A history of logic in Islam

1A tradition, not only a language

Historians use ‘Arabic logic’ for a tradition rather than simply for logic written in Arabic. In the definition of Tony Street and Nadja Germann (Stanford Encyclopedia of Philosophy), it is the tradition rooted in the Greek texts translated into Arabic from the eighth century, which settled on a set of technical terms and a set of problems and has continued ‘from the mid-eighth century down to today’. Works in Persian, Turkish or Urdu that use its terms and problems belong to it; the modern logic in the line of Frege, taught in Arabic at modern universities, does not.

Khaled El-Rouayheb’s survey, The Development of Arabic Logic (1200–1800), divides the later history into periods and regions: up to 1200; 1200–1350; 1350–1600, with separate Eastern and Western traditions; and 1600–1800, with Iranian, Indo-Muslim, Ottoman Turkish, North African and Christian Arabic traditions. This article follows a similar order.

2The Greek inheritance

Aristotle’s logical works were gathered after his death into a collection later called the Organon (‘instrument’). The late antique teachers of Alexandria read it in a fixed order, preceded by Porphyry’s Isagoge and followed, unusually, by the Rhetoric and the Poetics, which they counted as logic. It was this extended Organon, in the Alexandrian order, that the Arabic tradition eventually received.

Before the Arabic translations, Syriac-speaking Christians had taught a shorter course: the Isagoge, the Categories, On Interpretation and the first seven chapters of the Prior Analytics. According to Street and Germann this teaching continued undisturbed through the Arab conquests and the Umayyad period, but drew little interest from the new rulers. The story of how the Isagoge itself travelled from Greek into Syriac and Arabic is told in a separate article on The Mantiqi (‘How the Isagoge travelled’).

3The translations, eighth to tenth centuries

Interest grew under the Abbasids (from 750). An early Arabic logic treatise, probably based on a late antique introduction that reached its author through Syriac and Pahlavi, is attributed to Ibn al-Muqaffaʿ (executed around 756); it covers the Isagoge, the Categories, On Interpretation and the beginning of the Prior Analytics. The caliph al-Mahdī (r. 775–785) commissioned translations of the Topics and the Sophistical Refutations. By the 830s a circle of translators was gathered around al-Kindī (d. ca. 870), who wrote a short overview of the whole Organon. Dimitri Gutas’s Greek Thought, Arabic Culture (1998) remains the standard account of this ‘translation movement’ and its Abbasid setting.

From about the 850s Ḥunayn ibn Isḥāq (d. 873) and his son Isḥāq ibn Ḥunayn (d. 910), Syriac Christians, produced complete translations of Organon books, usually by way of Syriac. Isḥāq translated On Interpretation and revised the translations of the Topics and the Rhetoric; father and son both worked on the Posterior Analytics. Early in the tenth century Abū Bishr Mattā ibn Yūnus (d. 940), called ‘the chief of the logicians’, added translations of the Poetics and the Posterior Analytics. The TDV İslâm Ansiklopedisi names further translators, among them Yaḥyā ibn al-Biṭrīq, Abū ʿUthmān al-Dimashqī and Yaḥyā ibn ʿAdī (d. 974). The Baghdad school’s crowning achievement, in Street and Germann’s words, was ‘a superb and heavily glossed translation of the Organon’; it has been edited twice in modern times, by ʿAbd al-Raḥmān Badawī (Manṭiq Arisṭū, 1948–52) and by Jabre (1999).

Not everyone welcomed the newcomer. In a famous debate held before a vizier, the grammarian Abū Saʿīd al-Sīrāfī attacked Abū Bishr’s claim that speakers of Arabic needed Greek logic; Abū Bishr answered that logic deals with meanings and grammar only with expressions. Street and Germann note that the encounter at least shows that some were sceptical of Aristotelian logic, and that others went further and thought it impious because of its link with metaphysics.

4Al-Fārābī and the Baghdad Peripatetics

The Baghdad philosophers of the tenth century, Christians and Muslims, saw themselves as restoring Aristotle’s true teaching. Their outstanding figure was Abū Naṣr al-Fārābī (d. 339/950), whom Ibn Sīnā called ‘the Second Teacher’ after Aristotle. He wrote commentaries on the books of the Organon, many now lost, and shorter works such as the Enumeration of the Sciences (Iḥṣāʾ al-ʿulūm) and a treatise on the terms used in logic. He gave the classic answer to the grammarians: logic is to thought what grammar is to language, but its rules are common to all peoples. He also argued that logic was needed to analyse the arguments of theologians and jurists, and wrote a short treatise interpreting their reasoning as syllogisms.

His work aimed, in Robert Wisnovsky’s phrase, at a ‘Lesser Harmony’: a single consistent doctrine drawn out of Aristotle’s sometimes conflicting texts. This exegetical tradition continued after him, in Baghdad and then in al-Andalus, where its last great representative was Ibn Rushd (Averroes, d. 1198). His logical work was translated into Hebrew and Latin but, with few exceptions, neglected by later Arabic logicians; it continued to be read mainly in North Africa.

