Tahdhīb al-manṭiq wa-l-kalām: the book

1The title

The usual name, Tahdhīb al-manṭiq wa-l-kalām, means ‘the refinement (or pruning) of logic and theology’. It comes from the author’s preface, where he calls the book ‘the utmost in the refinement of speech (tahdhīb al-kalām), in setting out logic and kalām, and in bringing the aim within reach, in establishing the creeds of Islam’. The phrase plays on the two senses of kalām, ‘speech’ and ‘theology’. The book is therefore also called Tahdhīb al-kalām, and the logic part on its own, which travelled separately, is simply Tahdhīb al-manṭiq or ‘the Tahdhīb’. In the Indian madrasas the commentary of ʿAbd Allāh al-Yazdī is so closely tied to it that ‘Sharḥ-i Tahdhīb’ means that commentary.

M. Sait Özervarlı, in the TDV İslâm Ansiklopedisi, notes that the key word of the preface, taḥrīr, is usually understood as clearing logic and kalām of unnecessary detail. The plain style fits this reading, and so does the preface itself, which offers the book as ‘an insight for those of understanding and a reminder for those who wish to remember’ and dedicates it in particular to ‘the dearest son, worthy of honour, who bears the name of God’s beloved’: the author’s son Muḥammad, for whom he had already written a grammar, al-Irshād.

2Date and place

Kâtib Çelebi’s bibliography Kashf al-ẓunūn (I, 515–517) states that al-Taftāzānī wrote the book in 789. Brockelmann’s Geschichte der arabischen Litteratur is more precise: the book was completed in Samarqand in Rajab 789, that is, July–August 1387. Şükrü Özen’s TDV article on al-Taftāzānī also says it was finished in Samarqand, and the biography printed in the 2018 Amman edition of al-Yazdī’s commentary (summarising Ḍiyāʾ al-Dīn al-Qālish’s study of al-Taftāzānī) gives ‘789, in Samarqand’. The sources consulted agree on both year and place.

This places the book in the last years of al-Taftāzānī’s life, after Timur had brought him from Khwarazm to his capital. It was in the same year 789, according to the TDV, that Timur took Shiraz and brought al-Sayyid al-Sharīf al-Jurjānī to Samarqand, where the two scholars later disputed at court. Al-Taftāzānī had finished his large theological work al-Maqāṣid with its commentary in Samarqand in 784/1383, and Kashf al-ẓunūn says that the kalām part of the Tahdhīb abridges al-Maqāṣid. The Tahdhīb is thus the late summary of a scholar who had already written at length on both its subjects.

3Two parts: logic and kalām

The book has two parts (qism). The first is on logic and the second on kalām. According to the TDV the kalām part is divided into six sections: the definition, subject and methods of kalām; the ‘general matters’ (existence, essence, necessity and possibility, unity and plurality, causality); the accidents, treated along the lines of the categories; substances, including the indivisible atom, body, soul and intellect; God’s essence, attributes and acts, with good and evil, obligation, guidance and error, provision and the term of life; and finally prophecy, the afterlife, grave sins, repentance, faith and the imamate.

Although the kalām part was printed and commented on (for example by the Kurdish scholar ʿAbd al-Qādir al-Sanandajī, whose Taqrīb al-marām fī sharḥ Tahdhīb al-kalām appeared at Būlāq in 1318–1319/1900–1901), it is the logic part that made the book famous. Özervarlı remarks that the logic part is short but attracted most of the commentaries, and Kashf al-ẓunūn says that because its logic was ‘the best that had been written in its field, it became famous and spread to every horizon’. Brockelmann records copies that contain the logic part alone. The printed matn of the Cairo edition of 1330/1912 shows the proportions: the logic part fills only a few pages before the much longer kalām part begins.

4The structure of the logic part

A gloss printed with al-Yazdī’s commentary in the Amman edition describes the arrangement: ‘the first part, on logic, is arranged in an introduction, two aims (maqṣad) and a conclusion. The introduction explains the nature, the aim and the subject of logic. The first aim is on the topics of conceptions; the second aim on the topics of assents; the conclusion on the parts of the sciences.’ Within these divisions the text runs on in short paragraphs, some headed faṣl (‘section’).

The introduction opens with the division of knowledge: if it is assent (idhʿān) to a relation it is taṣdīq, otherwise taṣawwur. Neither kind is wholly self-evident or wholly acquired; what is acquired is acquired by reflection (naẓar), defined as ‘attending to what is known in order to obtain what is unknown’. Because reflection can go wrong, a rule is needed to protect it, and that rule is logic. Its subject is the known, conceptual or assentive, insofar as it leads to an unknown: what leads to a conception is called a definition (muʿarrif), what leads to an assent an argument (ḥujja).

5Conceptions

The first aim begins with signification (dalāla) and its three kinds, by correspondence, inclusion and implication, and then classifies expressions: simple and composite, the noun, the verb and the particle, univocal and ‘graded’ (mushakkik) terms, homonyms, transferred words, literal and figurative use, and synonyms. A section on concepts follows: the universal and the particular, the four relations between two universals, and the five universals, genus, species, differentia, proprium and common accident, with the ordering of genera upwards and species downwards. The accidental is divided into the inseparable (lāzim) and the separable, and the inseparable into the ‘evident’ and the ‘non-evident’.

