AbÅ« Ê¿AlÄ« Ibn SÄ«nÄ (d. 428â¯AH, 1037â¯CE), usually known in English by his Latin name Avicenna, was the most influential thinker of the Islamic world. This book is the second volume in a series devoted to charting just a part of that influence, albeit the part when his direct impact was greatest. Indirectly, one can fairly say that philosophy in the Islamic world was done within a broadly Avicennan framework right down to the 19th century. Roughly speaking, our chronological span is the 12â13th centuries CE. More specifically, we begin after the death of al-GhazÄlÄ« in 1111â¯CE and end with the circle of thinkers gathered around Naṣīr al-DÄ«n al-ṬūsÄ«. As in the first volume, for reasons of time, space, and coherence we focus on the Islamic East. That first volume, which appeared in 2023, was devoted to topics in metaphysics and philosophical theology. Here we turn to the disciplines that Avicenna himself might have suggested starting with: logic and epistemology. He does, after all, follow the late ancient and previous Arabic philosophical tradition by treating logic as the introductory philosophical discipline. This is already the case in his first general summa of philosophy, the Philosophy for Ê¿ArÅ«á¸Ä« (al-Ḥikma al-Ê¿ArÅ«á¸iyya), and logic is also treated first in his magnum opus, the Healing (al-ShifÄʾ), and in his difficult but much-read final summa, the Pointers and Reminders (al-IshÄrÄt wa l-tanbÄ«hÄt).1 The points made in Avicennaâs logic ramify through his whole philosophy, so that no one who fails to come to grips with this part of his thought can fully understand his work on topics like metaphysics or the soul. Those who responded to Avicenna were fully aware of the importance of the topic, and duly devoted intense reflection to his syllogistic theory and other issues closely related to logic, including philosophy of language and epistemology.2
We ourselves are following late ancient precedent by covering in this volume more or less the range of topics that would have been dealt with under the heading of the logical corpus or organon in the Aristotelian tradition. This explains why the themes covered here may seem somewhat disparate from a contemporary philosophical point of view. In particular, our chapters on the nature of knowledge might seem more related to philosophy of mind, and therefore to belong in the following volume, which will include chapters on soul and intellect. Ultimately though, Avicenna approaches epistemology from the perspective of Aristotleâs Posterior Analytics, called simply Demonstration (BurhÄn) in Arabic,3 so it makes some sense to have material on knowledge in the same volume as the material on the Organon in general, and demonstration theory in particular (Chapter 13). Similarly, topics like semantic reference are covered in Aristotleâs On Interpretation and the corresponding part of Avicennaâs ShifÄʾ (the IbÄra), while paradoxes, the topic of our final chapter, are traditionally dealt with under âsophistic,â though this is more reliably true in the Latin tradition than in the Arabic one. If this provides us with a historical justification for grouping all this material into a single book, a philosophical rationale may be found with the central role that conception (taá¹£awwur) and assent (taá¹£dÄ«q) will play throughout. This terminological contrast, originally introduced by al-FÄrÄbÄ«, becomes definitive of logic itself,4 and the nature of knowledgeâeven from a psychological perspectiveâoften plays a crucial role in explaining the relation between conception and assent for post-Avicennan philosophers.
In the previous volume, there was almost no need to insist upon the pioneering nature of Avicennaâs achievement. The novelty and influence of such ideas as the essence-existence distinction, mental existence, and God as the Necessary Existent more or less speak for themselves. When it comes to logic and epistemology, by contrast, an appreciation of the Avicennaâs innovations has been restricted to a smaller circle of specialists. This is unsurprising given the often technical nature of Avicennaâs advances. His distinctive ideas will of course be charted in detail in the rest of the volume, and a first impression can be had by reading the introductions to each chapter (especially the first part of each introduction, since we usually start by outlining the views of Avicenna that prompted controversy, modification, and defense). But we should say just a little here to prepare the reader for what is to come.
