Orientations of Avicenna's Philosophy: Essays on His Life, Method, Heritage 1000246159, 9781000246155

Citation preview

Also in the Variorum Collected Studies Series:

DIMITRI GUTAS Greek Philosophers in the Arabic Tradition

RICHARD M. FRANK (ed. Dimitri Gutas) Philosophy, Theology and Mysticism in Medieval Islam Texts and Studies on the Development and History of Kalam, Vol. I

RICHARD M. FRANK (ed. Dimitri Gutas) Early Islamic Theology: The Mutazilites and al-Ash arl Texts and Studies on the Development and History of Kalam, Vol. II

RICHARD M. FRANK (ed. Dimitri Gutas) Classical Islamic Theology: The Ash arites Texts and Studies on the Development and History of Kalam, Vol. Ill

PETER ADAMSON Studies on Plotinus and al-Kindl

RICHARD SORABJI Perception, Conscience and Will in Ancient Philosophy

WILFERD MADELUNG Studies in Medieval Muslim Thought and History

WILFERD MADELUNG Studies in Medieval Shi‘ism

JOHN DILLON The Platonic Heritage Further Studies in the History of Platonism and Early Christianity

ANDREW SMITH Plotinus, Porphyry and Iamblichus Philosophy and Religion in Neoplatonism

CHARLES BURNETT Arabic into Latin in the Middle Ages The Translators and their Intellectual and Social Context

JULES JANSSENS Ibn Slna and his Influence on the Arabic and Latin World

VARIORUM COLLECTED STUDIES SERIES

Orientations of Avicenna’s Philosophy

Dimitri Gutas

Orientations of Avicenna’s Philosophy

Essays on his Life, Method, Heritage

O Routledge

Taylor & Francis Group

LONDON AND NEW YORK

First published 2014 by Ashgate Publishing 2 Park Square, Milton Park, Abingdon, Oxfordshire 0X14 4RN 52 Vanderbilt Avenue, New York, NY 10017 Routledge is an imprint o f the Taylor & Francis Group, an informa business

First issued in paperback 2020 This edition copyright© 2014 Dimitri Gutas Dimitri Gutas has asserted his right under the Copyright, Designs and Patents Act, 1988, to be identified as the author of this work. All rights reserved. No part of this book may be reprinted or reproduced or utilised in any form or by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying and recording, or in any information storage or retrieval system, without permission in writing from the publishers. Notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identification and explanation without intent to infringe. British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library. The Library of Congress has cataloged the printed edition as follows: 2014947945 ISBN 13: 978-1-4724-3633-7 (hbk) ISBN 13: 978-0-367-60000-6 (pbk) VARIORUM COLLECTED STUDIES SERIES CS1050

CONTENTS Foreword

IX

Acknowledgments L if e

I

and

W

xiv

orks

Avicenna: biography

1-11

Encyclopaedia Iranica, III. London: Rout Iedge, 1987, pp. 67 - 70

II

Avicenna’s Madhab, with an appendix on the question of his date of birth

323-336

Quaderni di StudiA rabi 5 - 6 [Atti d e lX III Congresso d e ll ’Union Europeenne d 'A rab is ants et d ’Islamisants. Venice, 29 September ­ 4 October 1986], 1987 - 1988

III

Texts from Avicenna’s library in a copy by cAbd-ar-Razzaq as-Signahl (Notes and Texts from Cairo Manuscripts, II)

1-25

Manuscripts o f the M iddle E ast 2, 1987, pp. 8 - 1 7

IV

Avicenna’s marginal glosses on De anima and the Greek commentatorial tradition

77-88

Philosophy, Science & Exegesis in Greek, Arabic & Latin Commentaries (Essays in H onour o f Richard Sorabji), eds P. Adamson, H. Baltussen, and M. W.F. Stone (Bulletin o f the Institute o f Classical Studies Supplement 83.2). London: Institute o f Classical Studies, 2004, vol. I I

V

Avicenna’s Eastern (“Oriental”) philosophy: nature, contents, transmission Arabic Sciences and Philosophy 10, 2000

159-180

vi M

CONTENTS ethod

VI

The logic of theology (kalam) in Avicenna

59 - 72

Logik und Theologie. Das Organon im arabischen und im lateinischen Mittelalter, eds D. Perler and U. Rudolph. Leiden: Brill, 2005

