February 1963, The American Presidency Project.
Steven M. Southwick et al., "Resilience definitions, theory, and challenges: interdisciplinary perspectives", European Journal of Psychotraumatology 5, 2014.
F. Ozbay et al., "Social support and resilience to stress: from neurobiology to clinical practice", Psychiatry (Edgmont) 4(5), 2007.
M. R. Leary et al., "Self-compassion and reactions to unpleasant self-relevant events", Journal of Personality and Social Psychology 92(5), 2007.
B. N. Macnamara and A. P. Burgoyne, Psychological Bulletin 149, 2023; J. L. Burnette et al., Psychological Bulletin 149, 2023.
HarperCollins, 13 Things Mentally Strong People Don't Do by Amy Morin.
Al Jazeera, "Marquez knocks out Pacquiao", 9 December 2012; Associated Press photo report of Pacquiao's proclamation, 19 May 2016, via Bleacher Report.

Checked September 2026. The D'Amato "fear is like fire" passage is still labelled as not matched to a primary source.

Finding Meaning in Life's Suffering: Life Lessons and Life Hacks from The Last of Us

Hacks Vitae — Mon, 13 Mar 2023 16:00:00 GMT

The Last of Us: the game and the show

The Last of Us is a work of fiction. The video game, made by Naughty Dog for the PlayStation 3, came out on 14 June 2013, and BAFTA credited its story to Neil Druckmann and Bruce Straley. Troy Baker and Ashley Johnson played Joel and Ellie in the game and in its sequel, The Last of Us Part II, which followed in 2020. HBO's television series, co-created by Craig Mazin and Neil Druckmann, began on 15 January 2023, with Pedro Pascal as Joel and Bella Ramsey as Ellie.

The story is the same in both. After a fungal infection has wrecked the world, Joel, a hardened survivor who has lost his daughter Sarah, has to take Ellie, who is immune, across a ruined United States to a group called the Fireflies.

Every line of dialogue on this page is credited to the character who says it and to the game or the show. The lessons are our reading of a story, not advice.

Life Lessons and Life Hacks from "The Last of Us"

Each lesson below says what happens in the story, what we take from it, and one way to use it.

Lesson 1: Always be prepared for the worst

In the story. Joel's job is to get Ellie across the country "all while dodging clickers, bandits and all sorts of danger", as Variety put it when the series' release date was announced.

One way to use it. A written plan for an emergency at home is the ordinary version. The US government's Ready.gov, for example, suggests working out how you will "contact one another and reconnect if separated" and agreeing on "a family meeting place that's familiar and easy to find".

Lesson 2: Trust your instincts

In the story. The story's biggest decision comes at the Fireflies' hospital, where, in The Ringer's words, "everything unfolds (and unravels) quickly".

Our reading. Instinct is worth more in some places than others. The psychologists Daniel Kahneman and Gary Klein concluded in 2009 that judging an intuition means looking at "the predictability of the environment in which the judgment is made" and at the person's "opportunity to learn the regularities of that environment". They add that "subjective experience is not a reliable indicator of judgment accuracy": feeling sure is not the same as being right.

One way to use it. Before acting on a gut feeling, ask whether you have done this often, with quick feedback on whether you got it right.

Lesson 3: Never give up

In the story. In the game's final scene, just before Ellie asks him to swear that what he told her about the Fireflies is true, Joel tells her that he struggled a long time with surviving, and that she, no matter what, keeps finding something to fight for. Transcriptions of the exact wording differ, so it is paraphrased here.

Our reading. The point is less about toughness than about having something to fight for.

One way to use it. Small goals you can finish are one way to keep moving when big ones stall. There is more on keeping going in Motivation vs Discipline.

Lesson 4: Find meaning in suffering

In the story. When Joel suggests turning back and going to his brother Tommy's instead, Ellie refuses:

"After all we've been through. I mean, everything that I've done. It can't be for nothing." — Ellie, The Last of Us (2013 game), as transcribed on IMDb

Our reading. This is the story's central claim, and the one to hold loosely. Suffering does not come with meaning attached; people tend to build it afterwards, if they look for it.

One way to use it. Writing down what happened, and what you made of it, is one way to do that looking.

Lesson 5: Learn to let go

In the story. For much of the journey Joel will not talk about Sarah: in the series' sixth episode he shuts Ellie down as soon as she mentions her. In the final episode, The Ringer's recap notes, Joel and Ellie "discuss his daughter freely", and "it's clear that Joel is no longer repressing his past".

One way to use it. Epictetus opens his Enchiridion by separating the things "within our power" from those "beyond our power"; there is more on that in Stoicism.

Lesson 6: Value your relationships

In the story. By the end, in The Ringer's words, "Ellie is Joel's surrogate daughter".

One way to use it. Setting aside regular time for the people who matter, and being fully present for it, is the plain version.

Lesson 7: Embrace change

In the story. Nothing in Joel and Ellie's journey stays the same for long; they cross the country, from one place and set of dangers to the next.

One way to use it. A short list of things you have wanted to try, started with the smallest, makes change something you choose rather than something that only happens to you.

Lesson 8: Face your fears

In the story. The game is full of things to be afraid of: the infected, and human enemies that include, in Tech Digest's description, "military forces on the lookout for the runaways" and "scavengers".

One way to use it. Taking on a fear a step at a time, starting small, is the everyday version. If a fear is affecting your life, the UK's NHS suggests seeing a GP; it lists talking therapies "such as cognitive behavioural therapy (CBT)" among the treatments for phobias. There is more on fear, from a different angle, in Conquer Your Fears.

Lesson 9: Be resourceful

In the story. The game makes resourcefulness the way you play. A 2013 review in Tech Digest says Joel and Ellie "will constantly be scavenging, always on the verge of running out of ammunition", and that "scavenging parts littered around levels will let Joel craft new projectiles".

One way to use it. When a problem is stuck, list several ways round it before choosing one.

Lesson 10: Cherish the simple things

In the story. Near the end, Joel and Ellie come across a group of giraffes grazing, which The Ringer calls "one of the game's most timeless and uplifting scenes".

One way to use it. Noticing, each day, a few things you are glad of is a small habit with the same idea behind it.

Finding Meaning in Life's Suffering

Our reading. The ending complicates every lesson above. At the Fireflies' hospital, Joel learns that the surgery meant to lead to a cure "will kill her in the process", in The Ringer's summary. He kills the Fireflies to save her, then tells Ellie what The Ringer calls an "improbable tale of many immune survivors" and of doctors who had given up on a cure. When she asks him to "swear to me that everything you said about the Fireflies is true", he does. The story does not say whether he was right, and it leaves that judgement to the player and the viewer.

"Endure and survive."

It is not Joel's motto; in the game he only repeats it back to Ellie as a question. It is the catchphrase of Savage Starlight, the comic series Ellie reads in the game ("It's what the hero says after a big battle"), and in the HBO series Sam teaches Ellie the sign-language actions for it.

Lessons from King Leonidas and the 300 Spartans: Life Hacks, Mindset Hacks, and Life Lessons

Sources

PlayStation.Blog: "The Last of Us is Coming in June" (February 2013); "'Heads-up' PlayStation Store update, 12th June 2013"; "The Last of Us Part II is now available" (June 2020).
Variety, "'The Last of Us' Dominates BAFTA Games Awards" (13 March 2014); Variety, "HBO's 'The Last of Us' Sets Early 2023 Release Date" (2 November 2022).
IMDb, "The Last of Us (Video Game 2013) – Quotes" (user-transcribed).
Ben Lindbergh, "The 'Last of Us' Finale Is Just as Ambiguous and Agonizing as the Game's Indelible Ending", The Ringer (13 March 2023).
Edmond Tran, "The Last of Us – Episode 5 recap – 'Endure and Survive'", GamesHub (2023).
Tech Digest, "Review: The Last Of Us (PS3)" (June 2013).
Daniel Kahneman and Gary Klein, "Conditions for intuitive expertise: a failure to disagree", American Psychologist 64(6), 2009 (PubMed abstract).
Ready.gov (US Department of Homeland Security), "Make a Plan".
Epictetus, Enchiridion, trans. T. W. Higginson (Project Gutenberg #45109).
NHS, "Phobias".

Checked September 2026. Release dates, credits and plot points are from the sources above. The game's dialogue is quoted from a user transcription, and one line whose transcriptions disagree is paraphrased.

Lessons from King Leonidas and the 300 Spartans: Life Hacks, Mindset Hacks, and Life Lessons

Hacks Vitae — Wed, 01 Mar 2023 15:24:21 GMT

(Best played in the background while reading the article)

LESSONS FROM KING LEONIDAS AND THE 300 SPARTANS

In 480 BC a Greek force led by the Spartan king Leonidas held the pass of Thermopylae against the army of the Persian king Xerxes, and lost. It is one of the most retold stories from the ancient world, and much of the picture people carry today comes from the film 300 (2006), which is fiction built on the history rather than a record of it.

This page starts with what the ancient sources say happened, then draws the lessons from that version. It is a less tidy story than the film's, and a more useful one.

First, what happened

The main ancient account is Herodotus, Histories, Book 7. The short version:


When	480 BC. Herodotus says Xerxes let four days go by, expecting the Greeks to leave, and attacked on the fifth; the fighting lasted three days
Where	Thermopylae, a narrow pass in central Greece between a steep hillside and the sea
Greek force	Herodotus names contingents adding up to about 5,200, plus the Opuntian Locrians, whom he does not count. The later historian Diodorus gives figures that come to about 7,400. Britannica puts the Greek army at about 7,000 men
Persian force	Herodotus gives 2,641,610 fighting men, with as many again in camp followers. Britannica gives modern estimates of anywhere from 70,000 to 300,000
How it ended	A Malian named Ephialtes told Xerxes about a path over the mountain; Persian troops went round it and the Greeks were surrounded
Result	A Persian victory. The Persians went on to take Athens and set fire to the Acropolis. The war turned at Salamis (480 BC) and Plataea (479 BC)

Herodotus also explains why the force was so small. The Spartans sent Leonidas ahead and meant to follow in full strength once the festival of the Carneia was over, and the other allies planned the same because the Olympic festival fell at the same time: "they did not suppose that the fighting in Thermopylai would so soon be decided."

Two parts of the story are worth holding on to before the lessons start.

The Thespians stayed by choice. When the path was betrayed, most of the allies left or were sent away. Herodotus is specific about who remained: the Thebans "stayed against their will and not because they desired it, for Leonidas kept them, counting them as hostages; but the Thespians very willingly." There were 700 of them, under their commander Demophilos, and they "stayed behind and died with them." If the story is about choosing to stand, they have as good a claim to it as anyone.

The helots were not volunteers. Herodotus mentions them twice: a Spartan named Eurytus had his helot lead him into the fighting, and after the battle, visitors who came to look at the dead found that "the Helots also were among those that they saw." He does not say how many there were. Helots were the enslaved population whose labour left Spartan citizens free for war — in Plutarch's words, "the Helots tilled their ground for them." The full-time training the story admires rested on other people's forced work, and that belongs in the picture.

LIFE HACKS

1. Prepare before the hard day, not during it

When Xerxes sent a horseman to look at the Greek camp, he saw "some of the men practising athletic exercises and some combing their long hair," and Xerxes took it for foolishness. The exiled Spartan king Demaratus told him it was a custom: "whenever they are about to put their lives in peril, then they attend to the arrangement of their hair." Preparation often looks like routine from the outside. The part worth borrowing is the unglamorous one: the ordinary preparations, done before the day that counts.

2. Choose ground where your disadvantage matters least

The Greeks chose to hold a narrow pass rather than meet the Persian army on open ground. Herodotus says the Persians gained nothing at the pass, "seeing that they were fighting in a place with a narrow passage, using shorter spears than the Hellenes, and not being able to take advantage of their superior numbers." When you are outnumbered or under-resourced, the choice of where to compete can matter as much as effort.

3. Know what you are good at

Herodotus describes the Spartans as "men perfectly skilled in fighting opposed to men who were unskilled." Their advantage was training, and they fought in the way that made the most of it. The everyday version is to know which of your skills are real and to lean on those, rather than on the ones you wish you had.

4. Discipline as a rule decided in advance

Demaratus explained Spartan discipline to Xerxes as obedience to law: "though free, yet they are not free in all things, for over them is set Law as a master." The law, he said, "bids them not flee out of battle from any multitude of men, but stay in their post." Discipline, in this picture, is less about willpower in the moment and more about a rule settled before the moment arrives.

5. Keep a sense of humour under pressure

Herodotus reports that a Spartan called Dienekes was said to be the bravest man of all. Told that the Persian arrows were so many they would hide the sun, he answered that this was good news, "for if the Medes obscured the light of the sun, the battle against them would be in the shade and not in the sun." Unlike “Come and take them”, it comes from Herodotus rather than a later collection, and it is a joke.

