Al-Risāla al-Shamsiyya: the book

1The title and the dedication

The full title is al-Risāla al-Shamsiyya fī al-qawāʿid al-manṭiqiyya, ‘the Shamsian treatise on the rules of logic’. It is usually called simply al-Shamsiyya, in Turkish Şemsiyye and in Persian and Urdu Shamsiyya. The name has nothing to do with the sun (shams): the book is named after the man for whom it was written. Ḥājjī Khalīfa says in his Kashf al-ẓunūn that al-Kātibī ‘wrote it for Khwāja Shams al-Dīn Muḥammad and named it after him’.

This patron was Shams al-Dīn Muḥammad b. Bahāʾ al-Dīn Muḥammad al-Juwaynī, the brother of the historian ʿAṭāʾ Malik al-Juwaynī and for more than twenty years the head of the civil administration (ṣāḥib-dīwān) of the Mongol Ilkhans in Iran. According to the TDV İslâm Ansiklopedisi Hülegü appointed him in 661/1263 and he kept the office under Abaqa (1265–1282); he was known as a generous patron of scholars and poets, and Naṣīr al-Dīn al-Ṭūsī and Ṣafī al-Dīn al-Urmawī also dedicated works to him. He fell from favour under Arghun and was executed near Ahar, between Qazvin and Zanjan, on 4 Shaʿbān 683/16 October 1284.

The dedication gives the only firm date for the book. Tony Street, who has edited and translated it, points out that it must have been ‘written or redacted after Shams al-Dīn’s accession to power in 1262’ (the TDV gives the appointment as 661/1263). Al-Kātibī himself died in 675/1276–77 according to most sources, so the Shamsiyya belongs to the last fifteen years of his life, the years of the Maragha observatory.

The preface opens with praise of knowledge: the sciences, ‘especially the certain ones, are the highest of aims and the most splendid of merits’; their possessor is ‘the noblest of human individuals’, whose soul is ‘quickest to join the angelic intellects’; and their subtleties can only be grasped through ‘the science called logic’, by which one tells the sound from the unsound. Then comes a long string of titles for the patron, ‘Shams al-Milla wa-l-Dīn … Muḥammad’, son of the ṣāḥib-dīwān Bahāʾ al-Ḥaqq wa-l-Dīn Muḥammad, who ‘despite his youth’ excelled in virtue and who asked for ‘a book on logic that gathers its rules and contains its principles and canons’. Al-Kātibī says that he set out not to leave out any rule of consequence, to add ‘noble additions and subtle points of my own’, and to follow no one ‘but the plain truth’. He names the book and gives its plan: an introduction, three treatises and a conclusion. The TDV sums this up by saying that he did not bind himself to the arrangement of earlier logic books and added material not found in them.

2Size and structure

The Shamsiyya is short, though several times longer than al-Abharī’s Īsāghūjī. Street counts about 120 numbered sections (lemmata), ‘just over 12000 words in English translation’; the Egyptian Dār al-Iftāʾ calls it a treatise of about twenty pages. The Calcutta edition of 1854, Arabic text with Aloys Sprenger’s English translation, has ii + 36 + 29 pages.

The book has an introduction (muqaddima), three treatises (maqālāt) and a conclusion (khātima), and the headings of the text itself give a clear plan. The introduction has two discussions (baḥth): what logic is and why it is needed, and what its subject is. The first treatise, ‘on simple terms’ (al-mufradāt), has four sections (faṣl): on expressions; on simple meanings, where the universal, the particular and the five universals are introduced; on further questions about universal and particular, five in number; and on definitions. The second treatise, ‘on propositions and their rules’, has an introduction and three sections: the categorical proposition (in four discussions: its parts and kinds, the four quantified propositions, metathesis, and the modal propositions); the kinds of hypothetical proposition; and the rules of propositions (in four discussions: contradiction, simple conversion, contraposition, and the implications between hypotheticals). The third treatise, on the syllogism, has five sections: the definition and kinds of syllogism, with the four figures; the mixed modal syllogisms (mukhtaliṭāt); conjunctive syllogisms made of hypotheticals; the exceptive (istithnāʾī) syllogism; and the four ‘appendages’ of the syllogism. The conclusion has two discussions: on the matter of syllogisms and on the parts of the sciences. Street’s edition follows the same divisions.