VII

The empiricism of Avicenna

1 -53

Oriens 40, 2012, pp. 391 - 436

VIII

Medical theory and scientific method in the age of Avicenna

145 - 162

Before and A fter Avicenna (Proceedings o f the First Conference o f the Avicenna Study Group), ed. D. C. Reisman. Leiden: Brill, 2003 E p is t e m o l o g y

IX

Avicenna: the metaphysics of the rational soul

417 - 425

The Muslim World 102, 2012

X

Intuition and thinking: the evolving structure of Avicenna’s epistemology

1 - 38

Princeton Papers. Interdisciplinary Journal o f Middle Eastern Studies 9, 2001, 1 - 38; repr. in Aspects o f Avicenna, ed. R. Wisnovsky. Princeton: M arkus Wiener, 2001

XI

Imagination and transcendental knowledge in Avicenna

337 - 354

Arabic Theology, Arabic Philosophy. From the M any to the One: Essays in Celebration o f Richard M. Frank, ed. J.E. Montgomery. Leuven: Peeters, 2006

XII

Intellect without limits: the absence of mysticism in Avicenna

1-22

Intellect and Imagination in M edieval Philosophy (Proceedings o f the XI. International Congress o f M edieval Philosophy, Porto 26 - 31 A ugust 2002), eds M.C. Pacheco and J.F. Meirinhos. Turnhout: Brepols, 2006, pp. 351 - 372

XIII

Avicenna: mysticism and the question of his “Oriental” philosophy Encyclopaedia Iranica, III, 1987. London: Routledge, pp. 79 - 83

1-10

CONTENTS T he S tudy

XIV

of

A

v ic e n n a a n d h is

H

vii

e r it a g e

The heritage of Avicenna: the golden age of Arabic philosophy, 1000 -ca. 1350

81-97

Avicenna and H is Heritage. Acts o f the International Colloquium, Leuven - Louvain - la - Neuve, September 8—11, 1999. Leuven: Leuven University Press 2002

XV

Philosophy in the twelfth century: one view from Baghdad, or the repudiation of al-Ghazall

9-26

In the Age o f Averroes: Arabic Philosophy in the Sixth/Twelfth Century, ed. P. Adamson. London - Turin: The Warburg Institute - Nino Aragno Editore, 2011

XVI

Ibn Tufayl on Ibn Sma’s Eastern philosophy

222-241

Oriens 34, 1994

XVII The study of Avicenna. Status quaestionis atque agenda

45-69

Documenti e Studi sulla Tradizione Filosofica M edievale 21, 2010

Index of subjects and Avicenna’s works

1-4

Index of names and places

1-7

Index of principal terms (Arabic, Greek)

1-2

Index of manuscripts

This volume contains xiv + 368 pages

1

PUBLISHER’S NOTE The articles in this volume, as in all others in the Variorum Collected Studies Series, have not been given a new, continuous pagination. In order to avoid confusion, and to facilitate their use where these same studies have been referred to elsewhere, the original pagination has been maintained wherever possible. Articles I, III, VII, XII and XIII have necessarily been reset and given a new pagination. The original page numbers are indicated in square brackets within the text. Each article has been given a Roman number in order of appearance, as listed in the Contents. This number is repeated on each page and is quoted in the index entries.

FOREWORD In the course of the research on my book on Avicenna, which was first published in 1988,1 I came across a number of issues which could not be adequately treated then within the scope of that project; others were kindly noted by some of the reviewers of the book. In the quarter century that has elapsed since then, I have investigated them at greater depth and tried to clarify them in a number of articles; these are now conveniently collected here. Their main results have been incorporated in the second edition of the book that just appeared,2 to which this complementary volume thus forms a companion. The issues have to do with situating Avicenna in his historical, social, and philosophical context as a framework for reading his philosophical works, as the subtitle of the book has it. It is hardly a novel observation that authors - any authors, and not only philosophical ones - are received, interpreted, and valorized differently and variously at different times even within their own tradition (depending on the ideological orientations and social needs of that tradition at each point in time), let alone outside it, and this observation - this understanding of the distinction between authors and their heritage has not only benefited the historical study of them all but has itself grown into an independent field in the humanities in the form of reception studies and cultural studies. To take Aristotle as a commensurate example, one need only recall the title of the path-breaking publication edited by Richard Sorabji, Aristotle Transformed,3 and the immensely fruitful era of the study of Aristotelian philosophy in late antiquity (and early Islam, as well) it inaugurated. So just as we have learned to separate the historical Aristotle from Aristotelianism(s) down the centuries, so also must we disentangle Avicenna from self-proclaimed Avicennism(s) and re-created Avicenna(s). If anything, this has to be done with greater intensity and diligence in his case: he was a thinker of such overwhelming authority and brilliance - and not only within the Islamic tradition - that he was pulled and pushed by posterity here and 1 Avicenna and the Aristotelian Tradition. Introduction to Reading Avicennas’ Philosophical Works [Islamic Philosophy and Theology, Vol. 4], Leiden: Brill, 1988. 2 Avicenna and the Aristotelian Tradition. Introduction to Reading Avicennas’ Philosophical Works. Second, Revised and Enlarged Edition, Including an Inventory o f Avicenna s Authentic Works [Islamic Philosophy, Theology, and Science, Vol. 89], Leiden: Brill, 2014. 3 Aristotle Transformed: the Ancient Commentators and Their Influence, edited by Richard Sorabji, London: Duckworth, and Ithaca, N.Y.: Cornell University Press, 1990.