MINDSET HACKS

1. Sacrifice for the greater good — and mixed motives

Plutarch's collection of Spartan sayings has Leonidas answer an offer from Xerxes that "for me to die for Greece is better than to be the sole ruler over the people of my race." Herodotus gives a less tidy picture. He reports an oracle that either Sparta would be destroyed or a Spartan king would die, and says Leonidas stayed partly "desiring to lay up for himself glory above all the other Spartans." Real sacrifice can come with mixed motives, and it can still be real.

2. Lead by example

Herodotus believed Leonidas sent most of the allies away; he did not leave himself. Herodotus says he fell in the fighting on the last day, "having proved himself a very good man," and adds that he himself had learned the names of all three hundred Spartans.

3. Let people go who do not want to stay

Herodotus says Leonidas "perceived that the allies were out of heart and did not desire to face the danger with him to the end," and so "he ordered them to depart." The lesson we draw from it: letting people leave, and being clear about why you are staying, is a way of knowing the limits of what you can ask of others.

4. Courage in the face of fear

The Greeks were afraid, and Herodotus says so: when the Persians came near, they "were in dread, and deliberated about making retreat." Courage in this story is not the absence of fear; it is what they did after the deliberation. Plutarch's collection gives Leonidas the most quoted reply of all: when Xerxes wrote "Hand over your arms," he answered, "Come and take them."

5. Honour and integrity

The clearest case of integrity in the story is the Thespians, who were free to go and stayed. Herodotus records the name of the Thespian who won most honour: Dithyrambos, son of Harmatides.

LIFE LESSONS

1. Sacrifice and Betrayal

The Greek position was lost because a local man hoped to be paid for what he knew. Herodotus names him as Ephialtes, son of Eurydemos, a Malian, who came to Xerxes "supposing that he would win a very great reward." A price was later put on his head, and he was later killed, for a different reason, by a man of Trachis named Athenades. Herodotus also records another version, which named two other men as the guides, and says he "can by no means accept" it.

2. Unity and Cooperation

The force at the pass was a coalition: Spartans, Tegeans, Mantineans, Arcadians, Corinthians, men of Phlius and Mycenae, Thespians, Thebans, Phocians and Locrians. Herodotus says they were "ordered by companies as well as by nations, and they fought successively each in turn." Demaratus's description of the Spartans makes the same point about working together:

"So also the Lacedemonians are not inferior to any men when fighting one by one, and they are the best of all men when fighting in a body." — Herodotus, Histories 7.104, spoken by Demaratus to Xerxes, in G. C. Macaulay's translation.

3. Persistence and Determination

For two days the Persians attacked and got nowhere. On the second day, Herodotus says, they came on expecting the defenders to be worn out, and "finding nothing different from that which they had seen on the former day, retired back from the fight." Persistence here looked like doing the same hard thing well on the second day as on the first.

4. Adaptability and Flexibility

The Spartans varied their tactics. Herodotus describes them turning their backs "and make a pretence of taking to flight," then turning to face their pursuers once the Persians came after them "with shouting and clashing of arms." On the last day, knowing they were surrounded, they changed again and "advanced out much further than at first into the broader part of the defile." The lesson is that discipline and flexibility work together: you can change tactics without changing your commitments.

5. Self-Sacrifice

The best-known line about Spartan self-sacrifice is not from Thermopylae at all:

"Another, as she handed her son his shield, exhorted him, saying, 'Either this or upon this.'" — Plutarch, Moralia, "Sayings of Spartan Women" 16, trans. F. C. Babbitt.

Plutarch gives it to an unnamed mother. The Loeb editor notes that the anthology of Stobaeus quotes Aristotle giving it to Gorgo, Leonidas's wife, and that it is often spoken of as a general Spartan custom. Plutarch lived around 45–120 CE, so he was writing some five to six centuries after the battle, collecting sayings that were already old stories.

6. Respect for the people beside you

The coalition held only as long as its members stood together, and its most generous act came from the Thespians, who could have left. The Spartans were not simply admirable: their society ran on enslaved labour, as the helots at the pass show. The discipline and the coalition are worth keeping; the whole package is not.

7. Honour and Loyalty

One line in Herodotus about fighting for honour rather than money is not about Spartans at all. In Herodotus (Histories 8.26), a Persian, Tigranes son of Artabanos in Macaulay's translation, hears that the prize at the Olympic Games is only an olive wreath and says: "what kind of men are these against whom thou hast brought us to fight, who make their contest not for money but for honour!"

8. The Price of Freedom

Thermopylae was a defeat. The Persians went on to Athens, where they "plundered the temple and set fire to the whole of the Acropolis." What the stand bought was time and an example; the war turned at Salamis later that year and at Plataea in 479 BC. The sacrifice mattered, and it did not by itself save anything — a more useful shape for a lesson than the version where courage is rewarded straight away.

(Best played in the background while reading the article)

The epitaph

Herodotus records three inscriptions at the pass. One was for the whole force:

"Here once, facing in fight three hundred myriads of foemen, / Thousands four did contend, men of the Peloponnese."

Another was for the Spartans alone, and it is the most quoted line about the battle:

"Stranger, report this word, we pray, to the Spartans, that lying / Here in this spot we remain, faithfully keeping their laws." — Herodotus, Histories 7.228, trans. G. C. Macaulay

Herodotus says the Amphictyons, whose councillors had their seats at Anthela beside the pass, put up the inscriptions, and he names the poet Simonides only as the author of a third, for the seer Megistias.

Note what the Spartan couplet does not say. No defiance, no glory — only that they kept their laws. It matches what Demaratus told Xerxes in the same book: that the Spartans are free, but law is a master "whom they fear much more even than thy people fear thee."

What to take from it

The sayings are the part people remember, but the record offers something more practical:

Do the preparation while it still looks like routine. The Spartans combed their hair and exercised before the fight; the scout saw nothing impressive.
Pick the ground. A small side did well where numbers stopped counting, and failed only when that ground was taken away.
Let the unwilling go, and know why you stay. Leonidas sent the allies home; the Thespians chose to remain.
Check the story before borrowing its lesson. The film version rewards courage at once. The historical one does not, and it is clearer about what courage costs.

One more line from Plutarch's collection fits the story, though Plutarch's collection gives it to a later Spartan king, Agis, son of Archidamus — Agis II, who ruled from about 427 BC, according to the Loeb editor, who adds that some sayings in that section belong to the younger Agis III: "the Spartans did not ask 'how many are the enemy,' but 'where are they?'"

Sources

Herodotus, The History of Herodotus, vol. 2, trans. G. C. Macaulay — Project Gutenberg. Book 5.48; Book 7.104, 7.185–186, 7.202–229, 7.239; Book 8.25–26 and 8.53.
Diodorus Siculus, Library of History 11.4, Loeb Classical Library edition (1946) — LacusCurtius.
Plutarch, Moralia, "Sayings of Spartans" (Leonidas 10–11; Agis son of Archidamus 3) and "Sayings of Spartan Women" 16, trans. F. C. Babbitt, Loeb Classical Library edition (1931), with the editor's notes 6 and 23 — Sayings of Spartans, Sayings of Spartan Women.
Plutarch, Life of Lycurgus 24, Loeb Classical Library edition (1914) — LacusCurtius.
Battle of Thermopylae — Britannica, for the size of the armies, the dates of Salamis and Plataea, and the film.
Agis II — Britannica.
Plutarch — Stanford Encyclopedia of Philosophy, for Plutarch's dates.

Checked September 2026.

Motivation vs Discipline: How to Cultivate Both for Success ∙ Mindset Hack ∙ Motivation Hack

Hacks Vitae — Wed, 01 Mar 2023 03:46:45 GMT

Motivation vs Discipline

Motivation is why you act and how much you want to. Discipline is sticking to a plan when the wanting runs low. Motivation picks the goal and gets you started; discipline carries you through the ordinary days when you do not feel like it. Most goals need both, and most of the practical advice about "discipline" is really about making the right action easier, so it depends less on how you feel.

"Motivation is what drives you. Discipline is what keeps you." A common saying of uncertain origin, repeated in several versions.

What is Motivation?

The APA Dictionary of Psychology defines motivation as "the impetus that gives purpose or direction to behavior", and also as "a person's willingness to exert physical or mental effort in pursuit of a goal or outcome". It separates internal motivating forces from external factors such as rewards or punishments.

What is Discipline?

The same dictionary defines self-discipline as "the control of one's impulses and desires, forgoing immediate satisfaction in favor of long-term goals", and as "resolute adherence to a regimen or course of action in order to achieve one's goals".

Put side by side: motivation is about wanting to do something, discipline about doing it whether or not you want to at that moment.

Why motivation runs out, and what to build instead

Motivation rises and falls, so a plan that only works on high-motivation days will stall. One research tradition asks which kind of motivation holds up.

Self-determination theory, developed by Edward Deci and Richard Ryan, looks at the type of motivation, not just the amount. It distinguishes autonomous motivation (doing something because you value it or find it interesting) from controlled motivation (doing it because of rewards, pressure or fear of what others will think). The theory's own summary puts it this way: people are "often moved by external factors such as reward systems, grades, evaluations, or the opinions they fear others might have of them", and just as often "motivated from within, by interests, curiosity, care or abiding values".

In a 2000 paper, Ryan and Deci proposed three basic psychological needs that, when satisfied, "yield enhanced self-motivation and mental health and when thwarted lead to diminished motivation and well-being":

Autonomy — a sense that you chose this, or at least endorse it.
Competence — a sense that you can do it and are getting better.
Relatedness — a sense of connection to other people.

What follows from it: if a goal keeps stalling, ask which of the three is missing. Is it a goal you chose? Can you see progress? Is anyone else involved or aware?

Start with why, then work out how

The "why" is the reason the goal matters to you; the "how" is the method. It is common to spend a long time comparing methods for a goal you are not sure you want. Self-determination theory argues that conditions supporting autonomy, competence and relatedness foster "enhanced performance, persistence, and creativity", which suggests that a goal you do not actually endorse will be harder to keep going.

One way to find the why: write the goal down, then ask "why does that matter?" four or five times, until the answer is something you actually care about rather than something you think you should say.

How to Cultivate Motivation and Discipline for Success

These methods each have research behind them.

Make if-then plans

Peter Gollwitzer calls them implementation intentions: instead of "I will study more", you decide "If it is 7 p.m. on a weekday, then I will sit at my desk and open the textbook." A 2006 meta-analysis by Gollwitzer and Paschal Sheeran, pooling 94 independent tests, found that these plans had "a positive effect of medium-to-large magnitude" on reaching goals. A 2024 update by Sheeran, Olivia Listrom and Gollwitzer, pooling 642 tests, found them effective for "cognitive, affective, and behavioural outcomes", with effect sizes from d = .27 to .66 depending on the outcome, and larger when plans took an if-then form, people were highly motivated to pursue the goal, and plans were rehearsed. They helped people start, stay on track despite distractions, and drop failing courses of action.

Build habits in a fixed context

In habit research, a habit is a behaviour that has become automatic in a particular context. In a 2010 study, Phillippa Lally and colleagues asked 96 volunteers to repeat an eating, drinking or physical activity behaviour every day in the same context (for example "after breakfast") for 12 weeks. For the people whose data could be modelled, the time to reach 95% of their peak automaticity ranged from 18 to 254 days. Two findings are worth knowing:

It varies a great deal, and it can take a long time. There is no fixed number of days.
"Missing one opportunity to perform the behaviour did not materially affect the habit formation process." One missed day does not reset you.

The Stoic teacher Epictetus made a similar point about practice: "if you would make any thing a habit, do it" (Discourses 2.18, trans. George Long).

Set clear goals and make progress visible

Specific goals tell you whether you have hit them. Visible progress matters too: Teresa Amabile and Steven Kramer, analysing the diaries of knowledge workers, found that "nothing contributed more to a positive inner work life (the mix of emotions, motivations, and perceptions that is critical to performance) than making progress in meaningful work". Their name for it is the progress principle.

Stay accountable

Telling someone your goal and arranging regular check-ins brings in the "relatedness" part of self-determination theory, and it creates a small external reason to keep going on days when the internal one is weak.

Procrastination is often an emotion problem, not a time problem

The psychologists Fuschia Sirois and Tim Pychyl argue that procrastination, as a failure of self-regulation, "has a great deal to do with short-term mood repair and emotion regulation". It tends to happen with tasks that feel aversive. You put the task off because putting it off makes you feel better now, and the cost lands on your future self.

That helps explain why a better calendar often does not fix it: the problem is how the task feels, not where it sits in the week.

What follows from it:

Name the feeling. Bored, anxious, resentful, afraid of doing it badly? Knowing which one you are avoiding tells you what to change about the task.
Forgive the last time. In a study of first-year psychology students at Carleton University, Michael Wohl, Tim Pychyl and Shannon Bennett (2010) found that students who reported more self-forgiveness for procrastinating before their first midterm procrastinated less before the second. This was a correlation, not an experiment, but it fits the mood-repair idea: self-blame is one more bad feeling to avoid.
Shrink the first step until it is not worth avoiding: open the document, write one rough sentence.
Use an if-then plan for the moment you usually put it off (see above).
Work in short timed blocks. Francesco Cirillo's Pomodoro Technique is built around a timer, and its official site describes the full method as including "daily planning, interruption management, and effort estimation".
Remove the friction that comes just before the avoidance: the phone in another room, one window open.