As Khaled El-Rouayheb has succinctly explained in his invaluable recent overview of post-Avicennan Arabic logic, there are several crucial respects in which Avicennaâs logic differs from previous Aristotelian logic, even if it is grounded in that tradition and carries on innovations made by earlier thinkers, especially al-FÄrÄbÄ«.5 First, and as we explain more fully in the introduction to the Modal Syllogistic chapter below, for him all propositions have some âmodality,â even if this is left unstated. Thus an apparently âplainâ statement not marked as necessary or possible, like âSocrates is walking,â would for Avicenna be considered an âabsoluteâ proposition. For him, this means that the proposition asserts the predication of walking for Socrates at least once.
Second, Avicenna works to integrate into a generally Aristotelian framework inferences of the kind that were studied in antiquity not by Aristotle, but by the Stoics. An example would be good old modus ponens: If A then B, but A, therefore B. Avicenna applies quantification here too, so that hypothetical syllogistic becomes modalized just like predicative logic. To use the same example, Avicenna might take the syllogism:
If Socrates is walking, then Socrates is movingBut Socrates is walkingTherefore Socrates is moving
and say that the second premise is âabsolute,â and should be read as âSocrates is walking at least once,â licensing the conclusion âSocrates is moving at least once.â And the same for other modalities, like necessity, possibility, perpetuity (what is always the case), etc. Also, Avicenna incorporates predicative statements into hypothetical argument forms by considering inferences like this one:
If A is B, then C is DIf C is D, then E is FTherefore if A is B, then E is F
Here the proposition âC is Dâ plays the role of a âmiddle termâ in a typical predicative syllogism (like Y in âAll X is Y, all Y is Z, therefore all X is Zâ).6
Third, Avicenna introduces an important distinction between two ways that a proposition can be interpreted: substantially (dhÄtÄ«) or descriptionally (waá¹£fÄ«). Here the idea is that some predicates belong to their subject as such, while others only belong to it when the subject is described a certain way. For example it is only true under a description that Socrates is moving, e.g. so long as he is walking. But it is true âsubstantiallyâ that Socrates is rational since this belongs to him in himself. This is important for several reasons; one of them is that it explains how science can be charged with understanding accidental properties. Consider the same example of walking and moving. Since science should concern itself with necessary and universal propositions that are always true, it is not at first clear how a scientific demonstration could say anything about these featuresâafter all nothing is walking necessarily or always. But it is necessarily and always true that anything walking is moving, so long as it is walking (that is, when it is âdescribedâ as walking). So in the waá¹£fÄ« reading, âwhatever is walking is movingâ becomes a perfectly respectable scientific truth, one that could be âdemonstratedâ in the strict and proper sense.
If this represents an expansion of logicâs remit, a final change undertaken by Avicenna, which we deal with in our fifth chapter, is that he narrows the scope of logic in another way. In part precisely because he takes so seriously the traditional idea that logic culminates with the theory of demonstration, he makes certain parts of the organon no longer a part of logic. In particular, he rejects the relevance of the Categoriesâa significant step, since philosophers had been treating this as the first treatise of logic (apart from Porphyryâs general introduction) for the better part of a millenium. Having said that, he does make a concession to tradition by retaining a section on the categories in his works on logic.
As the post-Avicennan tradition evolved, it became even more focused on the issues that we would nowadays expect to discuss under the heading of âlogic.â Basically this would take in theory of the proposition, argument forms and their validity, modalities, and invalid arguments. In a development that would surely have dismayed Avicenna himself, his heirs become increasingly uninterested in scientific demonstration, touching upon it only briefly or not at all. Effectively, logic becomes a study of the issues connected to syllogistic (what is covered in Aristotleâs Prior Analytics, and correspondingly Avicennaâs QiyÄs), not issues connected to science and epistemology (what is covered in Aristotleâs Posterior Analytics, and correspondingly Avicennaâs BurhÄn). This is remarkable, since as we have said demonstration was for Avicenna, as for al-FÄrÄbÄ«, the members of the Baghdad School of Aristotelians, and late ancient commentators, really the whole point of the logical project.