X

FOREWORD

there and in all directions in order to fit everybody’s agenda, philosophical or otherwise; as I remarked elsewhere, everybody wanted a piece of Avicenna. In this mesh of ideas and beliefs, the historical Avicenna has been lost. There are many reasons for this loss. First and foremost is what I have just alluded to: his immense popularity and authority led to the generation of a sizable body of pseudepigraphous writings which we have barely begun sorting out, something which constitutes the primary task of future scholarship.4 To the unsuspecting and (necessarily) uncritical reader (because of the lack of secondary scholarship), these tendentious writings have presented a view of Avicenna, understandably believed to be well founded, which has been difficult to dislodge. There is thus an Isma III Avicenna, a (Twelver) Shi I Avicenna, a mystical Avicenna, and all sorts of gradations within the Sunni As arl, Safi I and Hanafi, tradition (the Malikls and especially the Hanballs, disapproving of philosophy in general, have not indulged in this game of appropriating Avicenna for their own use).5As a minimal example I can cite the very famous and popular poem, the 'Ayniyya, the poem rhyming in (ayn, on the fall of the soul into the body in this world. A manifest Isma III fabrication, it has been ubiquitously ascribed to Avicenna except for the lone voice of an early literary (mind you!) critic, as-SarlsI (d. 619/1222), who pointed out its spuriousness, but who remained without following.6 There are other reasons for the loss of historical Avicenna beyond the posthumous traditions about and refashionings of him, those due to modem scholarship and its practitioners when dealing with subjects having to do with religious beliefs, nationalist feelings, and orientalizing (pro and con) approaches, like that on Avicenna. The study of philosophy in the Islamic world is a highly charged, even contested, field, and the personal beliefs, religious affiliations, and cultural and educational backgrounds of individual scholars have tended to cloud what at this point must be called “objectivity” euphemistically. The myth of a mystical Avicenna has been particularly irresistible, both for those who espouse orientalist approaches (clearly - the argument would go - Muslims, Arabs, orientals, etc., cannot have philosophy, a product of the (Aryan, European, etc.) Greek mind, since their greatest philosopher, Avicenna, was a mystic), and for those who oppose them on chauvinistic grounds (Avicenna, an oriental, was a mystic because mysticism 4 See the presentation of this case in the introduction to the inventory of his authentic works in my Avicenna (220 14), pp. 389-91, 411-13. 5 See the discussion of this point in my Avicenna (22014), pp. 367-8. 6 See the discussion about its pseudepigraphic nature in my Avicenna (220 14), pp. 453-6. There is an excellent new translation of the poem by G.J. van Gelder (in his Classical Arabic Literature, New York and London: NYU Press, 2013, 73—4), who, to his credit, does mention as-SarlsT’s doubts about its authenticity.