When it is not motivation — it is energy

Before treating low energy as a motivation problem, it helps to know that tiredness has ordinary causes. The NHS lists not getting enough sleep, an unhealthy diet and lack of exercise, stress, and depression among the common ones.

When it may not be about motivation at all. If you have felt tired for a few weeks and are not sure why, if the tiredness is affecting your daily life, or if it comes with other symptoms, such as weight loss or mood changes, the NHS says to see a GP; a productivity system is the wrong tool for that. Depression, an underactive thyroid, anaemia, sleep apnoea, diabetes and some medicines can all cause tiredness, and a better goal-setting method will not fix any of them. This page is not medical advice.

Momentum: small, visible, and soon

Amabile and Kramer's finding points to a practical habit: make the progress small enough to happen today, and make it visible.

Start smaller than you think you should. Five minutes still counts, and it is the version you will actually do on a bad day.
Track it where you can see it. A row of marks is evidence of progress.
Finish one thing before starting three.
Protect the streak, but do not worship it. One missed day did not derail habit formation in Lally's study; the risk is letting one miss become a new pattern.

What setbacks can teach

Setbacks do not always come with an upside. Stories that say every closed door hides an open one are built from the cases where it did; the people whose setbacks simply cost them years are rarely the ones telling the story. A hard thing is not automatically a gift, and telling someone in the middle of one that it is can be cruel.

What holds up better is narrower, and still useful:

Rejection is information about fit, and sometimes about the other side. It is only sometimes about you.
Losing a role you were comfortable in removes the reason not to move. For some people that is the push they needed; for others it is just a bad year. Both happen.
Encouraging other people costs little, and it is the one item on this list with no obvious downside.

Take the lesson if there is one. You are not required to be grateful for the thing that hurt.

Motivation vs Discipline Quotes

These four are traced to their original documents:

"Discipline is the soul of an army. It makes small numbers formidable; procures success to the weak, and esteem to all." — George Washington, Instructions to Company Captains, 29 July 1757
"Generally then if you would make any thing a habit, do it; if you would not make it a habit, do not do it, but accustom yourself to do something else in place of it." — Epictetus, Discourses 2.18, trans. George Long
"Hold every hour in your grasp. Lay hold of to-day's task, and you will not need to depend so much upon to-morrow's." — Seneca, Letters to Lucilius 1.2, trans. Richard M. Gummere
"No longer talk at all about the kind of man that a good man ought to be, but be such." — Marcus Aurelius, Meditations 10.16, trans. George Long

Many popular lines on this subject, credited to speakers, coaches and athletes, are left out because we could not trace them to a dated source.

Mindset and Motivation Hack

A way to put this together, described as the research frames it rather than as a prescription:

Pick a goal you actually endorse, and check it against the three needs: your choice, visible progress, other people involved.
Write one if-then plan for when and where you will do the work.
Do it in the same context each time, and expect it to take weeks or months to become automatic, not days.
When you catch yourself putting it off, name the feeling and shrink the first step.
Forgive the missed day and pick up at the next one.
If you are tired all the time for weeks, look at energy before discipline.

Sources

American Psychological Association, APA Dictionary of Psychology: motivation, self-discipline.
Ryan, R. M., & Deci, E. L. (2000). Self-determination theory and the facilitation of intrinsic motivation, social development, and well-being. American Psychologist, 55(1), 68–78. PubMed 11392867.
Deci, E. L., & Ryan, R. M. (2008). Self-determination theory: A macrotheory of human motivation, development, and health. Canadian Psychology, 49(3), 182–185. doi:10.1037/a0012801.
Center for Self-Determination Theory, The Theory.
Gollwitzer, P. M., & Sheeran, P. (2006). Implementation intentions and goal achievement: A meta-analysis of effects and processes. Advances in Experimental Social Psychology, 38, 69–119. doi:10.1016/S0065-2601(06)38002-1.
Sheeran, P., Listrom, O., & Gollwitzer, P. M. (2024). The when and how of planning: Meta-analysis of the scope and components of implementation intentions in 642 tests. European Review of Social Psychology. doi:10.1080/10463283.2024.2334563.
Lally, P., van Jaarsveld, C. H. M., Potts, H. W. W., & Wardle, J. (2010). How are habits formed: Modelling habit formation in the real world. European Journal of Social Psychology, 40(6), 998–1009. doi:10.1002/ejsp.674.
Amabile, T. M., & Kramer, S. J. (2011). The Power of Small Wins. Harvard Business Review.
Sirois, F., & Pychyl, T. (2013). Procrastination and the priority of short-term mood regulation: Consequences for future self. Social and Personality Psychology Compass, 7(2), 115–127. White Rose repository.
Wohl, M. J. A., Pychyl, T. A., & Bennett, S. H. (2010). I forgive myself, now I can study: How self-forgiveness for procrastinating can reduce future procrastination. Personality and Individual Differences, 48(7), 803–808. doi:10.1016/j.paid.2010.01.029.
The Pomodoro Technique, official site.
NHS, Tiredness and fatigue and Underactive thyroid: symptoms.
George Washington, Instructions to Company Captains, 29 July 1757, Founders Online, National Archives.
Epictetus, Discourses 2.18, trans. George Long, Wikisource.
Seneca, Letters to Lucilius 1, trans. Richard M. Gummere, Wikisource.
Marcus Aurelius, Meditations, trans. George Long, Project Gutenberg #15877, 10.16.

Checked September 2026.

The Ripple Effect: Gandhi's Principles in Creating Positive Change in the World

Hacks Vitae — Tue, 28 Feb 2023 16:02:19 GMT

The Ripple Effect of Creating Positive Change

The idea behind this page is simple: change starts with one person's conduct and spreads to the people who see it. Gandhi's name is often attached to it, usually through one line.

"Be the change you wish to see in the world." — widely credited to Gandhi, but not found in his writings.

The nearest passage he wrote is in a 1913 article on snake-bite in his South African newspaper, Indian Opinion, published in Gujarati and translated in his Collected Works:

"We but mirror the world. All the tendencies present in the outer world are to be found in the world of our body. If we could change ourselves, the tendencies in the world would also change. As a man changes his own nature, so does the attitude of the world change towards him. … We need not wait to see what others do."

Gandhi died in 1948. The earliest close match to the popular line that Quote Investigator found is from 1974: "be the change you want to see happen", in a book chapter by the educator Arleen Lorrance.

The 1913 passage says more than the slogan does. It ties change in the world to change in yourself, and it ends with a reason to start now rather than wait for others.

The words Gandhi used

The four principles on this page (nonviolence, truth, love and self-discipline) are popular English shorthand. The terms Gandhi used are more specific.

Satyagraha. In September 1906, Indians in the Transvaal pledged to resist a new registration ordinance. Gandhi at first called the campaign "passive resistance", then ran a small prize in Indian Opinion for a better name. Maganlal Gandhi suggested sadagraha, "firmness in a good cause", and Gandhi changed it to satyagraha: "Truth (Satya) implies love, and firmness (agraha) engenders and therefore serves as a synonym for force." He called it "the Force which is born of Truth and Love or non-violence".
Accepting the penalty. In Hind Swaraj he put the method this way: "If I do not obey the law and accept the penalty for its breach, I use soul-force." In 1920 he added that non-violence "does not mean meek submission to the will of the evil-doer, but it means the putting of one's whole soul against the will of the tyrant."
Swaraj. Self-rule, in two senses: independence for India, and rule over oneself. "It is Swaraj when we learn to rule ourselves," he wrote in Hind Swaraj.
Brahmacharya. The "self-discipline" of the popular version. Gandhi described it as the "control in thought, word and action, of all the senses at all times and in all places" (Young India, 1924).

Gandhi's Principles in Action

Each principle below is paired with an example. The examples illustrate the idea; except where the page says so, they are not claims that the people involved took their lead from Gandhi.

Principle 1: Nonviolence

For Gandhi, nonviolence was a choice made from strength, not a way of avoiding a fight. In his 1920 essay "The Doctrine of the Sword" he went as far as this: "I do believe that where there is only a choice between cowardice and violence I would advise violence." What he asked for instead was the harder third option.

The clearest case of his method travelling is the American civil rights movement. Martin Luther King Jr. first used nonviolent direct action in the Montgomery bus boycott of 1955–56, and wrote in 1960: "Christ furnished the spirit and motivation while Gandhi furnished the method."

Principle 2: Truth

In Gandhi's own gloss, truth and love come together: "Truth (Satya) implies love." A modern parallel is investigative journalism, which works by publishing what can be shown and letting it change what people do.

Principle 3: Love

The example here is Médecins Sans Frontières (Doctors Without Borders), awarded the Nobel Peace Prize in 1999 "in recognition of the organisation's pioneering humanitarian work on several continents."

The same 1920 essay has Gandhi's view of forgiveness: "forgiveness is more manly than punishment", but "abstinence is forgiveness only when there is the power to punish".

"The weak can never forgive. Forgiveness is the attribute of the strong." — Gandhi, answering a question from the press in Karachi after the executions of Bhagat Singh and his comrades, as reported in Young India, 2 April 1931. He had been asked whether it was impolitic to forgive a government "guilty of a thousand murders".

Principle 4: Self-Discipline

"Strength does not come from physical capacity. It comes from an indomitable will," Gandhi wrote in "The Doctrine of the Sword". The example here is environmental work that depends on steady, unglamorous effort, such as the Green Belt Movement below.

The Ripple Effect of Gandhi's Principles in Action

Example 1: The Salt March

Gandhi and a group of volunteers left Sabarmati on 12 March 1930 and walked to Dandi on the coast, in protest at the British salt laws, under which the colonial government collected a tax on salt. Sources give the distance as between 200 and 241 miles. They reached Dandi on 5 April, and early the next morning Gandhi walked down to the sea to make salt.

It was easy to copy. In the coastal cities of Bombay and Karachi, crowds made salt too, and British authorities arrested more than 60,000 people. It did not end British rule by itself: after about a year of struggle, Gandhi negotiated a truce with the British government.

Example 2: The Green Belt Movement

The Green Belt Movement was founded in 1977 by Wangari Maathai, under the auspices of the National Council of Women of Kenya. It encouraged women to plant trees in their local environments; it spread to other African countries and contributed to the planting of over thirty million trees. Maathai won the Nobel Peace Prize in 2004, the first African woman to receive it.

Example 3: The Legacy of Dr. Martin Luther King Jr.

King travelled to India in 1959 and met the Gandhi family. He received the Nobel Peace Prize in 1964. His own account of his thinking, written for The Christian Century, credits Gandhi with the method and Christ with the spirit.

The fuller picture

Gandhi's methods changed the twentieth century. His record also includes things that are hard to defend, and they belong on the page too.

Race. In an 1894 open letter from Durban he complained that the colony regarded Indians as "little better, if at all, than savages or the Natives of Africa", and that the Indian was "being dragged down to the position of" Black Africans, for whom he used a racial slur. In December 2018 the University of Ghana removed a statue of him after lecturers petitioned against it, calling him "racist".
His marriage. The historian Vinay Lal notes that Gandhi's own autobiography records sharp disagreements with his wife, Kasturba, in the first two decades of their marriage, when he "unreasonably sought to bring her under his control".
The 1940s experiments. After Kasturba's death in 1944, Gandhi took young women, including his grandnieces Abha and Manu, to bed with him at night, naked, as a test of his brahmacharya, according to Lal's study. Close co-workers objected, and Kishorelal Mashruwala gave up editing Gandhi's weekly, Harijan.
Caste. In 1936, replying to an address by B. R. Ambedkar in a piece he titled "Dr. Ambedkar's Indictment", Gandhi wrote that "there is nothing in the law of varna to warrant a belief in untouchability", yet defended varna itself, which he said teaches that "we have each one of us to earn our bread by following the ancestral calling". In the same passage he wrote that varna "defines not our rights but our duties" and that all callings "are good, lawful, and absolutely equal in status." He said Ambedkar had "over-proved his case".

None of this unmakes the Salt March. It does suggest that what lasted was a method rather than a man: King took the method and gave it his own spirit.

What you can do with this

Start where you are. The 1913 passage ends with "We need not wait to see what others do." The idea is that your own conduct is the one part of the ripple you control.
Make it easy to copy. Making salt was something anyone near the coast could do. A change that others can repeat without asking permission does not need a leader present.
Know the cost you are willing to carry. Satyagraha rested on accepting the penalty openly rather than avoiding it. Whatever you stand for, it helps to decide in advance what you will accept for it.
Check the quote before you share it. "Be the change" sounds like Gandhi and is not in his writings. Looking for a date, a place and a text takes a minute and keeps the real passage in circulation.

