What is manṭiq?

1The word

The Arabic word manṭiq comes from the root n-ṭ-q, ‘to speak’; nuṭq is speech. The TDV İslâm Ansiklopedisi notes that the word matches the Greek logos, which means both reason and speech, and that the Qurʾān uses manṭiq for the ‘speech of the birds’ that Solomon was taught (al-Naml 27:16). Al-Fārābī (d. 339/950) distinguished three senses of nuṭq: the faculty by which we grasp intelligible things, the intelligible things formed in the soul (an ‘inner speech’), and the expression of thought in language (‘outer speech’). Logic takes its name from the first two.

The science has other names as well. The TDV article lists titles used for logic books such as ʿilm al-mīzān (‘the science of the balance’), miʿyār al-ʿilm (‘the yardstick of knowledge’), miftāḥ al-funūn (‘the key of the arts’) and al-qisṭās al-mustaqīm (‘the straight scale’). The present-day logic textbook of al-Azhar’s secondary schools adds miʿyār al-ʿulūm, mīzān al-ʿulūm, ʿilm al-istidlāl (‘the science of inference’) and qānūn al-fikr (‘the law of thought’), and remarks, in the traditional manner, that many names point to the nobility of the thing named.

2The classic definition

The definition that later handbooks repeat is short: logic is āla qānūniyya taʿṣimu murāʿātuhā al-dhihna ʿan al-khaṭaʾ fī al-fikr, ‘a canonical instrument whose observance protects the mind from error in thinking’. It is quoted, for example, by al-Suyūṭī (d. 911/1505) in his little dictionary of the sciences, Muʿjam maqālīd al-ʿulūm, and by al-Tahānawī in his encyclopaedia of technical terms, Kashshāf iṣṭilāḥāt al-funūn, who explains it word by word: ‘instrument’ is the genus, and ‘canonical’ (that is, consisting of general rules) is the differentia.

Each part of the definition does work. ‘Instrument’ means that logic is not studied for its own sake but for the sake of something else; al-Tahānawī says plainly that logic ‘in itself is not intended’. ‘Canonical’ (qānūniyya) means that it consists of general rules (qawānīn), which the Azhar textbook glosses as ‘universal rules’. ‘Whose observance protects the mind’ means that knowing the rules is not enough: they protect only when they are actually kept. And ‘error in thinking’ (fikr) limits the protection to reasoning, which the same textbook defines as ‘arranging things known by conception and by assent so as to reach something unknown’.

The idea of logic as a ‘canonical instrument’ goes back to Ibn Sīnā (d. 428/1037). Riccardo Strobino, in the Stanford Encyclopedia of Philosophy, notes that Ibn Sīnā calls logic a normative or, literally, ‘canonical’ instrument in his Pointers and Reminders (al-Ishārāt wa-l-tanbīhāt), while also treating it as a science with a subject of its own. Other handbooks put the same idea differently. According to the TDV, Quṭb al-Dīn al-Rāzī (d. 766/1365) in his commentary on the Shamsiyya describes logic as the science that leads from the known to the unknown, and al-Sharīf al-Jurjānī (d. 816/1413) in his Taʿrīfāt as a science whose rules, when followed, protect the mind from error. Ṣiddīq Ḥasan Khān’s Abjad al-ʿulūm, a nineteenth-century compilation drawing on older reference works, defines it as ‘a science from which one learns how to acquire unknown conceptions and assents from known ones’.

Al-Akhḍarī put the definition into the opening of his verse primer, al-Sullam al-munawraq: ‘Logic stands to the mind as grammar stands to the tongue; it protects thought from the error of going astray, and lifts the veil from subtle understanding.’ The comparison with grammar is older. Al-Fārābī wrote that the art of logic is to the intellect and its objects what grammar is to language and its expressions, with the difference that grammar gives rules for one people’s language, while logic gives rules common to all; he also compared logic to the ruler, compass and plumb line of a builder.

3What logic is about

Every science has a subject (mawḍūʿ), the thing whose properties it investigates: the body for medicine, the word for grammar. What is the subject of logic? Two answers circulated, and the handbooks often report both.