FOREWORD

xi

- the argument would go - is a form of knowledge superior to that provided by rationalist philosophy, a product of the Aryan, European, etc., mind). A variant of this latter position adds the further myth that Avicenna, who is taken to be Iranian, had access to a non-Greek “oriental”, pre-Islamic Iranian, mystical philosophy or theosophy ( Hrfan).1 And there are finally issues of confusion, by modem scholars, about the aims of historical research and the methods used to conduct it. It needs to be stated, though one would think hardly so at this point in time, that historical research, like that trying to establish the historical Avicenna, does not aim to “prove” or “disprove” any tradition’s or society’s system of beliefs and ideas, or show that they are “mistaken” or “right.” From the point of view of historical research, such beliefs and ideas - ideologies, generally - are historical facts or realities to be studied and are neither “right” nor “wrong:” they just are; scholarship simply tries to understand them and establish, on the available evidence, who thought, said, and did what, when, and why. If what historical research establishes turns out to be different from the ideological use to which a tradition or society put its history - which, as it turns out, is universally the case, no society ever having, or even proposing to have, a historian-king (Plato of course having proposed exactly the opposite) - that does not invalidate the ideology; it simply allows the historian to understand better its historical context, purposes, and directions. Historical research is not in the business of judging history but understanding it. A pertinent example of this confusion in the case of Avicenna is provided by the implied rhetorical question posed by Ch. Jambet, a French scholar and follower of H. Corbin, while reviewing my entry on Avicenna and mysticism in the Encyclopaedia Iranica (article XIII below) which denied Avicenna having mystical or illuminationist views. Jambet said that if I am right, then an entire Iranian philosophical tradition, led by SuhrawardI, must be wrong.8This expresses in a nutshell the problem with the tendentious - sectarian, really approaches to Avicenna that I have been decrying, for it shows a complete lack of understanding of the function and aims of historical research and scholarship on the one hand and of ideologies on the other, confusing the two - indeed, it negates scholarship, because the implication is clearly that the ideologically

7 I have discussed some of these approaches in my article, “The Study of Arabic Philosophy in the Twentieth Century. An Essay on the Historiography of Arabic Philosophy,” in British Journal o f Middle Eastern Studies 29 (2002) 525; and in the introduction to the Avicenna book (220 14), pp. xix-xxiii; see also article V below. A more sustained study of modem scholarship on Avicenna would be worth the effort. 8 “II faut que les avicenniens d ’ Orient (Sohravardi en tete) et toute une tradition de la philosophie iranienne se trompe, et que D. Gutas ait raison,” Abstracta Iranica 13 (1990) 81-2.

xii

FOREWORD

motivated traditional views of a society about its past are ipso facto “correct” and scholarship has no business questioning or investigating them. The studies collected in this volume look at the historical record and the texts of Avicenna himself, closely read, in order to resolve some of these issues about the historical man and his philosophy. Divided into four parts, they deal with his life and works, his method, his epistemology, and his later reception in the Islamic world, ending with a programmatic essay on the state of the field of Avicennan studies and future agenda.9 We are a long way from having a comprehensive biography of Avicenna and an accurate list of his authentic works. The studies collected in the first part pave the way towards that goal by reviewing the sources available to us for his biography, disposing of some problematic beliefs about his religious affiliation (he was Sunni Hanafi, not Isma III or Twelver SIT), highlighting points that have yet to be resolved, like his date of birth, and providing reliable information about some of his works that have been the cause of much misunderstanding, notably his texts on what he called the Eastern philosophy. As a scientist, Avicenna was a rationalist empiricist. Two of the three essays collected in the second part investigate the two aspects of his method, logic and empiricism, while the third charts the broader, historically determined scientific context in which these methods were intended (or allowed) to operate, that of the particular understanding of the classification of the sciences and their hierarchical interrelationship. This comprehensive evaluation of both the details and the general context of Avicenna’s methodology allows us to gauge with greater precision the possibilities and limitations of Avicenna’s science and, more generally, the science in the Islamic world of the tenth century. Epistemology engaged particularly the attention of Avicenna, a subject to the elucidation of aspects of which the essays in the third part are devoted. How we know depends on the faculties of the soul through which we know, and Avicenna accordingly allocated a significant amount of space in his works to the investigation of the senses, both external and internal, and of the intellect. Avicenna’s epistemology and his understanding of the faculties of the soul are also the means - the sole means - for analyzing whether it is possible for him to entertain, within the structure of his philosophical system, the existence of supra- or extra-rational knowledge, i.e., mysticism. A close reading of all the relevant passages in the works of Avicenna, and in particular the Isarat, 9 A collection of roughly the same articles as in this collection, and arranged in about the same parts as here under these headings, was published in Turkish, edited and translated by M. Ciineyt Kaya, to whom I am indebted, as before: ibn Sina ’nm Mirasi, Geni§letilmi§ 2. Baski, Istanbul: Klasik Yaymlan, 220 10. The Turkish edition has the additional feature of a bibliography, also prepared by Ciineyt Kaya, of Avicenna’s works published and translated in Turkey (Ibn Sina’nm Ttirkiye’de Ne§ir ve Tercume Edilen Eserleri), pp. 345-57.