It is an interesting historical question why post-Avicennan logic becomes so exclusively focused on what we might now think of as the âcoreâ of logic. One hypothesis could be that logic was seen as important for such culturally resonant activities as legal reasoning. This is not something we have attempted to represent systematically, though we make some reference to it in chapter 4 and quote a few exemplary texts. Logic became a standard part of education for jurists and religious scholars in this period, something comparable to what we find in European Christendom, from the contemporaneous schools and universities of the 12th and 13th centuries down to Jesuit âcollegesâ beginning in the 16th century.7 A sign of logicâs pedagogical importance is the production of handbooks or overviews of logic by authors like al-KhÅ«najÄ«, al-UrmawÄ«, and al-KÄtibÄ«, whose Epistle for Shams al-DÄ«n (al-RisÄla al-Shamsiyya) would become a standard text used for centuries to come.8
These logicians were not content merely to repeat Avicennaâs logic and present it in more accessible form. To the contrary, this was a stunningly original period in the history of logic, something that can again be compared to what we see in the somewhat later Latin scholastic tradition. While the innovations cannot be credited to just one individual, there is a figure who emerges as the most original mind of the period: al-KhÅ«najÄ«, rightly called by El-Rouayheb âarguably second only to Avicenna in the history of Arabic logic.â9 In a stark contrast with the previous volume, where we did not translate a single passage from his works, here he plays a leading role in every chapter on the core areas of logic. In this respect he occupies something like the place of Fakhr al-DÄ«n al-RÄzÄ« in our survey of metaphysics in the period. More than anyone else, it is around al-KhÅ«najÄ« that the debate in logic revolved. Impressively, though, al-RÄzÄ« is another one of the main figures, from whom we will be quoting extensively on all issues in logic, philosophy of language, and epistemology. To give just one example of his significance, it was al-RÄzÄ« who introduced the key distinction between ḥaqÄ«qÄ« and khÄrijÄ« (âessentialistâ and âexternalistâ) readings of a proposition. On the latter reading, âA is Bâ is true only if B is said of A in extramental reality, whereas on the ḥaqÄ«qÄ« reading, âA is Bâ is true just in respect of a essential connection between A and B. As al-RÄzÄ« says, âA is Bâ is true ḥaqÄ«qÄ« regardless of whether A exists extramentally; it suffices that if A were to exist, then it would be B. This decisive intervention is only one of the numerous moves by al-RÄzÄ« weâll be charting in the book. In this sense, our second volume confirms his status as the greatest all-around thinker of the age (the third volume will offer further confirmation still.)
Al-RÄzÄ«âs importance notwithstanding, logic displays a rather different development than what we found with metaphysics. In the introduction to the first volume, we proposed subdividing our period into three stages: formation, culmination, and refinement. The âculminationâ comes with al-RÄzÄ« (d. 1210) and al-SuhrawardÄ« (d. 1191) towards the end of the 12th century, with an honorable mention for the somewhat earlier AbÅ« l-BarakÄt al-BaghdÄdÄ« (d. ca. 1165). Though we did find much of interest in 13th century metaphysics, that period emerged as one in which the positions staked out by 12th century figures were explored, critiqued, and modified, rather than as a period of further paradigm shifts. That broad picture remains accurate for some of the topics covered in this book too, especially in the chapters on epistemology; chapter 14 provides a good example.
But when it comes to logic, most of the action is in the 13th century. This is above all thanks to of al-KhÅ«najÄ« (d. 1248), but we also have other significant contributors like al-UrmawÄ« (d. 1283), Naṣīr al-DÄ«n al-ṬūsÄ« (d. 1274), and al-KÄtibÄ« (d. 1276). So our selection of material agrees with the later traditionâwhich produced many commentaries and super-commentaries on the works of these figuresâin giving a special status to these 13th century thinkers. For the details of their innovations, the reader is referred to the individual chapters in what follows. But if one had to say briefly in what the breakthrough consisted, one might simply reiterate that they began to delve deeply into logic as a subject worthy of consideration in its own right. To take just one telling example, al-KhÅ«najÄ« was fascinated by arguments with premises that have impossible subjects, like âround square is a figure.â This is not a topic that would attract the attention of someone who believed that the whole point of logic is to do science. Rather, al-KhÅ«najÄ« wants to chart all possible valid argument forms, and does not particularly mind whether they are forms one would use outside of a purely logical discussion.