Street, who counts the conclusion as a fourth part, notes that this plan is the end point of a long compression of Ibn Sīnā’s al-Ishārāt wa-l-tanbīhāt. Ibn Sīnā had arranged his logic in ten ‘paths’ (nahj); Fakhr al-Dīn al-Rāzī reduced them to six in his digest of the book, and al-Kātibī merged some more. Using the distinction between the form and the matter of an argument that al-Ṭūsī had sharpened, he moved everything that concerns the matter of arguments, including demonstration and the sophistical fallacies, into the conclusion.

3Contents: knowledge, terms and definitions

The introduction starts from the nature of knowledge. Knowledge is either conception alone (taṣawwur), ‘the occurrence of the form of a thing in the intellect’, or conception together with a judgement, which is called assent (taṣdīq). Neither all conceptions nor all assents are self-evident, or we would be ignorant of nothing; nor are all of them acquired by reasoning, or there would be a circle or an infinite regress. Since reasoning often goes wrong, as people contradict one another and even themselves, a canon is needed, and that is logic, which ‘they have described as an instrumental canon whose observance protects the mind from error in thinking’. The subject of logic, al-Kātibī says, is ‘known conceptions and assents’ insofar as they lead to an unknown conception or assent. This was a disputed point: El-Rouayheb notes that al-Taftāzānī later defended al-Kātibī’s view against al-Ṭūsī’s objections, and the Stanford Encyclopedia of Philosophy traces it back to al-Khūnajī.

The first treatise covers the ground of the Īsāghūjī and a little more. Its first section explains the three kinds of verbal signification, by correspondence (muṭābaqa), inclusion (taḍammun) and implication (iltizām), with the example of ‘human’ signifying ‘rational animal’, ‘animal’ alone and ‘capable of knowledge’; then simple and composite expressions and the kinds of simple words. The second section defines universal and particular and derives the five universals: species, genus (near and remote), differentia, proprium and common accident, with the distinction between inseparable and separable accidents. The third section adds five further discussions, among them the different ways in which a universal can be instantiated (impossible, possible but not existing, existing once or many times), the ‘natural’, ‘logical’ and ‘mental’ universal, and the four relations between two universals. Al-Kātibī holds that the natural universal exists outside the mind, ‘because it is part of this existing animal’, and leaves the existence of the other two as disputed and ‘outside logic’; El-Rouayheb notes that al-Taftāzānī later argued against this position. The fourth section treats definition: the complete and incomplete essential definition, the descriptive definition, and the faults to avoid, such as defining by what is equally unknown, circular definitions and rare or obscure words. Unlike many earlier logic books, and unlike the later Indian al-Mirqāt, the Shamsiyya says nothing about the ten categories.

4Contents: propositions

The second treatise is where the Shamsiyya is most technical. Its first section analyses the categorical proposition into subject, predicate and the relation between them, whose expression is the copula; al-Kātibī observes that ‘in some languages’ the copula is dropped because the mind is aware of its meaning. It then explains the four quantified propositions, read either ‘according to the real nature’ (bi-ḥasab al-ḥaqīqa: whatever would be J, if it existed, would be B) or ‘according to the external world’ (bi-ḥasab al-khārij), a distinction that Quṭb al-Dīn al-Rāzī’s commentary develops at length, metathesis, in which the negation becomes part of the subject or predicate, and the modal propositions. ‘The modal propositions which it is customary to discuss’, he writes, ‘are thirteen’: six simple ones, which are only affirmative or only negative, and seven compound ones, made up of an affirmation and a negation. Later handbooks repeat and enlarge this list (al-Mirqāt has fifteen).

The second section treats conditional (muttaṣila) and disjunctive (munfaṣila) propositions with their kinds and their quantifiers; Street counts twenty-one lemmata on them, far more than Ibn Sīnā gave them in the Ishārāt. The third section deals with contradiction, simple conversion, contraposition and the implications between hypothetical propositions, including the conversion of the modal propositions. Here al-Kātibī sometimes leaves a question open. Street notes that he declares the conversion of the possibility propositions ‘unknown’, and that on this and a few other points, such as the subject matter of logic and conversion with negative terms, he departs from Ibn Sīnā and follows al-Khūnajī.

5Contents: syllogisms and the matter of arguments

The third treatise defines the syllogism and divides it into the conjunctive (iqtirānī) and the exceptive (istithnāʾī) kind. It sets out the four figures of the conjunctive syllogism, determined by the position of the middle term, with their conditions and productive moods. The TDV stresses that al-Kātibī includes the fourth figure, which al-Fārābī and Ibn Sīnā did not treat, and gives it eight productive moods where earlier authors had five; the text indeed says ‘its productive moods are eight’. The second section treats the ‘mixtures’ (mukhtaliṭāt), syllogisms whose premises have different modalities, with the modal conditions of each figure; this short section later produced a literature of its own. The third section shows that conditional and disjunctive propositions form conjunctive syllogisms in five combinations and in the same four figures; the fourth treats the exceptive syllogism. The last section gives four ‘appendages’ (lawāḥiq) of the syllogism: the compound syllogism, the reductio (qiyās al-khalf), induction and analogy. Induction, al-Kātibī warns, does not give certainty: all animals seem to move their lower jaw when they chew, but the crocodile does not. Of the two methods used to find the common cause in an analogy, ‘co-presence’ (dawarān) and elimination, he says that both are weak.