Sources

The 1913 passage — Gandhi, "General Knowledge About Health [-XXXII]", Indian Opinion, 9 August 1913, in The Collected Works of Mahatma Gandhi, vol. XII (1964), p. 158 (scan on Wikimedia Commons).
"Be the change", 1974 — Quote Investigator, "Be the Change You Wish To See in the World".
Satyagraha — Gandhi, Satyagraha in South Africa, chs. 12–13 (Navajivan).
Soul-force and swaraj — Gandhi, Hind Swaraj, chs. 14 and 17.
"The Doctrine of the Sword" — Young India, 11 August 1920, in The Collected Works of Mahatma Gandhi, vol. XVIII, pp. 131–134 (scan on Wikimedia Commons).
"The weak can never forgive" — Young India, 2 April 1931 (Internet Archive).
King and Gandhi — Martin Luther King Jr., "Pilgrimage to Nonviolence", 13 April 1960; King Institute, "Gandhi, Mohandas K."; NobelPrize.org, Martin Luther King Jr., biographical.
The Salt March — King Institute (above); HISTORY, "Mohandas Gandhi begins 241-mile civil disobedience march".
Doctors Without Borders — NobelPrize.org, The Nobel Peace Prize 1999.
The Green Belt Movement and Wangari Maathai — The Green Belt Movement; NobelPrize.org, Wangari Maathai, facts.
The 1894 open letter — The Collected Works of Mahatma Gandhi, vol. I, "Open Letter", Durban, December 1894, p. 177 (Internet Archive).
The Ghana statue — BBC News, "'Racist' Gandhi statue removed from University of Ghana", 13 December 2018.
Kasturba and the 1940s experiments — Vinay Lal, "Nakedness, Nonviolence, and Brahmacharya: Gandhi's Experiments in Celibate Sexuality", Journal of the History of Sexuality 9, no. 1–2 (2000).
Caste — Gandhi, "Dr. Ambedkar's Indictment-II", Harijan, 18 July 1936, in The Collected Works of Mahatma Gandhi, vol. LXIII, pp. 153–154 (Internet Archive).

Checked September 2026.

Overcoming Survivorship Bias: Making Impartial Decisions by Understanding the Psychology Behind it.

Hacks Vitae — Sat, 25 Feb 2023 10:12:48 GMT

Survivorship Bias Definition

Survivorship bias means drawing conclusions from the cases that made it through while the ones that failed have dropped out of view. Investopedia defines it, in investing, as "the tendency to evaluate the performance of existing investments without considering those that have failed, leading to an overestimation of historical performance". The same trap works anywhere the failures quietly disappear from the sample.

It is closely related to what researchers call selection bias. The Catalogue of Bias, published by the Centre for Evidence-Based Medicine, defines that as a sample that differs "systematically from the population of interest leading to a systematic error in an association or outcome". A group made only of survivors is a sample like that.

This page covers where the idea is usually illustrated, what is documented about the most famous example, the mental habits that make it easy to fall into, and a few questions that help.

Examples of Survivorship Bias

Most Famous Example of Survivorship Bias

The most famous example is Abraham Wald, a statistician at the Statistical Research Group at Columbia University during the Second World War. The usual telling goes like this: the military looked at the damage on bombers that came back and planned to protect the areas with the most hits, and Wald pointed out the flaw.

The reasoning is sound: the damage you can see is damage a plane survived. Hits in places that brought planes down are under-represented among the planes that return, because those planes are not there to be counted.

What the retelling adds

The most repeated example of survivorship bias has picked up some details of its own along the way.

What is documented:

Wald's work survives as "a series of eight memoranda originally published by the Statistical Research Group at Columbia University for the National Defense Research Committee in 1943", reprinted by the Center for Naval Analyses in 1980.
The memoranda are mathematics: methods for estimating how vulnerable each part of an aircraft is from the damage on the survivors. In a worked example with made-up figures ("Of 400 planes on a bombing mission, 359 return"), he concludes that "for the observed data of this hypothetical example, the engine area is the most vulnerable".
The report names the practical use itself: conclusions from its vulnerability tables "can be used as guides for locating protective armor and can be used to make a prediction of the estimated loss of a future mission".
The 1980 reprint says the work "was never published externally", and that his method "has been employed in the analysis of data from both the Korean and Vietnam Wars".

What the popular version adds:

The famous diagram is not his. The aeroplane outline covered in red dots is the work of Cameron Moll, who wrote in 2022 that "sometime around 2005 I hastily plotted fictitious red dots on a poorly-chosen commercial aircraft outline". He adds that the image seen most often on social media "is from Wikipedia (creator unknown) and is a recreation of my diagram".
The tidy ending. The story often ends with the armour being moved in 1943 and crews being saved. The reprinted memoranda say the results can guide where armour goes; they do not report what was actually done with them. Marc Mangel and Francisco Samaniego, who studied the memoranda, wrote in 1984: "We do not know whether it was used during World War II". They add that in the Vietnam War, analysts used Wald's techniques on the A-4 aircraft, and that their analysis "led to structural modifications that improved the A-4's survivability".

None of this weakens the idea. The logic holds whether or not anyone moved the armour in 1943.

Survivorship Bias in Finance

The clearest everyday case is investment funds. Funds that do badly are often closed or merged into other funds, and in most cases, Investopedia notes, a closed fund's performance "is not integrated into future reporting". A list of today's funds, and their average past returns, therefore leaves out many that did badly, which is why Investopedia describes the result as "an overestimation of historical performance".

Survivorship Bias in Business

The same shape appears in business stories. The companies that get studied and written about are the ones still standing. If their habits are copied without looking at firms that had the same habits and failed, the habits can look more decisive than they were.

Survivorship Bias in Personal Relationships

Advice from couples married for fifty years is worth hearing, but it comes only from the couples who stayed together. Couples who followed similar advice and split up are not usually asked. The advice may still be good; the sample on its own cannot show that it is.

Psychology of Survivorship Bias

Survivorship bias starts as a problem with the sample, but two well-documented mental habits make it easy to fall into.

The two habits behind it

The availability heuristic. In a 1974 paper in Science, Amos Tversky and Daniel Kahneman described people judging "the frequency of a class or the plausibility of a particular development" by the "availability of instances or scenarios": how easily examples come to mind. If successes are the examples that come to mind, the mental sample is skewed before any reasoning starts.
Confirmation bias. The Catalogue of Bias defines it as "the search for and use of information to support an individual's ideas, beliefs or hypotheses". A survivors-only sample supplies plenty of that.

Overcoming Survivorship Bias

The practical version is one question: before drawing a conclusion from a set of examples, ask who is missing from it, and why.

A few follow-ups make that question easier to use:

Where did this list come from? A "best funds", "most successful founders" or "happiest couples" list has already filtered out the failures. Check what the filter was.
What happened to the ones that failed? If they had the same habits, strategy or advice, the habit is not what separated them.
When and where did the success happen? Conditions such as a rising market or a new industry can do much of the work that later gets credited to a person's choices.
Is there a count of the whole group? A figure that includes the failures, such as all funds launched in a year rather than those still open, gives a truer picture than any list of winners.

Start Making Impartial Informed Decisions

Survivorship bias is easiest to see in other people's reasoning and hardest to see in your own. The missing cases do not announce themselves; that is the whole problem. Asking "who is not in this picture?" is a small habit, and it applies to a fund brochure, a founder's life story, a pile of success advice and, as the Wald story shows, to the stories we tell about the idea itself.

Sources

Abraham Wald, A Method of Estimating Plane Vulnerability Based on Damage of Survivors, reprinted as Center for Naval Analyses Research Contribution 432 (July 1980), from memoranda of 1943.
Marc Mangel and Francisco J. Samaniego, "Abraham Wald's Work on Aircraft Survivability: Rejoinder", Journal of the American Statistical Association 79 (1984).
Cameron Moll, "Abraham Wald and the airplane diagram with red bullet holes – here's the origin story" (24 March 2022).
Investopedia, "Survivorship Bias: Definition and Use in Investing" (updated April 2026).
Catalogue of Bias (Centre for Evidence-Based Medicine), "Selection bias" and "Confirmation bias".
Amos Tversky and Daniel Kahneman, "Judgment under Uncertainty: Heuristics and Biases", Science 185 (1974).

Checked September 2026. Wald's 1943 memoranda at Columbia's Statistical Research Group are documented in the 1980 reprint; the red-dotted aircraft diagram is a modern illustration, drawn around 2005 by its creator's own account.

The Principles of Sun Tzu: The Art of War in Business

Hacks Vitae — Fri, 17 Feb 2023 05:47:50 GMT

THE PRINCIPLES OF SUN TZU

The Art of War is a short Chinese book on strategy, traditionally credited to a general called Sun Tzu ("Master Sun"). It is divided into 13 chapters, and some of its admirers apply it to business, politics and sport.

This page groups its ideas into seven principles. The grouping is ours, not a list from the book, and so is every "in business" reading below. The quotations are from Lionel Giles's English translation of 1910, the first annotated English translation, which is free to read on Project Gutenberg. Each one carries its chapter and verse, so you can find it in a couple of minutes.

Know Your Enemy and Yourself

"If you know the enemy and know yourself, you need not fear the result of a hundred battles. If you know yourself but not the enemy, for every victory gained you will also suffer a defeat." — Sun Tzu, The Art of War III.18 (Giles)

The verse goes on: "If you know neither the enemy nor yourself, you will succumb in every battle." The Chinese says a hundred battles too (百戰不殆), so the popular "thousand battles" version is not what the text says.

Our reading, for a business: know your own costs, strengths and limits as well as you know your competitors'. The verse gives the two halves equal weight.

Understand the Terrain

"The natural formation of the country is the soldier's best ally; but a power of estimating the adversary, of controlling the forces of victory, and of shrewdly calculating difficulties, dangers and distances, constitutes the test of a great general." — X.21 (Giles)

Chapter X opens with six kinds of terrain, and chapter XI lists nine varieties of ground. As we read this verse, ground helps, but judgement decides.

Our reading: in business the "terrain" is the market you are standing in, meaning its size, its customers, its rules and who else is in it. Study it before you commit money or time to it.

Focus on Your Strengths

"The good fighters of old first put themselves beyond the possibility of defeat, and then waited for an opportunity of defeating the enemy." — IV.1 (Giles)

The next verse, IV.2, adds: "To secure ourselves against defeat lies in our own hands, but the opportunity of defeating the enemy is provided by the enemy himself."

Our reading: secure the part you control first, such as a product that works, costs you can carry and customers you already serve well. Then watch for the opening a competitor gives you.

The line "Opportunities multiply as they are seized" is often credited to Sun Tzu. It is not in Giles's translation: we searched the full text and it does not appear.

Exploit Your Competitors' Weaknesses

"Military tactics are like unto water; for water in its natural course runs away from high places and hastens downwards." — VI.29 (Giles)

"So in war, the way is to avoid what is strong and to strike at what is weak." — VI.30 (Giles)

Our reading: look for the need a competitor serves badly, rather than attacking the thing they do best.

Adapt Your Strategy to the Situation

"Therefore, just as water retains no constant shape, so in warfare there are no constant conditions." — VI.32 (Giles)

Earlier, in the first chapter, the book says: "According as circumstances are favourable, one should modify one's plans." (I.17). Chapter VIII, "Variation of Tactics", is about the same idea. Giles notes that its heading means literally "The Nine Variations".

Our reading: have a plan, and expect to change it.

The line "The wise adapt themselves to circumstances, as water molds itself to the pitcher" is sometimes credited to Sun Tzu. It is not in Giles's translation. The book does use water to describe tactics, in VI.29 to VI.32, but not in these words.

Deception and Surprise

"All warfare is based on deception." — I.18 (Giles)

"Let your plans be dark and impenetrable as night, and when you move, fall like a thunderbolt." — VII.19 (Giles)

Our reading: the useful business version is surprise, not deception. Keep plans private until they are ready, and move quickly once you commit. We would not read this as permission to mislead customers or partners, who are not the enemy in this picture.

Build Alliances

"We cannot enter into alliances until we are acquainted with the designs of our neighbours." — VII.12 (Giles)

The book is choosy about allies. In chapter XI it says of a warlike prince who attacks a powerful state: "Hence he does not strive to ally himself with all and sundry, nor does he foster the power of other states." (XI.55).

Chapter III also ranks the ways of winning. The best is to win without a battle at all:

"Hence to fight and conquer in all your battles is not supreme excellence; supreme excellence consists in breaking the enemy's resistance without fighting." — III.2 (Giles)

The popular "the greatest victory is that which requires no battle" is a paraphrase of this verse, not a translation of it.

Our reading: partners are worth having when you know what they want. A deal that settles a contest can be worth more than winning it.

What the book contains

The Art of War is short: 13 chapters, each a run of numbered sayings. The titles below are Giles's, with one verse from each chapter to show what it is about.