5Ibn Sīnā: a new system

The decisive turn came with Ibn Sīnā (d. 428/1037), working far to the east, in Khurasan. He treated Aristotle’s logic not as a text to be interpreted but as a system to be assessed and repaired, setting aside parts of Aristotle’s modal logic as mere ‘tests’ for students or as mistakes. His logic survives in several works: the vast Healing (al-Shifāʾ, from the 1020s), the shorter Salvation (al-Najāt), the Persian Dānishnāma, and his shortest and probably last exposition, Pointers and Reminders (al-Ishārāt wa-l-tanbīhāt, 1030s), which arranged the material in a new way that shaped later logic.

Among his innovations Street and Germann list a temporal reading of ‘absolute’ propositions, new attention to propositions of the form ‘every J is B while J’, and a new division of syllogisms into connective (iqtirānī) and repetitive (istithnāʾī) instead of categorical and hypothetical. Riccardo Strobino calls him ‘the most influential logician in the Arabic tradition’: every later logician confronted him, as follower or critic, and Aristotle almost vanished from view, so that Arabic logic divides into pre-Avicennan and post-Avicennan. The Pointers became, in Ibn Taymiyya’s words, ‘the Qurʾān of the philosophers’.

6The twelfth century: into the madrasa

The eleventh century saw the rise of the madrasa, and al-Ghazālī (d. 505/1111), who taught at the most prestigious of them, made logic acceptable there. He wrote handbooks of logic, argued in Miʿyār al-ʿilm that legal reasoning differs from philosophical reasoning only in where it takes its premises from, and opened his late work on legal theory, al-Mustaṣfā, with an introduction to logic. Ibn Khaldūn later wrote that since al-Ghazālī and Fakhr al-Dīn al-Rāzī scholars had been ‘more lenient’ towards logic. Meanwhile logicians such as ʿUmar ibn Sahlān al-Sāwī (d. 1148) began to rework Ibn Sīnā’s modal syllogistic. (The debate about whether logic may be studied is the subject of a separate article.)

71200–1350: the great age of the handbooks

The thirteenth century was the most creative period of Arabic logic. El-Rouayheb has observed that logic was then more thoroughly detached from commentary on older texts than at any time before or since; the major works were independent treatises, and many of the textbooks taught for centuries afterwards come from this period.

Ibn Khaldūn named two men as authors of the new way: Fakhr al-Dīn al-Rāzī (d. 606/1210), who criticised Ibn Sīnā sharply in works such as the Mulakhkhaṣ and the Lubāb al-Ishārāt, and Afḍal al-Dīn al-Khūnajī (d. 646/1248–9), ‘on whom Eastern scholars rely even now’. ‘The books and ways of the ancients’, he added, ‘have been abandoned, as though they had never been.’ Al-Khūnajī’s Kashf al-asrār ʿan ghawāmiḍ al-afkār devotes, by El-Rouayheb’s count, about seventy per cent of its length to modal and hypothetical propositions and syllogisms; it also held that the subject of logic is ‘conceptions and assents’.

Al-Khūnajī’s work set the agenda for a group of logicians connected with the observatory founded at Maragha in 1259: Naṣīr al-Dīn al-Ṭūsī (d. 1274), who defended Ibn Sīnā; Athīr al-Dīn al-Abharī (d. 663/1265); and al-Abharī’s student Najm al-Dīn al-Kātibī (d. 675/1276–7). Al-Abharī’s Īsāghūjī and al-Kātibī’s al-Risāla al-Shamsiyya, written for the Ilkhanid vizier Shams al-Dīn al-Juwaynī, became the two mainstays of madrasa teaching. Sirāj al-Dīn al-Urmawī (d. 1283), whom the TDV calls the most important logician of his century, wrote Maṭāliʿ al-anwār, and Shams al-Dīn al-Samarqandī wrote Qisṭās al-afkār, which replaced the Topics and Sophistical Refutations with a section on the etiquette of debate. The sources differ on al-Samarqandī’s death: around 1310 according to Street and Germann, 1322 in El-Rouayheb’s survey.

81350–1600: commentaries, glosses and regional paths

From the fourteenth century most logic was written as commentary (sharḥ) and gloss (ḥāshiya) on the thirteenth-century handbooks. Quṭb al-Dīn al-Rāzī al-Taḥtānī (d. 766/1365) wrote the standard commentaries on the Shamsiyya and on the Maṭāliʿ. Saʿd al-Dīn al-Taftāzānī (d. 792/1390) wrote the compact Tahdhīb al-manṭiq; El-Rouayheb treats him as a ‘verifier’ (muḥaqqiq) with original views, not a sterile commentator. Al-Sayyid al-Sharīf al-Jurjānī (d. 816/1413) glossed Quṭb al-Dīn’s commentaries and wrote two Persian primers, the Ṣughrā and the Kubrā. Jalāl al-Dīn al-Dawānī (d. 908/1502) left an unfinished commentary on the Tahdhīb that was widely studied in Ottoman Turkey and Mughal India; his glosses on a commentary on al-Ṭūsī’s Tajrīd drew objections from his Shirazi contemporary Ṣadr al-Dīn (al-Dashtakī), and the two answered each other in several further glosses.