A short passage distinguishes the logical universal (the concept ‘universal’ itself), the natural universal (what it is predicated of, such as ‘human’) and the mental universal (the two together), and adds, in four words, a position on one of the great disputes of philosophy: ‘the truth is that the natural universal exists in the sense that its individuals exist’. The aim ends with definition: a definition must be coextensive with what it defines and clearer than it; a definition by the proximate differentia is a ḥadd, by the proprium a rasm, each complete or incomplete according to whether the proximate genus is added; definitions by the common accident are not counted; and a merely verbal definition explains what a word means.

6Assents and arguments

The second aim covers in a few pages what al-Kātibī’s al-Risāla al-Shamsiyya, the handbook al-Taftāzānī had himself commented on, treats at length. It defines the proposition, divides it into categorical and conditional, names the parts of each, and distinguishes singular, quantified, unquantified and ‘natural’ propositions, propositions about externally existing, real and mental subjects, and propositions with a negated term. The modal propositions follow in a dense list: eight simple modals and seven compound ones, each named in a phrase. Separate sections deal with the kinds of conditional propositions, contradiction, simple conversion and contraposition, the last two with their rules for the modal propositions.

The section on the syllogism defines it as ‘a discourse composed of propositions from which, by itself, another discourse follows’, divides it into the exceptive (istithnāʾī) and the conjunctive (iqtirānī), and gives the four figures with the conditions under which each produces a conclusion and the ways of proving the productive moods. Here al-Taftāzānī adds something of his own: a single ‘rule (ḍābiṭa) of the conditions of the four figures’, a general formula that covers in one sentence all the conditions listed one by one. The editor of the Amman edition calls it one of al-Taftāzānī’s innovations; it attracted separate commentaries, including one by al-Yazdī, and Indian scholars were still writing on it in the nineteenth century.

Shorter paragraphs treat conditional syllogisms, the reductio (qiyās al-khalf), induction and analogy (with its two methods, dawarān and tardīd). The last section of the aim turns to the ‘matter’ of arguments: the kinds of certain premises (first principles, observations, experience, intuition, multiply transmitted reports), demonstration and its two kinds, ‘why’ (limmī) and ‘that’ (innī), and the four other arts, dialectic, rhetoric, poetics and sophistry, each defined by the kind of premises it uses. The conclusion explains the three parts of every science: its subjects, its principles (definitions and premises) and its problems.

7Style

The Tahdhīb is written in the most compressed style of the madrasa handbook. A whole doctrine is often stated in a single sentence; technical terms are used without being explained; proofs are reduced to a hint, as in the formula for conversion, ‘the proof in every case is that the contradictory of the converse, together with the original, yields an absurdity’. Some topics are dismissed explicitly: of the conjunctive syllogisms made of conditional propositions the author says only that ‘their details are long’.

This concision is the reason for the book’s many commentaries. Already the Central Asian commentator al-Khabīṣī, as quoted in Kashf al-ẓunūn, explained his own work by noting that the Tahdhīb contains most of the questions of the Shamsiyya, while students ‘are in confusion about understanding its difficult questions because of the extreme brevity of its wording’. The Tahdhīb is not a primer for absolute beginners: in most curricula it was read after an elementary book such as the Īsāghūjī, and a teacher was expected to unfold it.

8Reputation and use

Kashf al-ẓunūn calls the Tahdhīb ‘a solid text’ and reports that ‘the exacting scholars applied themselves to it in study and teaching’. The TDV says that it became a standard madrasa textbook because of its concise treatment of its topics, that it survives in very many manuscripts and that library records list more than fifty commentaries and glosses; it also sets the book beside two earlier handbooks, al-Āmidī’s Rumūz al-kunūz and al-Bayḍāwī’s Miṣbāḥ al-arwāḥ. A later fatwa on the status of logic, translated by Khaled El-Rouayheb and quoted in the Stanford Encyclopedia of Philosophy, names ‘Saʿd al-Dīn’, with the Īsāghūjī and the works of al-Kātibī and al-Khūnajī, as examples of logic ‘not mixed with philosophy’, which it is permissible and even a collective duty to study.

Each region read the book with its own commentary. In the Ottoman lands and in Iran the incomplete commentary of Jalāl al-Dīn al-Dawānī and the glosses on it were the advanced route; in Egypt al-Khabīṣī’s al-Tadhhīb, glossed by Azhar scholars such as Ḥasan al-ʿAṭṭār and Muḥammad al-Dasūqī, was the usual commentary; in Safavid and later Iran and in India the commentary of ʿAbd Allāh al-Yazdī became the usual way into the book. Francis Robinson’s list of the logic books of the Indian Dars-i Niẓāmī includes the Tahdhīb, ‘Sharḥ-i Tahdhīb’ (al-Yazdī) and ‘Mullā Jalāl’, which he describes as a gloss on al-Dawānī’s commentary, and his list for Safavid Iran includes al-Yazdī’s commentary. The current eight-year course of the Dār al-ʿUlūm at Deoband still has Sharḥ al-Tahdhīb in the third year, after Taysīr al-manṭiq and al-Mirqāt and before Quṭbī. The commentaries are described in ‘Al-Yazdī’s Sharḥ al-Tahdhīb’ and ‘The commentaries on the Tahdhīb’.

On The Mantiqi the Tahdhīb is read at level 4, with al-Yazdī’s commentary, after the Īsāghūjī (level 1), al-Kātī’s commentary on it (level 2) and al-Mirqāt (level 3). This follows the order of many South Asian courses, in which the Tahdhīb comes after al-Mirqāt.