Laying Plans: the general who wins "makes many calculations in his temple ere the battle is fought" (I.26).
Waging War: "There is no instance of a country having benefited from prolonged warfare." (II.6)
Attack by Stratagem: winning without fighting (III.2), and knowing the enemy and yourself (III.18).
Tactical Dispositions: first make yourself impossible to defeat, then wait for the opening (IV.1).
Energy: "Energy may be likened to the bending of a crossbow; decision, to the releasing of the trigger." (V.15)
Weak Points and Strong: avoid what is strong and strike at what is weak (VI.30).
Manœuvering: tactical manœuvering, "than which there is nothing more difficult" (VII.3).
Variation of Tactics: the general who understands varying his tactics "knows how to handle his troops" (VIII.4). Its heading means literally "The Nine Variations".
The Army on the March: encamping the army and "observing signs of the enemy" (IX.1).
Terrain: "six kinds of terrain" (X.1).
The Nine Situations: "nine varieties of ground", from dispersive ground to desperate ground (XI.1).
The Attack by Fire: "five ways of attacking with fire" (XII.1), and a warning that "no general should fight a battle simply out of pique" (XII.18).
The Use of Spies: the foreknowledge a general needs "cannot be elicited from spirits" (XIII.5), and "Knowledge of the enemy's dispositions can only be obtained from other men." (XIII.6)

The two "nine" titles are easy to mix up. In Giles's translation, "The Nine Variations" is the literal meaning of chapter VIII's heading; chapter XI is "The Nine Situations".

A note on the author. Sima Qian's history, the Shiji, records a tradition that the book was written by one Sun Wu of the state of Qi, and read by King Helü of Wu, who reigned about 514 to 495 BC. From about the 12th century, Chinese scholars began to question whether Sun Tzu existed at all, mainly because an older history, the Zuo zhuan, never mentions him. When the book was written is still discussed. Wikipedia's article says it is "now generally accepted" that the earlier text was "completed sometime between 500 and 430 BCE"; elsewhere the same article says the earliest parts "probably date to at least 100 years later" than the 6th century when Sun Tzu is said to have lived, and in the early 20th century Liang Qichao argued that the book was written in the 4th century BC, by Sun Bin. Giles himself argued that the book was already in existence towards the end of the 5th century BC.

In 1972, bamboo slips found in two Han dynasty tombs at Yinqueshan, in Shandong, included a text attributed to Sun Tzu, corresponding to the received text, and a separate text attributed to Sun Bin. Wikipedia's article says this showed that much of the confusion came from there being two texts, not one. The tombs had been sealed between about 134 and 118 BC (Wikipedia's page on the slips gives about 140 or 134 BC as the earlier date). None of this makes the book less useful; it does mean "Sun Tzu said" is a looser claim than it sounds.

THE ART OF WAR IN BUSINESS

A few ways to use the book without over-reading it:

Check the verse before you quote it. Giles's translation is free online and numbered by chapter and verse. If a "Sun Tzu" line has no chapter and verse, search the text for it first.
Start with yourself. IV.1 and IV.2 put your own position first: what you control, what you can afford, what you do well.
Read the market as ground. Chapters X and XI are about reading the ground before moving; the business version is learning a market before committing to it.
Treat "deception" as timing. Keeping plans quiet until they are ready is fair; misleading the people you serve is a different thing.
Remember it is a book about war. Much of it, from spies to attacking with fire, has no business equivalent in our reading, and that is fine.

Sources

Sun Tzu, The Art of War, translated by Lionel Giles (London: Luzac, 1910), Project Gutenberg eBook #132 (all quotations and verse numbers).
孫子兵法 (the Chinese text), Wikisource (chapter 3, 知彼知己，百戰不殆).
Wikipedia, "The Art of War" (the Shiji tradition, authorship and dating, the Yinqueshan slips, Giles's 1910 translation).
Wikipedia, "Yinqueshan Han Slips" (the 1972 discovery).

Checked September 2026. Every quotation is from Giles's 1910 translation on Project Gutenberg, with its chapter and verse; the Chinese wording is from Wikisource; the history of the text follows Wikipedia's articles on The Art of War and the Yinqueshan slips, which give more than one view of when the book was written.

Gaius Julius Caesar: Summary ∙ Life Lessons from the Rise and Fall of Julius Caesar

Hacks Vitae — Thu, 02 Feb 2023 20:27:30 GMT

Gaius Julius Caesar conquered Gaul, won a civil war, made himself dictator for life, and was stabbed to death by a group of senators on the Ides of March, 44 BC. He was never a Roman emperor: Rome was still a republic, and the first emperor was his adopted heir, Augustus. This page covers his life in brief, the lines people quote (and who really wrote them), and what his rise and fall can teach.

GAIUS JULIUS CAESAR SUMMARY

Caesar was born in July 100 BC in Rome. That is the date most modern accounts give, though Britannica marks the year with a question mark; its 1911 edition, following the historian Theodor Mommsen, preferred 102 BC. The ancient biographers agree on his age at death: Suetonius says he "died in the fifty-sixth year of his age".

His family was patrician, one of Rome's old noble houses, and claimed descent from Iulus, son of Aeneas and grandson of the goddess Venus.

RISE TO POWER: CONQUEST OF GAUL

Between 58 and 50 BC Caesar conquered the rest of Gaul up to the left bank of the Rhine. He wrote his own account of the campaigns, the Commentaries on the Gallic War.

"I am not much in fear of these fat, long-haired fellows, but rather of those pale, thin ones." — Caesar, as reported by Plutarch (Caesar 62, Bernadotte Perrin's translation), meaning Brutus and Cassius

CIVIL WAR

In January 49 BC Caesar crossed the Rubicon, the river that separated Cisalpine Gaul from the rest of Italy, with a single legion, and began a civil war against Pompey. Plutarch describes him stopping at the river and wavering for a long time before crossing with the phrase "Let the die be cast"; Suetonius gives it as "The die is cast."

The civil war ran from 49 to 45 BC and ended with his victory over Pompey's sons at Munda, in Spain, on 17 March 45 BC. In 47 BC, in the middle of it, he beat Pharnaces, son of Mithridates the Great, at Zela so quickly that he summed it up in three words: veni, vidi, vici.

The Romans then appointed him dictator for life. Plutarch is blunt: "This was confessedly a tyranny, since the monarchy, besides the element of irresponsibility, now took on that of permanence."

FALL: ASSASSINATION ∙ TRAGIC DEATH

More than sixty men joined the conspiracy, led by Gaius Cassius and by Marcus and Decimus Brutus. They chose a meeting of the Senate called for the Ides of March in the Hall of Pompey. Caesar died on 15 March 44 BC, stabbed with twenty-three wounds.

Both biographers record a warning. Plutarch writes that "a certain seer warned Caesar to be on his guard against a great peril" on the Ides of March; Suetonius names him as Spurinna. On the day, Caesar joked that the Ides had come, and the seer answered, in Plutarch's version: "Ay, they are come, but they are not gone."

Killing him did not end the fighting. Suetonius notes that hardly any of the assassins survived Caesar by more than three years. In his will Caesar had adopted his great-nephew Gaius Octavius, and it was Octavius who came out on top: in January 27 BC the Senate gave him the title Augustus (on the 16th, according to two surviving Roman calendars; on the 17th, according to the later writer Censorinus), and he is counted as the first of the Roman emperors.

"Beware the ides of March." — the Soothsayer, in Shakespeare's Julius Caesar, I.ii. Caesar is the one being warned; the famous wording is Shakespeare's.

GAIUS JULIUS CAESAR QUOTES

Several of the most quoted lines credited to Caesar are best known from Shakespeare, not from Caesar. Here is where the familiar ones come from.

TRUE, as reported. "I came, I saw, I conquered" (veni, vidi, vici). After Zela in 47 BC. Plutarch says Caesar wrote the three words to a friend at Rome, Amantius; Suetonius says they were carried on an inscription in his Pontic triumph, "not indicating the events of the war, as the others did, but the speed with which it was finished." There is more on the phrase in our page on the Veni Vidi Vici method, where the "method" is this site's own framework, not Caesar's.

TRUE. "Experience is the teacher of all things." This is Caesar's own writing, Civil War 2.8: ut est rerum omnium magister usus. It is an aside in his account of a siege, where the soldiers found by experience that their small brick shelter would be more useful raised into a tower. It is not a free-standing motto.

TRUE. "Men willingly believe what they wish." Caesar, Gallic War 3.18: fere libenter homines id quod volunt credunt. W. A. McDevitte's translation reads "in most cases men willingly believe what they wish". He is explaining why an enemy fell for a trick.

TRUE, as reported. "I would rather be first here than second at Rome." Plutarch (Caesar 11) says Caesar said it while crossing the Alps and passing a poor village.

FICTION. "Et tu, Brute?" The famous version is Shakespeare's (Julius Caesar, III.i: "Et tu, Brute! Then fall, Caesar."), and the Latin phrase was already familiar on the English stage in his day: The True Tragedie of Richard Duke of York, printed in 1595, has "Et tu, Brute? wilt thou stab Cæsar too?". No ancient source gives Caesar these words. Suetonius says Caesar died "uttering not a word, but merely a groan at the first stroke, though some have written that when Marcus Brutus rushed at him, he said in Greek, 'You too, my child?'"

FICTION. "Let me have men about me that are fat." Shakespeare's version (I.ii) of the line Plutarch reports about "pale, thin" men.

More at: Gaius Julius Caesar Quotes ∙ Words of Wisdom

GAIUS JULIUS CAESAR LIFE LESSONS

These are our readings of his life, not rules he left behind. Each one rests on something the sources record.

Speed can be the message

Suetonius says the three words of veni, vidi, vici were chosen to show "the speed with which it was finished", not the events of the war. What he wanted remembered was how fast it was done.

Big decisions are rarely as bold as they look afterwards

"The die is cast" sounds like instant resolve. Plutarch's account is slower: Caesar stopped at the Rubicon and "communed with himself a long time in silence as his resolution wavered back and forth" before he crossed. The famous line came at the end of the doubt, not instead of it.

Words change how people act

Suetonius says Caesar addressed his soldiers "not as 'soldiers,' but by the more flattering term 'comrades'." When the Tenth legion demanded their discharge, "with a single word, calling them 'citizens,' instead of 'soldiers,' he easily brought them round", and they insisted on following him to Africa. One word told them what they would be giving up.

Mercy won loyalty, and did not protect him

Caesar was known for sparing enemies. Plutarch says "he pardoned many of those who had fought against him, and to some he even gave honours and offices besides, as to Brutus and Cassius". Suetonius adds that at Pharsalus he cried out, "Spare your fellow citizens." Two of the men he pardoned and promoted led the plot to kill him.

Power without limits made enemies

Plutarch calls the dictatorship for life "confessedly a tyranny" because it added permanence to power that already answered to no one. Suetonius reports that some thought Caesar trusted the senators' oath so fully "that he dismissed even the armed bodyguard of Spanish soldiers that formerly attended him". His rise shows what drive and speed can do; his fall shows what happens when nobody can check a leader's power, including the leader himself.

GAIUS JULIUS CAESAR LEADERSHIP IN TODAY'S WORLD

A few ways to put the lessons above to use, whether you lead a team or just your own week:

Say what the result was, briefly. Three words carried Zela. A short, plain report of what got done often lands better than a long one.
Let the doubt happen before the leap. Caesar's pause at the Rubicon is part of the story. Taking time to think before a big, hard-to-reverse choice is not the opposite of decisiveness.
Pick your words for the people in front of you. "Comrades" and "citizens" did more than a speech. Notice what the words you use tell people about where they stand.
Be generous, and still keep your eyes open. Pardoning opponents won Caesar support. It did not mean every pardoned opponent was a friend.
Build in checks on your own power. The lesson of the Ides of March is less about enemies than about a man nobody could say no to.

Sources

Plutarch, Life of Caesar, trans. Bernadotte Perrin (Loeb Classical Library), LacusCurtius: chapters 11, 32, 50, 57, 62, 63, 69.
Suetonius, Life of Julius Caesar, trans. J. C. Rolfe (Loeb Classical Library), LacusCurtius: chapters 32, 37, 67, 70, 75, 80–83, 86, 88, 89.
Caesar, Gallic War 3.18 and Civil War 2.8, Latin text at The Latin Library; Gallic War trans. W. A. McDevitte, Project Gutenberg #10657.
William Shakespeare, Julius Caesar, MIT Shakespeare text: I.ii, III.i.
Julius Caesar, ed. E. C. Black (New Hudson Shakespeare), introduction, "Et tu, Brute", Project Gutenberg #28334.
The Roman calendars (fasti) for 16 January, in Ehrenberg and Jones's translation (attalus.org); Censorinus, De die natali 21.8, Latin text at LacusCurtius.
Encyclopaedia Britannica, "Julius Caesar" (read in a browser; the site blocks scripts).
Encyclopaedia Britannica, 11th edition (1911), "Caesar, Gaius Julius" and "Augustus", Wikisource.

Checked September 2026. Dates, the assassination and the quotations were read in the ancient sources above and in the two editions of Britannica. If you can show any of this wrong, with a source, we want to see it.