In the Islamic East, El-Rouayheb finds a shift of interest after the fourteenth century away from modal and hypothetical syllogisms and towards semantics, rhetoric and philosophical theology: the later parts of Quṭb al-Dīn’s commentary on the Shamsiyya, on conversion and the syllogism, ‘appear not to have elicited a single gloss after the fourteenth century’. The shift was much less marked in North Africa. There, Ibn Rushd was still read until the end of the fourteenth century, al-Khūnajī’s short Jumal was commented on (by Ibn Khaldūn’s teacher al-Sharīf al-Tilimsānī, d. 1370), and new handbooks were written, above all the Mukhtaṣar of the theologian Muḥammad ibn Yūsuf al-Sanūsī (d. 895/1490). The Algerian ʿAbd al-Raḥmān al-Akhḍarī wrote al-Sullam al-munawraq, a verse primer traditionally described as a versification of the Īsāghūjī; his death is dated 953/1546 by El-Rouayheb and 983/1575 in the TDV and in al-Azhar’s textbook.

91600–1800: empires and regions

In the early modern period logical traditions followed the borders of the great empires. In Ottoman lands logic was taught intensively; El-Rouayheb notes that Turkish libraries hold more than four thousand manuscripts on logic copied between 1300 and 1800, and that Ottoman logicians after 1600 did impressive work on relational syllogisms. In Iran, Street and Germann note, a late fifteenth-century dispute between two leading scholars made it respectable to prefer the ancients to the moderns, and El-Rouayheb finds in the Safavid period, the age of Mīr Dāmād and Mullā Ṣadrā, efforts to reconnect logic with Aristotelian philosophy. In India logic was a central part of the Dars-i Niẓāmī curriculum, and Muḥibb Allāh al-Bihārī (d. 1119/1707) wrote Sullam al-ʿulūm, which Asad Q. Ahmed calls the most advanced logic textbook of Muslim South Asia; the Khayrābādī scholars of the eighteenth and nineteenth centuries continued this line (see the articles on al-Mirqāt).

In North Africa al-Ḥasan al-Yūsī (d. 1102/1692) commented on al-Sanūsī’s handbook and defended logic against its critics. His works were widely read in the Maghrib and taught in Egypt, where students of that line, Aḥmad al-Mallawī (d. 1181/1767) and Aḥmad al-Damanhūrī (d. 1192/1778), wrote much-read commentaries on al-Akhḍarī’s Sullam; al-Mallawī’s shorter commentary became, in the words of al-Azhar’s textbook, a mainstay of teaching at al-Azhar. Among Arabic-speaking Christians, some of whom trained in Europe, Buṭrus al-Tūlāwī (d. 1745) wrote an introduction to logic in 1688 that gives al-Khūnajī’s definition of the syllogism, a sign of a meeting of traditions that, Street and Germann note, has still to be studied.

10The nineteenth and twentieth centuries

Logic remained a madrasa subject everywhere. In Egypt three rectors of al-Azhar, Ḥasan al-ʿAṭṭār (d. 1250/1834–5), Ḥasan al-Quwaysinī (d. 1255/1839) and Ibrāhīm al-Bājūrī (d. 1276/1860), wrote on logic, and printing presses in Būlāq, Cairo, Istanbul and India issued the standard texts with their commentaries and glosses. In Shīʿī seminaries, al-Ṭūsī’s Tajrīd with the commentary of al-ʿAllāma al-Ḥillī was printed in the late nineteenth century and came into use as an introduction to logic.

New pressures also appeared. The Salafiyya movement of the nineteenth and twentieth centuries took up Ibn Taymiyya’s rejection of logic. European logic arrived in schools and universities: the TDV notes that Ottoman Turkish logic manuals multiplied from the second half of the nineteenth century, still based mainly on the Īsāghūjī, and that Ali Sedad’s Mîzânü’l-ukūl (1885) was the first Turkish book to take in the new European currents; modern symbolic logic was taught at Istanbul University from 1938.

The older tradition has nevertheless continued. Al-Azhar’s secondary schools today teach logic from a simplified version of al-Akhḍarī’s Sullam with al-Mallawī’s commentary; and, as the articles on the Īsāghūjī and al-Mirqāt in the madrasa describe, the madrasas of the Indian subcontinent still read the Īsāghūjī, al-Mirqāt and the Tahdhīb. How this long history has been studied, and how historians’ views of a supposed ‘decline’ after the thirteenth century have changed, is the subject of ‘Modern study of Arabic logic’.