Matn, sharḥ, ḥāshiya: how the literature works

1Why the literature comes in layers

A student in a madrasa did not begin logic with a long treatise. He began with a short base text, a matn, which could be memorised and read quickly with a teacher, and then read the same material again in a commentary, and perhaps a third time in a gloss on that commentary, before moving on to a longer handbook. The TDV İslâm Ansiklopedisi notes that sources on legal teaching describe exactly such a graded sequence: text, commentary, gloss, notes, so that the student advances step by step.

Commentaries were written for many reasons. The TDV’s article on commentary (‘Şerh’) lists them: readers differ in ability, so a text written for one level needs explaining for another; an author may leave out basics that belong to another discipline, or leave his reasons unstated; concise and figurative language needs unpacking; and earlier works contain gaps and errors that need correcting. Commentary also served debate: the commentator explains the author’s meaning, sets out other opinions, raises objections and gives his own view.

This made the commentary the main place where logic was done after the thirteenth century. Historians once took this as a sign of decline. Tony Street and Nadja Germann warn against that inference: ‘genre by no means dictates content, and we often find original work presented in this way.’

2Matn, mukhtaṣar and risāla

A matn is a base text, written to be learned, taught and commented on. Many matns are abridgements, mukhtaṣar. The TDV distinguishes a family of related terms: a mukhtaṣar abridges a work; a talkhīṣ or mulakhkhaṣ summarises it; a tahdhīb revises it, removing weak or superfluous material; a mujarrad strips out the evidence. Several famous logic books carry these names: Fakhr al-Dīn al-Rāzī’s al-Mulakhkhaṣ, al-Taftāzānī’s Tahdhīb al-manṭiq, and the Mukhtaṣar of the North African theologian al-Sanūsī.

Short works were also called risāla, ‘treatise’ or ‘epistle’. The most studied logic handbook of all is a risāla: al-Kātibī’s al-Risāla al-Shamsiyya, named after the Ilkhanid vizier Shams al-Dīn al-Juwaynī, who had asked for a treatise containing all the rules of logic. Al-Kātibī says in his preface that he did not tie himself to the arrangement of earlier logic books and added material not found in them.

Concision had its critics. Ibn Khaldūn argued that very short works make a subject hard to understand, and the TDV notes that some blamed the dominance of abridgements for a decline of effort in scholarship. But the abridgement and the commentary were two sides of one practice: the short text condensed the tradition, and commentaries opened it up again.

3Sharḥ: the commentary

A sharḥ, a word whose lexical sense is ‘opening up’ or ‘making clear’, explains a whole work. The TDV distinguishes three ways of laying one out. In the first, a passage of the text is quoted after qāla (‘he said’) and the explanation follows after aqūlu (‘I say’). In the second, only the words to be explained are quoted, introduced by qawluhu (‘his words …’). In the third, the sharḥ mamzūj (‘blended’ commentary), text and explanation are woven into one continuous sentence, a style that became common in later centuries. Commentaries also came in sizes: the TDV notes that the great commentators on Aristotle’s logic wrote ‘large, medium and short’ commentaries.

In logic the standard commentaries of the main handbooks became classics in their own right: Quṭb al-Dīn al-Rāzī’s Taḥrīr al-qawāʿid al-manṭiqiyya on the Shamsiyya and his commentary on al-Urmawī’s Maṭāliʿ, al-Dawānī’s unfinished commentary on the Tahdhīb, and, for the Īsāghūjī, the commentaries of al-Kātī, al-Fanārī and Zakariyyā al-Anṣārī (see ‘The commentary tradition’ for the Īsāghūjī). An author might also comment on his own text; the TDV gives Ibn Mālik and Ibn Hishām as examples from grammar, and in logic al-Akhḍarī commented on his own Sullam (see ‘The commentary tradition’).