IKIGAI: Living A Purpose-Driven Life ∙ Finding Meaning and Fulfillment in Work and Life

Hacks Vitae — Thu, 02 Feb 2023 19:14:03 GMT

Ikigai (生き甲斐) is a Japanese word for something that makes life worth living. In Japan it can be people, work or a hobby, and it can be small and private. In the West it has become a four-circle diagram about finding the right career. Those are two different ideas. This page covers both, starting with where the diagram came from, and ends with a few ways to use each.

First: the famous diagram is not Japanese

You have probably seen the four overlapping circles: what you love, what you are good at, what the world needs and what you can be paid for, with ikigai in the middle.

The ikigai version was published on 14 May 2014 by the blogger Marc Winn, in a post called "What is your ikigai?". In a note added to that post later, Winn writes: "The diagram is not mine. It began as a purpose Venn diagram by the Spanish astrologer Andrés Zuzunaga". His own part, in his words, was to "add a single word to a shape that already existed." Zuzunaga, commenting on the same post in 2017, dated his Spanish graphic to 2012.

Winn's post also pointed readers to Dan Buettner's TED talk about communities where people live far longer than average.

Ken Mogi, the Japanese neuroscientist who wrote The Little Book of Ikigai, rejects the diagram as a definition. In a talk transcribed by the site Ikigai Tribe in January 2022, he said that "for the western readership, probably this diagram is useful in structuring what motivates you and what gives you pleasure". But defining ikigai as the overlap of all four circles is, he said, "too narrow of a definition for ikigai". His example: "if you love to paint a picture, and you're not paid for that, that is ikigai". His conclusion: "ultimately, it's completely wrong. You shouldn't really take this diagram seriously."

Why the difference matters

The diagram turns ikigai into a career problem. Two of its four circles are about the market: what the world needs, and what you can be paid for.

In Mogi's account, neither is required. A hobby that pays nothing can be ikigai, and so can something you are not good at: "you might want to play musical instruments, and you might not play well, but that can also be your ikigai." He adds that it can be "something vocational", but that it can also be "your really personal hobby".

The Japanese government's English-language magazine describes ikigai the same broad way: "a broad concept, it refers to that which brings value and joy to life: from people, such as one's children or friends, to activities including work and hobbies."

Our view is a little gentler than Mogi's: as he allows, the chart can be useful as a Western career-planning tool, so long as it is not taken as the definition of the word. If your four circles do not overlap, nothing has gone wrong.

What is ikigai?

Meaning and pronunciation

The word joins iki (生き, "life; alive") and kai (甲斐, "effect"), which is voiced as gai in the compound. The Japanese dictionary Digital Daijisen defines it as 生きるに値するもの。生きていくはりあいや喜び。, roughly "something worth living for; the zest and joy of going on living".

The Oxford English Dictionary, as quoted on Wikipedia, defines it in English as "a motivating force; something or someone that gives a person a sense of purpose or a reason for living". The same page gives the English pronunciation as /ˈikiˌɡaɪ/, close to "EE-kee-guy".

Where the Japanese idea comes from

A classic Japanese book on the subject is 生きがいについて (Ikigai ni tsuite, "On ikigai") by the psychiatrist Mieko Kamiya, first published in 1966 by Misuzu Shobo. Kamiya worked as a psychiatrist at Nagashima Aiseien from 1957 to 1972; Wikipedia describes it as a sanatorium where she treated leprosy patients. Her publisher's page opens with her question: what is it that makes our everyday life feel worth living, and once a person has lost their ikigai, how do they find a new one?

Ken Mogi's five pillars

Mogi's The Little Book of Ikigai went on sale on 7 September 2017, according to its publisher, Quercus; in the United States it is titled Awakening Your Ikigai. It sets out five pillars: starting small, releasing yourself, harmony and sustainability, the joy of little things, and being in the here and now. None of them is about money or a career. Summarised from Ikigai Tribe's write-up of its interview with Mogi:

Starting small. Mogi links it to kodawari, "an approach where craftspeople take extraordinary care of the fine details of their work".
Releasing yourself. Accepting who you are. The book puts it this way: "accepting yourself is one of the most important and difficult tasks we face in our lives."
Harmony and sustainability. Getting on with other people and thinking about the wider effect of what you do: "you should show adequate consideration for other people and be mindful of the impact your actions might have on society at large."
The joy of little things. Small pleasures built into the day; Mogi's example is having favourite things, such as chocolate and coffee, soon after you get up.
Being in the here and now. Mogi's point is that children value the present because they have no definite idea of the past or future.

How to find your ikigai

It depends which ikigai you mean.

The Japanese sense is rarely worked out on paper. In Mogi's account, Japanese people are aware of the idea from childhood, and "we don't actually discuss ikigai in expressed, rule-based, algorithmic ways." The publisher's description of Mogi's book gives three people who each have one: "a cleaner of the famous shinkansen bullet train, the mother of a newborn child or a michelin-starred sushi chef."

The Western diagram is a set of career questions. Where two circles overlap, the usual version adds a label: passion, mission, vocation and profession. The labels were already attached to it by 2015, when a commenter on Winn's post asked where "the passion, mission, vocation, profession visual model" came from.

The four questions are:

What do you love?
What are you good at?
What does the world need?
What can you be paid for?

Used this way, the four questions are a starting point for thinking, not a measurement of anything.

How to apply ikigai in work and in life

The two ideas lead to different uses.

In life, the Japanese sense asks less of you. Mogi calls ikigai "a really wide spectrum of things that you love to do", so you can have more than one, and they can be small: people, a craft, a routine. Mogi's pillars describe a way of holding them: start with details, accept yourself, fit in with the people around you, notice small pleasures, and pay attention to the present.

In work, the diagram is a way of testing a job or a plan against four questions at once. If a job pays but you do not care about it, or you love something nobody pays for, the chart shows where the gap is. What it does not show is whether your life has meaning. In Mogi's account, that can sit entirely outside work.

Examples of ikigai

The Japanese government's 2022 article on ikigai visited Ogimi village in Okinawa, which it calls "a 'village of longevity'". Among the older residents it describes:

Taira Toshiko, described as a 101-year-old textile artisan who "has been making bashofu, a traditional fabric, for many years", and who was designated a Living National Treasure of Japan in 2000 for preserving the craft.
Tedokon Keiki, described as 91 years old and still growing crops.

The same article reports that Héctor García, co-author of Ikigai: The Japanese Secret to a Long and Happy Life, interviewed "more than 100 elderly residents" of Ogimi for the book.

Mogi's own examples are just as ordinary: painting a picture nobody pays for, or playing a musical instrument badly and loving it.

What you can do with this

Keep the two ideas apart. A list of what makes an ordinary day worth it is one exercise. A career plan tested against four circles is another. Both can be useful; mixing them makes the first feel like a failed version of the second.
Count unpaid things. On Mogi's account, a hobby that earns nothing, or a skill you have not mastered, can still be ikigai.
Allow more than one. The Japanese government's description covers people as well as activities, and work as well as hobbies.
Use the diagram as a check on work, not a test of a life. It is a 2014 adaptation of a purpose diagram, and it is good at showing where a job and your interests do not meet.

Sources

Marc Winn, "What is your ikigai?", The View Inside Me (14 May 2014, with a later note and reader comments, including one from Andrés Zuzunaga).
Ikigai Tribe, "Ken Mogi on what makes the Ikigai Venn Diagram wrong" (26 January 2022), transcript of Mogi's talk.
Ikigai Tribe, "Interview with Ken Mogi on the 5 Pillars of Ikigai" (podcast episode page).
Hachette India (Quercus), The Little Book of Ikigai by Ken Mogi: publisher's description; Quercus Books, ebook edition page (on-sale date).
The Government of Japan (JapanGov), "Ikigai: The Japanese Secret to a Joyful Life", Kizuna (2022).
Digital Daijisen (Shogakukan), "生き甲斐", via Kotobank.
Wikipedia, "Ikigai" (for the OED definition, pronunciation and etymology).
Misuzu Shobo, book page for 生きがいについて and author page for 神谷美恵子.

Checked September 2026. The diagram's history is taken from Winn's own post and note; Mogi's views from the transcript of his talk; the pillars from Mogi's publisher and Ikigai Tribe's interview write-up. We did not read The Little Book of Ikigai itself, so the book's lines above are as quoted by Ikigai Tribe.

Financial Education: The Rich Dad Poor Dad Blueprint for Financial Freedom ∙ Book Analysis & Review

Hacks Vitae — Tue, 31 Jan 2023 22:00:01 GMT

This is a review of what Rich Dad Poor Dad argues and what can be checked about it. It describes the book; it does not recommend any investment, and nothing on this page is financial advice. It is an independent review: Hacks Vitae is not affiliated with or endorsed by The Rich Dad Company.

I. Introduction

Rich Dad Poor Dad was first published in 1997. It is written by Robert Kiyosaki with Sharon L. Lechter, a CPA. Its full title sets out its promise: What the Rich Teach Their Kids About Money That the Poor and Middle Class Do Not.

The book is built as a story about two father figures. The Rich Dad company summarises its argument as a challenge to "conventional beliefs about money" that "advocates for financial education as a pathway to wealth." The company calls it an international bestseller. ABC News noted in 2012 that it became a New York Times best-seller "despite criticism of his personal finance tips, such as his emphasis on real estate investing."

II. The Rich Dad, Poor Dad Story

Poor dad is Kiyosaki's own father, Ralph Kiyosaki. The Honolulu Star-Bulletin described him in 2000 as "a teacher, state superintendent of schools in the late 1960s and an unsuccessful Republican candidate for lieutenant governor." The Rich Dad company describes him as "a highly educated man with a Ph.D. in Education". We could not find an independent record of an earned doctorate. The University of Hawaii's Board of Regents lists, among the honorary degrees it has conferred, "Ralph Kosaki, superintendent of education, 1991", and the university's archives hold a file titled "Honorary Degrees - Ralph Kiyosaki". His son told the Star-Bulletin that for all his public service his father "died broke", after losing his job following the election and putting his retirement money into a franchise that failed.

So "poor" in the title does not mean a low earner. Ralph Kiyosaki held one of the senior public education posts in Hawaii. The book's contrast is about attitudes to money, not about income.

Rich dad, in the Rich Dad company's account, was the father of Kiyosaki's best friend Mike, whom he met at the age of nine: "a successful entrepreneur and investor" who taught the boys about money while playing Monopoly. Kiyosaki has declined to name him. In the same 2000 interview Kiyosaki said: "I have an agreement with the family that I won't disclose who he is... but it's a true story." He added that in the book the man "asked that I say he said he died, but he's still alive."

A reader can take the story as Kiyosaki tells it, but the central figure of the book cannot be checked. Our reading: it works best as a parable, two ways of thinking about money given two faces, rather than as a documented biography.

III. The Importance of Financial Education

Kiyosaki's central complaint is that school does not teach people about money; the Rich Dad company's own tagline is "What school doesn't teach you about money." In the 2000 interview he put it this way: "what causes people to be rich or poor is their vocabulary, what they say", recalling that his father "used to say 'I'll never be rich,'" and that it became a self-fulfilling prophecy.

That is Kiyosaki's argument. What does research say about teaching people about money? Two large reviews point in different directions, and both are worth knowing.

A 2014 meta-analysis by Daniel Fernandes, John Lynch and Richard Netemeyer, covering 201 prior studies, found that interventions to improve financial literacy "explain only 0.1% of the variance in financial behaviors studied", with weaker effects in low-income samples, and that the effect decays: "even large interventions with many hours of instruction have negligible effects on behavior 20 months or more" later. They suggested a narrower role for "just-in-time" financial education tied to the specific behaviours it is meant to help.
A 2022 meta-analysis by Tim Kaiser, Annamaria Lusardi, Lukas Menkhoff and Carly Urban, covering 76 randomised experiments with over 160,000 people, found that financial education programmes have, on average, "positive causal treatment effects on financial knowledge and downstream financial behaviors", with effects "similar to those realized by educational interventions in other domains".

The later review is restricted to randomised experiments and accounts for publication bias, which is a reason to give it weight; its 2020 working-paper version reported effects "at least three times as large as the average effect documented in earlier work". The earlier one is a reason not to expect a single course to change behaviour for years. Neither tests the book's specific claims.

IV. Mindset and Attitude Towards Money

The book's contrast between the two dads is a contrast of attitudes: to security, to employment, and to learning about money. The Rich Dad company describes poor dad as "an employee his entire life" who "craved nothing more than security", and rich dad as the entrepreneur and investor who shared "crucial financial insights" with the boys.

These are the book's claims about mindset, drawn from one family's story. They are not research findings, and the book does not test them. What the story does is make a question visible: when you think about money, are you thinking about safety, or about what your money is doing?

V. Assets and Liabilities

The Rich Dad definition, in the company's own words, is "incredibly simple: an asset puts money into your pocket, while a liability takes money out of your pocket."

By that test, a car usually counts as a liability, and so does the home you live in. The Rich Dad site explains that Kiyosaki "argued that your home is not necessarily an asset if it keeps pulling money out of your pocket every month without putting any into your pocket," and that a rental property with positive cash flow is an example of an asset.