FOREWORD

xiii

much misunderstood or deliberately misinterpreted, reveals that it is not, and indicates the way in which Avicenna understood and scientifically explained mysticism, much as he did prophecy. The essays in the final part deal with some salient aspects of Avicenna’s heritage. His potent philosophical system, which presented a self-consistent understanding and logically cogent account of all reality, including religious reality, ushered in a long period of heightened philosophical discussion in the Islamic world which I have called (rightly, it seems, as all recent research indicates) a “golden age.” A large part of this discussion was directed at transforming Avicenna-re-working, re-directing, tweaking, despoiling and re­ packaging,10 even blunting some of his ideas (which is the reason why we are faced with the issues I briefly mentioned at the outset) - and the two articles on Abdallatlf al-Bagdadl’s repudiation of al-Gazall and on Ibn Tufayl’s original creation of a mystical Avicenna give a preliminary picture of how it was done. Writing the full account of these developments as they relate to Avicenna will be, in effect, writing the history of philosophy in the Islamic world during the three centuries following his death. The article on the state of the field of Avicennan studies - our path to the recovery of the historical Avicenna finally, closes the volume. For this reprint, I wish to thank Dr John Smedley for accepting the book in the Variorum series, and both him and the Variorum editor, Lindsay Farthing, for seeing it through the press with consummate professionalism and exemplary cooperative spirit. I am also grateful to the publishers of the original articles for permission to reproduce them. However, for technical and also copyright reasons, it proved necessary to re-set the text of five articles (nos. I, III, VII, XII, and XIII) rather than reproduce it as it appeared in the original publications, and this happily afforded the welcome opportunity to revise and update them. The usefulness of the volume is further increased by the addition of exhaustive indexes of names, places, subjects, and technical terms. It is hoped that the enhanced appearance of these studies here between two covers, and further historical research which they are intended to promote, will bring us closer to the day when we will be able to distinguish Avicenna from “Avicenna transformed.” DIMITRI GUTAS New Haven Bastille Day 2014 10 This is the felicitous and very apposite description of al-GazalT’s method, in the construction of his philosophy and theology, by Alexander Treiger, Inspired Knowledge in Islamic Thought, Fondon and New York: Routledge, 2012, p. 103.

ACKNOWLEDGEMENTS Grateful acknowledgment is made to the following publishers for kind permission to reprint the following articles in this collection: Encyclopaedia Iranica Foundation, Inc. and Ehsan Yarshater (articles I, XIII); Quaderni di Studi Arabi, formerly of Universita degli Studi di Venezia, now published by the Istituto per l’Oriente C.A. Nallino, Rome (II); Manuscripts o f the Middle East, Ter Lugt Press in Leiden, and J.J. Witkam (III); Institute of Classical Studies, University of London, and Wiley Publishers (IV); Cambridge University Press (V); Koninklijke Brill NV, Leiden (VI, VII, VIII, XVI); Wiley Blackwell Publishing and Hartford Seminary (IX); Princeton Papers/ Markus Wiener Publishers, Princeton (X); Peeters Publishers and the Department of Oriental Studies, Leuven/Louvain (XI); Brepols Publishers, Tumhout (XII); De Wulf-Mansioncentrum and Leuven University Press, Leuven (XIV); Warburg Institute Colloquia, The Warburg Institute, London, and Nino Aragne Editore, Turin (XV); SISMEL - Edizioni del Galluzzo, Florence (XVII). Every effort has been made to trace all the copyright holders, but if any have been inadvertently overlooked the publishers will be pleased to make the necessary arrangement at the first opportunity.

