Mental Models from Philosophy
11 Thinking Tools Checked Against the Texts

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.

Related: Mental models, checked · Deciding under uncertainty · Where the trolley problem comes from