This is not how accounting defines the word. Under the International Accounting Standards Board's 2018 Conceptual Framework, an asset is "a present economic resource controlled by the entity as a result of past events", and an economic resource is "a right that has the potential to produce economic benefits". On that kind of definition, a home you own counts as an asset whether or not it produces income.

The two definitions answer different questions. The accountant's asks what you own. Kiyosaki's asks what your possessions do to your cash each month. Both can be useful, as long as you know which one you are using.

VI. The Cashflow Quadrant (From the Follow-Up Book)

The Cashflow Quadrant comes from Kiyosaki's follow-up book, Rich Dad's CASHFLOW Quadrant, which the Rich Dad store presents as the place to "go next" after Rich Dad Poor Dad. It sorts people by where their income comes from:

E: employee
S: small business owner or self-employed
B: big business owner
I: investor

In the Rich Dad company's words, "where you are is determined by where your cash comes from." It places employees and the self-employed on the left side and says "the right side is reserved for those individuals who receive their cash from businesses or investments they own." It also claims that "the Es and Ss pay the most in taxes and trade their time for money"; tax rules differ by country and change over time, so that claim depends on where you live.

VII. Building Passive Income Streams

In the Rich Dad account, the point of acquiring assets is income that keeps arriving whether or not you are working: "an asset, whether you're working or not, is something that puts money in your pocket." The company's own example is "a rental property that has a positive cash flow."

Two things are worth keeping in view. Income that needs little work to keep can still take a lot of work, money and risk to set up, and any income from an investment can fall or stop. Our reading: the book describes the destination more vividly than the risks on the way.

VIII. Caveats

The rich dad cannot be checked. Kiyosaki has said he has an agreement with the family not to name him.
Business setbacks. Rich Global LLC, one of Kiyosaki's companies, filed for Chapter 7 bankruptcy in a Wyoming court in August 2012. ABC News reported that the company had been weighed down by a lawsuit from the Learning Annex, which had helped arrange his speaking events, and that a district judge in New York awarded the Learning Annex $23.7 million. ABC described it as a corporate bankruptcy filed through one of his companies.
One family's story is not evidence. The book's claims about how rich and poor people think are drawn from two men, one of them anonymous.
Tax and property rules are local. Anything the book says about tax or property is a starting point for questions, not an answer where you live.

IX. What a Reader Can Take From It

Described, not prescribed. These are ways of using the book's ideas as questions, not instructions.

What does each thing do to monthly cash? Kiyosaki's test asks whether something puts money in each month or takes it out. The question can be useful without accepting his label.
Which definition of "asset" is in use? A home can be an asset on a balance sheet and a monthly cost in a budget at the same time.
How close is the learning to the decision? The 2014 meta-analysis found that the effect of financial education fades over time and suggested "just-in-time" learning tied to the behaviour it is meant to help; the 2022 review found positive average effects. Learning about something close to when it is used would fit both findings.
Does the idea stand without the story? The central character is anonymous, so the ideas have to be judged on their own.
Tax, law and investing questions are for a qualified professional where you live. This page describes a book; it does not tell anyone what to do with their money.

Sources

Honolulu Star-Bulletin, "Features: Robert Kiyosaki", 10 July 2000.
Rich Dad, "Robert Kiyosaki" (company biography).
University of Hawaii Board of Regents, Honorary Degrees.
University of Hawaii at Manoa Library, archives catalogue: "Honorary Degrees - Ralph Kiyosaki" (file title; University President's records, 1984-1992).
Rich Dad, "Assets vs. Liabilities: The Difference is Life Changing" and "Your Home is Not an Asset… Could it Be?".
Rich Dad, "Rich Dad Fundamentals: The CASHFLOW Quadrant", and the Rich Dad Store page for Rich Dad's CASHFLOW Quadrant.
IFRS Foundation, Conceptual Framework for Financial Reporting: Project Summary, March 2018, p. 8.
Fernandes, D., Lynch, J. G., & Netemeyer, R. G. (2014). Financial Literacy, Financial Education, and Downstream Financial Behaviors. Management Science, 60(8), 1861–1883. doi:10.1287/mnsc.2013.1849.
Kaiser, T., Lusardi, A., Menkhoff, L., & Urban, C. (2022). Financial education affects financial knowledge and downstream behaviors. Journal of Financial Economics, 145(2), 255–272. doi:10.1016/j.jfineco.2021.09.022. Published abstract (RePEc). The 2020 working paper: NBER w27057.
Susanna Kim, "'Rich Dad, Poor Dad' Author Files for Bankruptcy for His Company", ABC News, 12 October 2012.

Checked September 2026.

The Godfather Life Lessons and Quotes: A Guide to Living a Life of Power, Respect, and Leadership

Hacks Vitae — Tue, 31 Jan 2023 10:41:04 GMT

The Godfather began as Mario Puzo's 1969 novel and became three films directed by Francis Ford Coppola: The Godfather (1972), The Godfather Part II (1974) and The Godfather Part III (1990). People have been pulling life lessons out of it for decades. Below are the ones most often taken from it, each with the line it is usually tied to, who says it and in which film or book, and a note on what the story itself shows.

One thing to keep in mind throughout: every line here is fiction, spoken by gangsters in a story about what that life costs them. The lessons are our reading, not advice the films give.

Respect is Earned, Not Demanded

The film opens with Don Vito Corleone hearing requests on his daughter's wedding day. The undertaker Bonasera asks for justice for his daughter; Vito first mildly chastises him for refusing his friendship in the past, then agrees to help in exchange for some future service. Respect in this world is built from favours given and owed.

"I'm gonna make him an offer he can't refuse." — Vito Corleone, The Godfather (1972). The American Film Institute ranks it No. 2 on its list of the 100 greatest movie quotes.

The line shows the other side of that respect: the "offer" is a threat. A reader can take the useful half, that respect follows from what you do for people, and leave the half that rests on fear.

Keep Your Friends Close and Your Enemies Closer

"Keep your friends close, but your enemies closer." — Michael Corleone, The Godfather Part II (1974). No. 58 on the AFI's list.

The film's version has "but", not "and". The idea it carries is about attention: know what the people who oppose you want and are doing, rather than keeping them out of sight.

Family is Everything

At the wedding, Vito asks the singer Johnny Fontane whether he spends time with his family, and adds:

"Because a man who doesn't spend time with his family can never be a real man." — Vito Corleone, The Godfather (1972)

A study of the scene from UC Berkeley's The Godfather: Anatomy of a Film project points out that Vito looks straight at his son Sonny as he says it, directing the line more at him than at Johnny: Sonny had slipped away from the wedding to be unfaithful to his wife. The line is advice, and also a public rebuke.

Always Keep Your Word

The whole Corleone system runs on promises. On the wedding day, following a Sicilian tradition, a father cannot refuse a request. Vito promises to help Johnny Fontane get a part in a film, and keeps his word in the most brutal way: the studio head who refused Johnny wakes up with the severed head of his horse in his bed. Each favour also creates a debt: Bonasera's help comes "in exchange for some future service".

Keeping your word builds trust in any setting. The film adds a warning worth remembering: be careful what you promise in return.

Always Be in Control of Your Emotions

Coppola's films show power as calm. In the meeting with the drug dealer Sollozzo, Sonny tries to argue for the deal his father is turning down. Vito tells him never to tell anyone outside the family what he is thinking again. The Berkeley project's study of gesture notes that Vito and Michael hold their power by controlling their bodies and holding back their urges. Sonny, who acts on his temper, is the one who dies: when he hears his sister has been beaten, he races off without his bodyguards and is ambushed at a highway toll booth.

The useful part for ordinary life is simple: decide after the anger passes, not during it.

Don't Ask Questions You Don't Want to Know the Answers To

The first film ends with this question being asked. Carlo, the husband of Michael's sister, has just been strangled after Michael promised him leniency. When Kay asks Michael whether he killed Carlo, he tells her that, just this once, she may ask about his business, and answers "No." Moments later she watches Clemenza kiss Michael's ring and call him "Don Corleone" before the study door closes on her.

The film does not recommend staying in the dark. It shows what it costs to live with someone whose answers you cannot trust.

Never Go Against the Family

When his brother Fredo sides with the casino owner Moe Greene, Michael warns him never again to take sides with anyone against the family.

"It's not personal, Sonny. It's strictly business." — Michael Corleone, The Godfather (1972)

Michael says it when he volunteers to kill Sollozzo, whose men have shot his father, and the corrupt police captain McCluskey. The line is Michael's. Tom Hagen has just argued that this is business, not personal, and near the end of the film Tessio asks Tom to tell Michael it was only business. No Film School reads the moment as Michael's devotion to his father; on that reading, it is about as personal as it gets.

Revenge is a Dish Best Served Cold

"Revenge is a dish that tastes best when it is cold." — Don Corleone, in Mario Puzo's novel The Godfather (1969), chapter 28

This is the book's line. We could not find it in the dialogue of any of the three films. The saying is much older than Puzo. Its earliest known printing, according to Quote Investigator, is in Eugène Sue's 1841 French novel Mathilde, where a character presents it as a common saying. English versions followed from 1846, including a translation of Sue's novel that gives it as "revenge is very good eaten cold, as the vulgar say". The wording "best served cold" appears in Star Trek II: The Wrath of Khan (1982), where Khan calls it a Klingon proverb.

The advice buried in it, to wait until you are calm, is the same as the lesson on emotions above. The films show where revenge leads.

What to take from this

The Godfather is a story about a family business that destroys the family. By the end of Part II, Kay has told Michael she is leaving him and taking the children, and his brother Fredo has been shot on a lake by Michael's enforcer Al Neri, on Michael's orders. That is our reading of the films, and it is worth holding onto when the lines are quoted as a management guide.

The lessons that survive outside the story are the modest ones: earn respect through what you do for people, keep your promises and be careful what you promise, pay attention to people who oppose you, and make decisions after the anger has passed.

Sources

American Film Institute, "AFI's 100 Years…100 Movie Quotes", with plot summaries of The Godfather (1972) and The Godfather Part II (1974).
AFI Catalog, The Godfather Part III (release date 25 December 1990).
The Godfather: Anatomy of a Film (UC Berkeley): Hansol Jung, "A Family in Celebration, and in Transition: The Godfather's Opening Wedding Scene"; Alex Chellsen, "Never Let Your Body Show What You're Thinking: Gesture and Masculinity in The Godfather".
Shortlist, "20 pieces of wisdom from The Godfather trilogy" (speakers of the Vito and Michael lines).
No Film School, "What Michael Really Meant When He Said 'It's Strictly Business' in 'The Godfather'", 18 November 2025.
Screen Rant, "Did Fredo Know He Was Going To Be Killed In The Godfather Part II? Ending Mystery Explained", 19 March 2025.
Quote Investigator, "Revenge Is a Dish Best Eaten Cold", 5 March 2025 (Puzo, The Godfather, G. P. Putnam's Sons, 1969, chapter 28, p. 403; Sue, Mathilde, 1841; Osborne's 1846 translation).

Checked September 2026. The lines were checked against the sources above, which name the speaker and the film; the novel's line is from Quote Investigator's record of the 1969 edition. If you can show any of this wrong, with a source, we want to see it.

FRIDAY THE 13TH ∙ Origins and Superstitions

Hacks Vitae — Thu, 12 Jan 2023 13:14:51 GMT

Friday the 13th is treated as an unlucky day in parts of Europe and North America. Fridays and the number thirteen each had a bad name long before anyone put them together, and the pairing is younger than it feels: the earliest clear references are French, from 1834.

Origins of Friday the 13th and superstitions surrounding it

The folklorists at the Library of Congress's American Folklife Center have looked into this more than once. Stephen Winick's conclusion there: "the belief in Friday the 13th is not ancient or medieval, although it does have roots in earlier European beliefs about Friday and the number 13."

The earliest clear references Winick found are two French texts from 1834:

In the Revue de Paris, the Marquis de Salvo, writing about a Sicilian count who killed his daughter on a Friday the 13th: "It is always Fridays and the number 13 that bring bad luck!"
In the play Les Finesses des Gribouilles, a character says he was "born on a Friday, December 13, 1813, from which come all my misfortunes."

Winick is careful about both. The first could mean Fridays and thirteen as two separate bad omens; the second could be about that one date only. But by the middle of the 19th century the idea that any Friday the 13th was unlucky was common in France. In the 1858 play Bloqué! by Henri Chivot and Alfred Duru, a character exclaims: "I was born on a Friday the 13th!"

One small twist in the 1834 play: 13 December 1813 was a Monday, not a Friday. The character is explaining himself to a judge after being arrested and is not telling the full truth, so the impossible date may be part of the joke.

In English, an early example comes from music. The composer Gioachino Rossini died on 13 November 1868, a Friday. His biographer H. S. Edwards wrote in The Life of Rossini (London, 1869): "if it be true that, like so many other Italians, he regarded Friday as an unlucky day, and thirteen as an unlucky number, it is remarkable that on Friday, the 13th of November, he died."