I

Avicenna: Biography 1

[67a] Sources. Avicenna’s biography presents the paradox that although more material is available for its study than is average for a Muslim scholar of his caliber, it has received until recently relatively little critical attention. The very existence of the Autobiography and Juzjanl’s Biography, both endlessly retold, paraphrased, and elaborated upon, seems to have inhibited from the very beginning further research into additional sources and critical analysis of those available. The assessment first made by Ebn Abl Osaybe'a in the middle of the 7th/13th century has been valid ever since for both Muslim and Western scholars: Avicenna “mentioned his personal circumstances and described his own life in a way that relieves others of describing it again.”2 A comprehensive and critical study of Avicenna’s life will have to draw on the following four categories of sources: 1. The Autobiography/Biography complex and its recensions and derivatives. Avicenna’s Autobiography covers the period from his birth until after his encounter with Abu cObayd 'Abd -al-Wahed JuzjanI, his disciple, amanuensis, and constant companion.3 JuzjanI wrote down the Autobiography (either from dictation, according to Ebn [67b] al-Qeftl,4 or from a draft originally penned by Avicenna), and appended to it the Biography, covering the rest of Avicenna’s life. This combined original document, in Arabic, exists in at least two recensions. One is embedded in the accounts of Avicenna by Ebn al-Qeftl (pp. 413 - 26) and Ebn Abl Osaybe'a (II, pp. 2 - 9), and the other exists independently in various manuscripts, in some of which it is also given the Persian title, Sargodast. The relationship of the two recensions to Juzjanl’s text and to each other is 1 This is a slightly revised and updated version of the entry in Encyclopcedia Iranica, Vol. Ill, Fasc. 1 (1987), pp. 67 -70. Its transliteration system has been retained. The page numbers in that publication are entered here in the text in brackets in bold. 2 Ketab "oyun al-anba ’fita b a q d t al-atebba \ ed. A. Muller, Cairo, 1882 -84, II, p. 2. 3 See on him the article “Jowzjani” by Robert Wisnovsky in Encyclopcedia Iranica, Vol. XV, Fasc. 1 (2009), pp. 82—4. 4 Ta 'rik al-hokama \ ed. J. Lippert, Leipzig, 1903, p. 413.

I 2

Avicenna: Biography

not entirely clear. W.E. Gohlman, the editor of the Sargodast, tends to think that it is closer to the original.5 Juzjam’s edition of the Autobiography/Biography forms the only source for all subsequent narrative accounts of Avicenna’s life. There is no indication that there ever was an original rival biography, while all later retellings merely repeat JuzjanT’s account in different degrees of fidelity to his (and Avicenna’s) wording. These retellings differ only in the various details they introduce, which, however, have to be proven authentic in each case before they can be accepted. Even during his lifetime Avicenna had achieved great fame or notoriety (depending on the viewer’s standpoint), and stories about him with essentially hagiographic or demonographic content doubtless began to circulate soon after his death. The trend continued until later popular tradition made of him either a saint and a mystic or a magician. The most important retelling of the Autobiography/Biography, which sometimes is erroneously treated as an independent source, is that by Zahlr-al-dln Abu’l-Hasan BayhaqI in the Tatemmat sewan al-hekma. BayhaqI recasts the Autobiography in the third person, slightly paraphrases the Biography, omits certain details, mostly of a bibliographic nature, and adds some others. These additions, which in SafT’s, edition (Lahore, 1935) are conveniently enclosed in double brackets, occur mostly in the Autobiography part, and are informative (provide names of personalities mentioned, refer to historical events), bibliographic (report on the fate of some of Avicenna’s lost books), calumniatory (blame Avicenna for being the first philosopher to indulge in wine and sex, and to frequent royal courts), and anecdotal (e.g., the story of cAla3-al-dawla’s sister and the Ghaznavid Sultan Mas'ud, pp. 55 - 6, and Avicenna and the youth from Ray, pp. 59 -61). Except for the bibliographic reports about the survival of some of Avicenna’s books, not much of this additional information can be taken at face value. BayhaqT’s gratuitous comments on Avicenna’s alleged debauchery, his arbitrary extrapolations from the material at hand, such as the mention that Avicenna used to study the R asa3el Ekwan al-Safa \ and the anecdotal material merit little credence; but even the information about the names of certain personalities should be questioned. There is no reason why the names of Avicenna’s mother, and of the greengrocer from whom Avicenna learned arithmetic as a young boy, should have survived 5 W.E. Gohlman, The Life o f Ibn Sina, Albany, N.Y., 1974, p. 6. Nothing definite, however, can be said before the entire biographical tradition of Avicenna ’s life and its manuscript transmission have been studied.