The first is Ibn Sīnā’s: the subject of logic is the ‘secondary intelligibles’ (al-maʿqūlāt al-thāniya), considered insofar as they lead from the known to the unknown. First-order notions are things like ‘human’, ‘animal’ or ‘stone’. When the mind works with them, they acquire properties that exist only in the mind: being a subject or a predicate, a genus or a species, a universal, a proposition, a premise. These second-order notions are what the logician studies. Fakhr al-Dīn al-Rāzī (d. 606/1210) explained it this way: a person first conceives the realities of things and then qualifies some by others; the properties that belong to a notion only after it is known are ‘second-order’, and logic studies them insofar as they allow a correct passage from the known to the unknown.

The second answer is that of Afḍal al-Dīn al-Khūnajī (d. 646/1248–9): the subject of logic is ‘conceptions and assents’ (al-taṣawwurāt wa-l-taṣdīqāt), because the logician also studies properties of first-order notions, such as being essential or accidental. As Tony Street and Nadja Germann explain, al-Kātibī and al-Abharī followed al-Khūnajī, while Naṣīr al-Dīn al-Ṭūsī defended Ibn Sīnā’s view. Al-Tahānawī records both: ‘it is said that its subject is conceptions and assents’, while the verifying scholars hold that it is the secondary intelligibles insofar as they lead to the unknown. Modern madrasa textbooks usually take the second, simpler line. The Azhar textbook gives the subject as ‘things known by conception and by assent (al-maʿlūmāt al-taṣawwuriyya wa-l-taṣdīqiyya), insofar as they lead to the unknown’, with the examples ‘human’ (a conception) and ‘Ḥamza has passed’ (an assent).

4Aim and benefit

The aim (ghāya) of logic, in the words of Ṣiddīq Ḥasan Khān’s compilation, is ‘protecting the mind from error in thinking’, and its benefit (fāʾida) is ‘getting things right in all the sciences’. Al-Fārābī called logic ‘the chief of the sciences’ (raʾīs al-ʿulūm), because it judges the soundness and weakness of every science; Ibn Sīnā called it ‘the servant of the sciences’ (khādim al-ʿulūm), because it is a tool and not an end. The two titles do not conflict: logic rules the sciences as a measure and serves them as a tool.

Why not rely on natural intelligence? Ibn Sīnā’s answer, as summarised in the TDV, is that natural intelligence may hit on correct conclusions, but only by chance; it cannot guarantee that it will not err. The Azhar textbook lists three benefits for students today: protection of the mind from error, the ability to answer opponents who play with words and terms, and the training of a critical judgement that weighs things correctly.

5The ten principles

Madrasa textbooks often begin a science by listing ten things a student should know about it before starting, the ‘ten principles’ (al-mabādiʾ al-ʿashara). The best-known statement is a short poem by al-Ṣabbān, quoted in the Azhar textbook: ‘The principles of every art are ten: the definition, the subject, then the fruit; its relation, its merit and its founder; the name, the source, the Lawgiver’s ruling; its questions. Some are content with some of them; whoever knows them all has gained honour.’ The list was used for every science, not only logic; a Shāfiʿī law book, for instance, quotes a similar verse by al-Khuḍarī and applies the ten to law.

For logic, the Azhar textbook fills in the ten as follows. Definition (ḥadd): a canon whose observance protects the mind from error in thinking. Subject (mawḍūʿ): things known by conception and assent, insofar as they lead to what is unknown. Fruit (thamara): protection from error. Relation (nisba) to other sciences: distinctness (tabāyun), since its questions differ from theirs. Merit (faḍl): it is among the noblest sciences because its benefit is the most general, every science consisting of conceptions and assents. Founder (wāḍiʿ): Aristotle, who ordered and refined what earlier thinkers such as Socrates and Plato had done only in scattered pieces. Name (ism): ʿilm al-manṭiq and the other names above. Source (istimdād): reason. Ruling (ḥukm): disputed, in three positions (see ‘Should a Muslim study logic?’). Questions (masāʾil): its propositions, such as those about definitions and their kinds, and about propositions and their rules.

The ‘ruling’ is the only principle that is a religious judgement rather than a description of the science. That it was included at all reflects the long debate over the study of logic, which is the subject of a separate article.