The belief then crossed to America. The Library of Congress's Stephen Winick traces it through a French play, Octave Gastineau's Lélia, whose two English translations came out in 1878 and 1879, and through newspaper jokes of the 1880s that treated it as "both foolish and typically French". Clubs formed to disprove beliefs like it: the Thirteen Club of New York first met on Friday, January 13, 1882. In the late 1880s and early 1890s, references begin to show up regularly in American newspapers.

By 1907 the date was familiar enough for Thomas W. Lawson to build a Wall Street novel around it. Friday, the Thirteenth calls it "the Wall Street hoodoo-day, Friday, the thirteenth of the month". The novel is sometimes given as the origin of the superstition, but its characters already believe in it, so it shows the belief was around rather than starting it.

So as a named superstition, Friday the 13th is about two hundred years old.

Why Friday, and why thirteen

Friday. Christian tradition holds that Jesus was crucified on a Friday, and Winick suggests this may be where the idea of Friday as unlucky came from; it also made Friday a fasting day across the Christian world. In many parts of Europe, the Folklife Center notes, Fridays were traditionally a bad day to begin new projects, get married or start a journey, and fishermen often did not go out to sea on Fridays.

Thirteen. In the United States before 1900, the most common belief about thirteen was that if 13 people dine together, one of them will die. This is associated with the Last Supper, where Jesus "sat down with the twelve" (Matthew 26:20) and one of them, Judas, betrayed him. The Bible does not say who arrived or sat in what order, so there is no sense in which Judas was "the thirteenth"; the link is that 13 were at the table.

Thirteen was tied to death in other places too. In Tarot, the Death card has carried the number 13 since at least 1565, when Francesco Piscina mentions it.

The Loki story. You will often read that twelve Norse gods were at a dinner party when Loki arrived uninvited as the thirteenth guest and brought about the death of the god Baldr. The medieval sources do not tell it that way. In Snorri Sturluson's Prose Edda (in Arthur Gilchrist Brodeur's translation), Baldr dies at the Thing, an assembly, where the gods are shooting at him for sport because nothing can hurt him; Loki puts a shaft of mistletoe in the hands of the blind god Hödr and guides his aim. Winick notes that neither the Poetic nor the Prose Edda sets Baldr's death at a dinner party or gives the number of gods present when he was killed. Where the modern version came from is unclear; Winick suggests it was made by combining Baldr's death with the Last Supper's "13th guest".

Two names you will see: triskaidekaphobia, "an irrational or disproportionate fear of the number 13", and paraskevidekatriaphobia, fear of Friday the 13th.

The Knights Templar story

The Templar explanation starts from a real event. On the night of Friday, 13 October 1307, the Templars in France were arrested on the orders of King Philip IV. The order was not ended that day: Pope Clement V dissolved it by a bull of 22 March 1312, and Jacques de Molay, its last grand master, was burned in Paris in 1314. There is more on the orders themselves in our overview of the crusader knights.

What is missing is a link between those arrests and the superstition. According to Winick, "no one seems to have connected this historical event to the belief in Friday the 13th until the 20th century", and the theory "was popularized by Dan Brown's novel The Da Vinci Code". Until evidence from the Middle Ages turns up, he writes, it "seems more like a modern guess." The two French references from 1834 are about a Sicilian count and a man's birthday; neither mentions the Templars.

Not unlucky everywhere

Which day counts as unlucky depends on where you are. In Spanish-speaking countries and Greece, the Folklife Center notes, Tuesday the 13th may be considered unlucky; in Italy it is Friday the 17th. And thirteen is not bad everywhere: in many parts of Europe it is a lucky number, and in the United States it has positive associations with the 13 original colonies, which is why the Great Seal has 13 stars, 13 stripes of white and red, and 13 arrows.

What to take from this

Friday the 13th is a good example of a habit worth having with any old custom: put the dates in order. Find the earliest dated appearance of the belief, then ask whether the explanation offered for it shows up before or after that. Here the belief first appears in 1834, while the Templar link and the Loki dinner party turn up much later.

It also shows how a real date can make a loose connection feel solid. The Templars really were arrested on a Friday the 13th. That fact is true; the link to the superstition is the part no one has found evidence for.

Sources

Stephen Winick, "On the Possible Origins of Friday the Thirteenth: Metafolklore, Fear, and Fun", Folklife Today, American Folklife Center, Library of Congress, 12 December 2024 (read in the Internet Archive's copy).
Stephanie Hall, "Who's Afraid of Friday the Thirteenth?", Folklife Today, Library of Congress, 12 January 2017, with Stephen Winick's reply in the comments on the 1813 date (read in the Internet Archive's copy).
H. Sutherland Edwards, The Life of Rossini (London: Hurst and Blackett, 1869), Project Gutenberg #45705.
English National Opera, "Gioachino Rossini" (born Pesaro 29 February 1792; died Passy 13 November 1868).
Thomas W. Lawson, Friday, the Thirteenth (1907), Project Gutenberg #12345.
Matthew 26:20–25, King James Version, Bible Gateway.
Snorri Sturluson, The Prose Edda, trans. Arthur Gilchrist Brodeur, Gylfaginning, the death of Baldr (sacred-texts.com).
"Templars", Encyclopaedia Britannica, 11th edition (1911), Wikisource.
"Knights Templars", Catholic Encyclopedia (1912), New Advent.
"Triskaidekaphobia", Dictionary.com.
Gaillard Hunt, The History of the Seal of the United States (1909), Project Gutenberg #73844 (the Act of Congress of 20 June 1782).

Checked September 2026. The French references, the Rossini biography, Lawson's novel and the Templar dates were read in the sources above; the weekday of 13 December 1813 was checked against the calendar. If you can show any of this wrong, with a source, we want to see it.

Stoicism Life Hacks ∙ What is Stoicism? How can I be Stoic?

Hacks Vitae — Thu, 12 Jan 2023 10:22:09 GMT

"Men are disturbed not by things, but by the views which they take of things." — Epictetus, Enchiridion 5 (tr. Thomas Wentworth Higginson)

What is Stoicism?

Stoicism is an ancient Greek school of philosophy that the Romans later took up. Its central claim is about what makes a life good. For the Stoics, virtue "is necessary and sufficient for happiness", in the Stanford Encyclopedia of Philosophy's summary: things like health, wealth and reputation are neither good nor bad in themselves. Only virtue is good.

That is why the Stoic advice most people meet is about attention and judgement: noticing which things are up to you, and how much of what upsets you comes from your view of an event rather than the event itself.

Where it came from

The school was founded around 300 BC by Zeno of Citium, who had studied with the Cynic Crates. Its name comes from the Stoa Poikilê, a porch in the agora at Athens decorated with mural paintings, where the first Stoics met and lectured.

Almost nothing of the early Stoics survives whole. The three voices most people read today are Roman-era, and very different from one another:

Epictetus, a teacher who spent part of his life as the slave of an official at Nero's court. His teaching survives through his student Arrian, in the Greek Discourses and a short digest of them, the Enchiridion or Handbook.
Seneca, Nero's tutor and later one of his advisers, who wrote letters, essays and tragedies.
Marcus Aurelius, the second-century emperor, whose Meditations were, as most scholars now read them, written for his own moral improvement.

The ideas behind most Stoic advice

Most modern Stoic advice rewords a handful of ideas.

Some things are up to us, and some are not. The Enchiridion opens with it: "There are things which are within our power, and there are things which are beyond our power." Within our power, it goes on, are "opinion, aim, desire, aversion"; beyond it are "body, property, reputation, office".
The virtues. In Diogenes Laertius's account of the Stoics (tr. A. A. Long and D. N. Sedley), the virtues are "prudence, justice, courage, moderation and the rest", and the Stoics held that they are the only true goods.
Picturing loss. Epictetus suggests reminding yourself what the things you love really are: "if you embrace your child or your wife, that you embrace a mortal—and thus, if either of them dies, you can bear it." Seneca makes the same point in Letter 98, about regarding "children, or wife, or property" with the idea that we will not necessarily have them always.
The judgement behind the feeling. Marcus Aurelius, 8.47: "If thou art pained by any external thing, it is not this thing that disturbs thee, but thy own judgment about it."
Remembering death. Marcus, 4.17: "Do not act as if thou wert going to live ten thousand years. Death hangs over thee. While thou livest, while it is in thy power, be good."

A line often credited to Marcus, "The happiness of your life depends upon the quality of your thoughts", comes from Jeremy Collier's free translation of 1701, at Meditations 3.9, where George Long's translation begins: "Reverence the faculty which produces opinion." Long's nearest line is 5.16: "Such as are thy habitual thoughts, such also will be the character of thy mind; for the soul is dyed by the thoughts."

Two labels often attached to Stoicism

Amor fati, "love of fate", is a phrase from Nietzsche. In The Gay Science (1882), §276, he wrote: "Amor fati: let that henceforth be my love!" The Latin words do not appear in Seneca's surviving works, and Epictetus taught in Greek. The attitude is close to Epictetus, though: "Demand not that events should happen as you wish; but wish them to happen as they do happen, and you will go on well" (Enchiridion 8).
"The inner citadel" is the title of Pierre Hadot's book on the Meditations. The image is Marcus's own: "the mind which is free from passions is a citadel" (Meditations 8.48).

A Stoic day, in practice

The texts describe a few small habits. None of them needs special equipment, and people take what is useful and leave the rest.

In the morning, expect friction. Marcus, 2.1: "Begin the morning by saying to thyself, I shall meet with the busybody, the ungrateful, arrogant, deceitful, envious, unsocial." He goes on to say that none of them can injure him.
Sort the day's worries into what is within your power and what is not, as the first line of the Enchiridion does.
Picture losing something you value, in the way Epictetus describes with the favourite cup: "it is but a cup of which you are fond of—for thus, if it is broken, you can bear it."
When something stings, look for the judgement underneath it (Meditations 8.47), and ask whether it holds.
Choose company with care. Seneca, Letter 7: "Associate with those who will make a better man of you."
Review the day at night. In On Anger 3.36, Seneca describes the custom of the philosopher Sextius, who asked himself each evening: "What bad habit of yours have you cured to-day? What vice have you checked? In what respect are you better?" Seneca says he does the same, once the lamp is out and his wife has fallen silent.

Seneca also names the trap the practice is meant to avoid (Letter 98): "It is tragic for the soul to be apprehensive of the future and wretched in anticipation of wretchedness, consumed with an anxious desire that the objects which give pleasure may remain in its possession to the very end."

Why it keeps coming back

Stoicism has a real place in the history of modern talking therapy. Albert Ellis, who founded rational emotive behaviour therapy, named Epictetus as a forerunner, and later said he had brought Epictetus "out of near-obscurity". Aaron Beck, the founder of cognitive therapy, wrote in his first book on it that its "philosophical underpinnings go back thousands of years, certainly to the time of the Stoics".

That lineage is also a limit. Stoicism is a philosophy, not a treatment. It can help with the ordinary friction of life; it is not a substitute for care if you are struggling with your mental health.

Where to start reading

All three are free online in older translations:

Marcus Aurelius, Meditations, tr. George Long (Project Gutenberg).
Epictetus, Enchiridion, tr. Thomas Wentworth Higginson (Project Gutenberg). It is short, and it is an abridgment: the Stanford Encyclopedia calls it "a much attenuated account" of the Discourses, which are the fuller record.
Seneca, Moral Letters to Lucilius, tr. Richard M. Gummere (Wikisource), and On Anger, tr. Aubrey Stewart (Wikisource).

Our verified Stoic quote of the day draws only on these translations, with a book and section number for every line.

Sources

Epictetus, The Enchiridion, tr. Thomas Wentworth Higginson (Project Gutenberg eBook 45109), sections 1, 3, 5 and 8.
Marcus Aurelius, Meditations, tr. George Long (Project Gutenberg eBook 15877), 2.1, 3.9, 4.17, 5.16, 8.47 and 8.48.
Seneca, Moral Letters to Lucilius, tr. Richard M. Gummere (Wikisource), Letters 7 and 98; Of Anger, Book III, tr. Aubrey Stewart (1900; Wikisource), section 36.
Stanford Encyclopedia of Philosophy: "Stoicism", "Epictetus", "Seneca" and "Marcus Aurelius"; the Diogenes Laertius passage on the virtues is quoted there in A. A. Long and D. N. Sedley's translation (The Hellenistic Philosophers, 1987, 58A).
Marcus Aurelius, Emperor Marcus Antoninus His Conversation with Himself, tr. Jeremy Collier (London, 1701), Book III, section IX (Internet Archive scan).
Friedrich Nietzsche, The Joyful Wisdom (The Gay Science), tr. Thomas Common (Project Gutenberg eBook 52881), §276; Seneca's Latin works as published on The Latin Library, searched for amor fati.
Pierre Hadot, The Inner Citadel: The Meditations of Marcus Aurelius, tr. Michael Chase (Harvard University Press).
Andrea E. Cavanna and colleagues, "The Western origins of mindfulness therapy in ancient Rome", Neurological Sciences (2023), for Ellis and Beck.

Checked September 2026. Every quotation on this page was matched against the translation named beside it.