THE ARTICLE · 9 MIN
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.
Related: Mansa Musa’s gold · The Library of Alexandria was not destroyed in one fire
- history
- mathematics
- myths
- middle east