The conclusion turns from form to matter. Its first discussion classifies premises into certain and non-certain ones. The certain premises are six: first principles, observations, premises based on repeated experience, intuitions, reports transmitted by many, and propositions ‘whose syllogisms come with them’. A syllogism made of them is a demonstration, either ‘why’ (limmī) or ‘that’ (innī); the six kinds of non-certain premise (widely accepted, conceded, received, presumed, imaginative and estimative ones) are the matter of dialectic, rhetoric, poetry and sophistry, and the discussion ends with the fallacy (mughālaṭa) and its sources, in the form and in the matter of an argument, with the school example of the picture of a horse on a wall: ‘this is a horse, and every horse neighs’. Ibn Sīnā had treated these premises before the syllogism; al-Kātibī puts them after it, combining material that the Ishārāt divides between its sixth and ninth paths (TDV). The second discussion gives the parts of every science: its subjects, its principles (definitions and premises, illustrated from geometry: ‘we may join any two points with a straight line’) and its problems.

6An Avicennan handbook of a new kind

Street calls al-Kātibī ‘fundamentally an Avicennan logician’: the Shamsiyya keeps Ibn Sīnā’s division of syllogisms, his temporal reading of unmodalised propositions and most of his doctrine. But the TDV, following Street, stresses that its arrangement is original, and calls it one of the best examples of the ‘formalisation’ of logic after Ibn Sīnā: the material topics shrink, the formal ones (modal propositions, conditionals, conversions, the moods of the four figures) grow and are set out with great precision. The Stanford Encyclopedia of Philosophy shows that the book was written after the debates of the Maragha circle, in which al-Abharī, al-Ṭūsī and al-Kātibī himself argued over al-Khūnajī’s challenges to Ibn Sīnā, and that it draws on their arguments.

The result was a book that suited the madrasa. It is short enough to be read in one course, yet it touches on almost every question that the longer works discuss, often in a single dense sentence that a teacher has to unpack. Street remarks that few people have read it without a teacher. That made it ideal ground for commentary: the commentaries of al-ʿAllāma al-Ḥillī and Quṭb al-Dīn al-Rāzī were written within decades, and the glosses on Quṭb al-Dīn’s commentary eventually outnumbered the commentaries on the Shamsiyya itself (see ‘Commentaries and glosses’).

Its success can be measured in several ways. The publisher of Street’s edition calls it ‘the most widely read introduction to logic in the Arabic-speaking world’, which ‘has probably enjoyed a longer shelf-life than any other logic textbook ever written’. Street notes that it effectively replaced the logic of the Ishārāt in teaching, so that later copyists of the Ishārāt often left out its logical part, and that specialists came to treat ‘Shamsiyya logic’ as the logic of Ibn Sīnā. In the twentieth century the logician Nicholas Rescher reconstructed from it what El-Rouayheb calls ‘a sophisticated system of temporal and modal logic’.

7The Shamsiyya and the other books of this site

For a student of The Mantiqi the Shamsiyya is the natural next step after the Īsāghūjī. In the old Ottoman order of study it followed al-Abharī’s primer and its commentaries, and in the Indian Dars-i Niẓāmī it was read, with Quṭb al-Dīn’s commentary, after the Tahdhīb (see ‘The Shamsiyya in the madrasa’). It also stands behind al-Mirqāt: the TDV states that Faḍl-i Imām Khayrābādī’s book is based mainly on the Shamsiyya and al-Taftāzānī’s Tahdhīb, and many of al-Mirqāt’s discussions, from the meanings of ‘knowledge’ to the modal propositions, follow the Shamsiyya and its commentators.

The text has been printed many times since 1815, in Calcutta, Istanbul, Lucknow, Cairo and Tashkent, usually inside its commentaries. The first modern edition, by Mahdī Faḍl Allāh (Casablanca 1998), was based on three printed copies; the first edition with an English translation and full commentary is Tony Street’s (Library of Arabic Literature, 2024). These and the other editions and translations are listed in the catalogue of this section.