6Two halves: conception and assent

All knowledge, according to Ibn Sīnā, is either conception (taṣawwur) or assent (taṣdīq). Conception is grasping the meaning of something without judging it true or false, as when one hears the word ‘human’ and understands it. Assent is judging that something is so, as when one hears ‘every whiteness is an accident’ and not only understands it but accepts it. Conception that is not immediate is gained through a definition (ḥadd) or a description (rasm); assent that is not immediate is gained through a proof, above all the syllogism (qiyās).

From this division comes the two-part shape of almost every Arabic logic book after Ibn Sīnā: a first part on what leads to conception (utterances and their meanings, universals, the five predicables, definition) and a second on what leads to assent (propositions, their contradiction and conversion, syllogisms of the various kinds). Street and Germann describe it as the structure of all Avicennan treatises: ‘a first section deals with definition, which conduces to conception, a second with proof, which conduces to assent.’

Al-Kātibī’s al-Risāla al-Shamsiyya, one of the most widely read handbooks, shows the pattern. According to the TDV, it has an introduction on the nature of knowledge, the need for logic and its subject; a first treatise on utterances, universals and definition; a second on propositions; a third on the syllogism; and a conclusion on the ‘matters’ of syllogisms and the parts of the sciences. The Īsāghūjī follows the same order in miniature. Older Arabic writers had mapped the same division onto the books of the Organon: the TDV counts Porphyry’s Isagoge and Aristotle’s Categories under conception and the remaining seven books under assent.

7The five arts

A syllogism can be valid in form and still weak in content. Logicians therefore sorted arguments by the kind of premises they use into ‘the five arts’ (al-ṣināʿāt al-khams), a term the TDV thinks al-Fārābī was the first to use: demonstration (burhān), built on certain premises; dialectic (jadal), on widely accepted or conceded ones; rhetoric (khiṭāba), on accepted and probable ones; poetics (shiʿr), on premises that move the imagination; and sophistry (mughālaṭa), on premises that only resemble the true ones.

In the Greek scheme these five corresponded to separate books of the Organon (Posterior Analytics, Topics, Rhetoric, Poetics, Sophistical Refutations). In the later Arabic handbooks they shrank to a closing chapter. Ibn Khaldūn (d. 1406) observed that later scholars treated the books on the ‘matter’ of arguments ‘as though they had never been’, and Street and Germann show how other disciplines took over their work: the etiquette of debate (ādāb al-baḥth), introduced as a section of logic by Shams al-Dīn al-Samarqandī, replaced the Topics and the Sophistical Refutations, and the rhetorical sciences (ʿilm al-maʿānī wa-l-bayān) took the place of the Rhetoric and the Poetics. Khaled El-Rouayheb describes the same development as logic being ‘naturalised’ in the madrasa.

8Among the instrumental sciences

Whether logic is a part of philosophy or only its tool was debated; the TDV notes that Ibn Sīnā saw no contradiction in calling it both, and that over time the view of logic as a tool for the other sciences prevailed. Al-Tahānawī places it among the ‘instrumental sciences’ (al-ʿulūm al-āliyya). Ibn Khaldūn explained the distinction in his Muqaddima: some sciences are wanted for their own sake, such as the religious sciences, physics and metaphysics; others are auxiliary, such as Arabic philology and arithmetic for the religious sciences, and logic, which serves philosophy and often also theology and legal theory. Because they are only means, he advised, they should not be studied at too great length.

In the madrasa logic took its place beside the other instrumental sciences: grammar and morphology, which govern the word; rhetoric, which governs eloquence; and the theory of law (uṣūl al-fiqh), which governs the derivation of rulings and itself borrowed much from logic. El-Rouayheb observes that handbooks of uṣūl al-fiqh and kalām adopted logical terminology, and al-Ghazālī (d. 505/1111) opened his late work on legal theory, al-Mustaṣfā, with a whole introduction on logic, calling it ‘an introduction to all the sciences’. The Ottoman scholar Ṭaşköprüzāde (d. 968/1560) praised logic as the highest of the instrumental sciences and the chief of the rational ones.

This is how the madrasa understood manṭiq: not as a branch of metaphysics, but as a short, disciplined training in defining and arguing, read early, usually through a small text such as the Īsāghūjī or the Sullam, and then put to work in law, theology and the Arabic sciences.