4Ḥāshiya, taʿlīq and taqrīr

A ḥāshiya (plural ḥawāshī) is literally the margin of a page, and originally a note written there. According to the TDV’s article on the term, from the eighth/fourteenth century onwards ḥāshiya came to mean an independent work on a commentary: a ‘commentary on a commentary’ that takes up difficult words and phrases of the commentary or the base text, adds information and sometimes criticises the commentator. The article suggests that many glosses grew out of teachers’ and students’ lesson notes, since most were written on textbooks used in teaching. Later glosses typically introduce each phrase they discuss with qawluhu. Some glosses became longer than the texts they explain.

A taʿlīq or taʿlīqa (plural taʿlīqāt), from a root meaning ‘to attach’, is used for shorter attached notes, often written between the lines or in the margin; authors also used the word modestly for a full commentary. A taqrīr is a further layer of notes, typically on a gloss. A Būlāq printing of about 1880 shows the three levels together: al-Akhḍarī’s Sullam as the base text, Ibrāhīm al-Bājūrī’s ḥāshiya on it, and in the margin the taqrīr of Muḥammad al-Anbābī (1824/5–1895/6). The word also had an oral sense: the TDV notes that in Ottoman palace lessons the scholar who delivered the lesson was called the muqarrir.

Two further genres belong here. An iʿrāb is a grammatical analysis of a text, word by word; The Mantiqi’s article on the commentary tradition notes one on al-Kātī’s commentary on the Īsāghūjī. And a short treatise could be written on a single question raised in the commentaries, so that the debate continued outside the layered books altogether.

5Naẓm: logic in verse

A naẓm or manẓūma (a poem) turns a science into verse so that it can be memorised. A poem in the rajaz metre is called an urjūza, and Ibn Sīnā himself wrote an Urjūza fī al-manṭiq. The most successful logic poem is al-Akhḍarī’s al-Sullam al-munawraq, written in Algeria in the sixteenth century and traditionally described as a versification of the Īsāghūjī; it was commented on by al-Mallawī, al-Damanhūrī, al-Bājūrī and many others, and al-Azhar still teaches logic from a simplified version of it with al-Mallawī’s commentary.

Shorter poems were written on single topics. Khaled El-Rouayheb mentions an eighteenth-century Egyptian scholar, Aḥmad al-Sijāʿī, who versified the moods of the syllogism, with a commentary by the Mālikī jurist and Sufi Aḥmad al-Dardīr, and the Damascene Sufi ʿAbd al-Ghanī al-Nābulusī, who summarised the basic rules of logic in a poem with his own prose commentary. A verse text was usually read with a prose commentary, so that poem and explanation formed one more pair of layers.

6One handbook and its layers: the Shamsiyya

The Shamsiyya shows how far the layering could go. The TDV’s article on it traces the main branches. At the root is al-Kātibī’s text (d. 675/1276–7). On it stands Quṭb al-Dīn al-Rāzī’s commentary Taḥrīr al-qawāʿid al-manṭiqiyya, printed in Calcutta in 1815 and later in Istanbul, Lucknow and Cairo. On that commentary glosses were written by al-Sayyid al-Sharīf al-Jurjānī, al-Taftāzānī, ʿIṣām al-Dīn al-Isfarāyīnī, al-Dawānī, Muḥammad al-Dasūqī and the Ottoman Dede Cöngî. On al-Jurjānī’s gloss there are further glosses by Mullā ʿAlī al-ʿAjamī, Molla Fenârî, Kara Dâvud of İzmit and ʿAbd al-Ḥakīm al-Siyālkūtī; and on al-Siyālkūtī’s, glosses and notes by ʿAbd al-Ḥamīd al-Kharpūtī and ʿAbd al-Raḥmān al-Shirbīnī. That is five levels: matn, sharḥ, ḥāshiya, a gloss on the gloss, and notes on that.