Our different thematic interest means that we will also be quoting from the works of a few authors who were not yet introduced at the beginning of the previous volume. Apart from the sundry grammarians and jurists referred to in chapters 3 and 4, we should mention here Ibn al-á¹¢alÄḥ and Ibn WÄá¹£il al-ḤamawÄ«, who have not yet appeared on the scene. About Ibn al-á¹¢alÄḥ we know very little, and most of his works remain unexplored. He was likely born in HamadÄn and trained in medicine, serving as the court physician for the Artuqid ruler of Mardin, IlghazÄ« (r. 1107â1122). He died in Damascus in 1154. He is perhaps best known for his astronomical work on Ptolemyâs Almagest, but he also wrote on mathematical and philosophical problems. In this volume he appears as the first scholar in the Arabic tradition to defend the fourth figure of the syllogism. As for Ibn WÄá¹£il, he was a student of al-KhÅ«najÄ«, but like al-UrmawÄ« and al-KÄtibÄ« was an independent-minded logician. Born in 1208, he studied the rational sciences in Damascus with Fakhr al-DÄ«n al-RÄzÄ«âs student Shams al-DÄ«n al-KhusrawshÄhÄ«, and later settled in Cairo. The MamlÅ«k ruler Baybars sent him on a mission to Sicily at the court of Manfred, son of Frederick II. He died in his hometown Hamah in 1298. Other than his commentary on al-KhÅ«najÄ«âs Jumal, which has recently been edited by El-Rouayheb in 2022, Ibn WÄá¹£ilâs work on logic remains largely unknown.
With these exceptions, the cast of characters is the same, and so is the method we have used. The material presented here was selected from the vast corpus of philosophical literature in the 12â13th centuries by postdoctoral researchers on the Heirs of Avicenna project, in this case Fedor Benevich and Dustin Klinger. They did draft translations and worked on these with Peter Adamson, who also did the initial drafts of the thematic introductions for each chapter. We were fortunate enough to get further contributions from three more outstanding scholars, who did the selection and translation of passages for further chapters: these were Nora Kalbarczyk (chapters 3 and 4), Alexander Kalbarczyk (chapter 5), and Hassan Rezakhany (chapter 17). As with the previous volume, all the texts quoted are being made available in the original languages as a free online resource, which can be found at:
For Avicennaâs works and their structure see D. Gutas, Avicenna and the Aristotelian Tradition: Introduction to Reading Avicennaâs Philosophical Works (Leiden: 1988, 2nd revised ed. 2014), part one section 2.
For a good starting point on reactions to the core logical theory, see T. Street, âMedieval and Modern Interpretations of Avicennaâs Modal Syllogistic,â in F. Opwis and D.C Reisman (eds), Islamic Philosophy, Science, Culture, and Religion: Essays in Honor of Dimitri Gutas (Leiden: 2012), 232â256.
For an important study of Avicennaâs theory of demonstration see R. Strobino, Avicennaâs Theory of Science: Logic, Metaphysics, Epistemology (Oakland: 2021). The fact that knowledge is associated both with Avicennaâs philosophy of mind and his âlogicâ (in the section on demonstration) reflects a tension in epistemology going back to Aristotle. One of the central problems he bequeathed to posterity is that he discussed knowledge in very different ways in On the Soul and in the Posterior Analytics: it is no easy matter to say how the soulâs âreception of formâ in the former context could be the same phenomenon as understanding propositions as the conclusion of a demonstrative syllogism in the latter context. In order to grasp the full dimensions of Avicennaâs epistemology and its reception, readers will thus need to consult both this and the following volume.
Al-KhÅ«najÄ« says that the âobjects of conception and assentâ are its subject matter at [T14] of our chapter on the Subject Matter of Logic.
K. El-Rouayheb, The Development of Arabic Logic (1200â1800) (Basel: 2019), 22â24.
In these examples one is to read e.g. âX is Yâ as meaning âY is said of X,â which is why the propositions are âpredicative.â
A classic study of this parallel is G. Makdisi, The Rise of Colleges: Institutions of Learning in Islam and the West (Edinburgh: 1981).
See now T. Street, Najm al-DÄ«n al-KÄtibÄ«âs al-RisÄlah al-Shamsiyyah: an Edition and Translation with Commentary (New York: 2024). Unfortunately this book appeared too late for us to make use of it in our volume.
El-Rouayheb, The Development, 44.