The printed books mirror the layers. An Istanbul printing of 1843 contains al-Siyālkūtī’s gloss on al-Jurjānī’s gloss on Quṭb al-Dīn’s commentary on the Shamsiyya. A two-volume Cairo collection, Shurūḥ al-Shamsiyya, gathers Quṭb al-Dīn’s commentary with the glosses of al-Jurjānī, al-Siyālkūtī, al-Dawānī, al-Dasūqī and al-Shirbīnī, and prints al-Kātibī’s own text at the end of the second volume, almost as an afterthought. The TDV dates it 1327–28/1909–10; a library copy of a two-volume Majmūʿ of the same commentary and glosses from Cairo’s Amīriyya press is dated 1905, so there may have been more than one printing.

The layers were not evenly spread. El-Rouayheb observes that roughly three quarters of al-Jurjānī’s gloss on the Shamsiyya commentary is devoted to conceptions, and that the later parts of Quṭb al-Dīn’s commentary, on conversion and the syllogism, appear not to have received a single gloss after the fourteenth century. Where the glossators wrote shows where the live questions were.

7Reading in class: dars and muṭālaʿa

The layered book was made for a classroom. Asad Q. Ahmed, studying the commentaries on Muḥibb Allāh al-Bihārī’s Sullam al-ʿulūm in India, describes commentary as moving constantly between the spoken and the written. A debate held at the court of Rampur in 1916, about positions taken by the Khayrābādī scholar ʿAbd al-Ḥaqq in his commentary on a commentary on a short treatise of Quṭb al-Dīn al-Rāzī, produced within months a chain of printed treatises and rebuttals by the debaters’ students and their students, each speaking in the voice of his teacher. Written commentary preserved the classroom’s dialogue, and new dialogue produced new written commentary.

Base texts were written to provoke this. Ahmed shows that expressions such as fa-taʾammal (‘reflect!’) and fa-tadabbar (‘consider!’), which the Sullam al-ʿulūm uses nine times, were read by commentators as hints: signals that the solution in the text was incomplete or open to objection and that a later commentator should take up its defence. A text was meant to remain elusive enough to demand the reader’s own effort.

Teachers guarded that effort. Ahmed quotes a modern scholar of the Dars-i Niẓāmī, ʿAbd al-Salām Khān, whose teachers told him not to consult the very clear commentaries and glosses of Mullā Mubīn while doing his own deep reading (muṭālaʿa) of a base text, because they made the text too easy: ‘Don’t look at Mubīn!’ (Mubīn-rā mabīn!). The aim of muṭālaʿa was to extract the author’s intention oneself; commentaries were for checking, not replacing, that work.

8How to read a page of commentary and gloss

Printed books lay out the layers in two main ways, and the TDV describes both. In one, the base text runs across the top of the page, the commentary below it, and the gloss and notes lower down or in the margins. In the other, common for glosses, the gloss fills the large central block, because it is the longest layer, and the commentary or base text runs around it in the margins. Manuscripts used other devices: letters such as mīm or ṣād for the text and shīn for the commentary, lines to separate or mark them, and reference marks, a number or letter placed in the text and repeated beside the note.

A practical way in is to work outward from the base text. First find the matn passage being discussed. Then find the commentary on it: in a qāla/aqūlu commentary look for qāla, in a blended commentary look for the words of the matn, marked by brackets, a different script or a line, inside the commentator’s sentences. Then find the gloss: each of its sections begins with qawluhu and a few words, the lemma, which may be taken from the matn or from the commentary; locating that phrase tells you exactly what the glossator is discussing. Finally, read the gloss with its question in mind: is it explaining a word, supplying a missing step, or raising an objection to the commentator and answering it?

It also helps to know whose voice is speaking. A gloss may discuss the words of the base text or those of the commentary, and, as Ahmed shows, commentators often speak in the voice of their predecessors, defending or extending them, before giving their own view. Reading a page of this kind is slow, which was exactly its purpose: it trained the student to follow an argument through several minds before adding his own.