I Avicenna: Biography

3

unaltered, or at all, for more than [68a] 150 years until Bayhaqi’s time. Such information, even if it had survived orally in the popular tradition, lends itself easily to falsification. Similar criticism must be applied to all details added in the derivatives of JuzjanT’s edition of the Autobiography/Biography. These derivatives are the accounts of Avicenna’s life by subsequent biographers, from Ebn Kallekan ( Wafayat al - ayan) and Sahrazuri {Nozhat al-arwah) to Ebn al- Emad {Sadarat al-dahab) and ICandamTr {Habib al-siar and Dostur al-wozard ). The manuscript tradition of JuzjanT’s edition also shows accretions and contaminations from various sources. In one instance, the recension in Ebn al-Qeftl/Ebn Abl Osaybe'a is contaminated with the version of BayhaqI, in the Vienna MS Ar. Mixt. 866, 8.6 In another, material intended to attribute his death to his alleged sexual depravity has been interpolated;7 etc. A critical edition of the entire Autobiography/Biography complex is a major desideratum. 2. Private writings by Avicenna and his disciples. These constitute a reliable source about numerous details of Avicenna’s life. They consist of his many autobiographical references in the prologues, epilogues, and occasionally even in the body of his own works, of personal writings in the form of correspondence, accounts of controversies in which he was involved, homilies and devotional pieces, and possibly even poetry,8 and of similar writings by his disciples, including JuzjanT’s introduction to the S efa ? This material, which has not been systematically tapped for Avicenna’s life, is preserved partly also in the collection of his Nachlass known under the title al-Mobahatat, two recensions of which were edited by cA. BadawT

6 H. Loebenstein, Katalog der arabischen Handschriften der osterreichischen Nationalbibliothek, Vienna, 1970, p. 212, no. 2430. 7 See Joep Lameer, “Avicenna ’s Concupiscence,” Arabic Sciences and Philosophy 23 (2013) 277 - 89. 8 See the list in the inventory of his works in D. Gutas, Avicenna and the Aristotelian Tradition. Introduction to Reading Avicenna s Philosophical Works. Second, Revised and Enlarged Edition, Including an Inventory o f Avicennas’ Authentic Works, Leiden, 22014, pp. 502 -11. 9 G.C. Anawati et al., eds, Ebn Sina. A l - Sefa\ al-Manteq, al-Madkal, Cairo, 1371/1952, pp. 1-4.

I 4

Avicenna: Biography

and M. Bidarfar,10 and partly independently in the manuscripts, yet to be fully researched.11 3. Historical works. As a rule they have little information to add to Avicenna’s life, but they are useful for providing the background for many of the social and political events referred to or hinted at in the Autobiography/Biography. 4. Legendary and hagiographic stories. These belong not to his biography proper but to a study of the transformation of his image in popular tradition after his death. This played a role in the reception of his authentic works in the Persian- and Turkish-speaking areas of the Islamic world and has to be studied as a separate subject. The pride of place among the legendary material belongs to Nezam IcAruzl’s Cahar maqala.12For later material see the article by Unver.13 Analysis o f the Autobiography. For the first part of Avicenna’s life, our sole source of information is the Autobiography. The real purpose of this document is philosophical: While purporting to give details about his early life - details which, in the absence of contrary, or any other, evidence, may have to be taken at face value - Avicenna is providing a concrete illustration of his epistemological theory. This centers on the ability of some individuals with powerful souls to acquire intelligible knowledge all by themselves and without the help [68b] of a teacher through their propensity to hit spontaneously upon and to guess correctly the middle terms of syllogisms, hads.u The Autobiography is written from the perspective of a philosopher who does not belong by training to any school of thought 10 A. Badawi, Arestu And al- Arab, Cairo, 1947, pp. 119-249; and M. Bidarfar, AlMobahatat, Qom, 1371 S/1992. 11 For which see D.C. Reisman, The Making o f the Avicennan Tradition: The Transmission, Contents, and Structure oflbn Sind s al-Mubahatat {The Discussions), Leiden, 2002 .

12 Edward G. Browne, Revised Translation o f the Chahdr maqala ( “Four discourses ”) ofNizami-i­ Arudi ofSamarqand, followed by an abridged translation ofMirzd Muhammad s notes to the Persian text, Cambridge, 1921. 13 A. Sliheyl Unver, “ §ark folklorun ’da Ibni Sina hakkmda ya§ayan ve kaybolan efsaneler,” in his ibni Sina, Hayati ve Eserleri Hakkmda (fahymalar, Istanbul, 1955, pp. 62 -70. 14 See the analysis of this key concept of Avicenna’s epistemology in Gutas, Avicenna and the Aristotelian Tradition (22014), pp. 179 -201.

I Avicenna: Biography

5

and is therefore not beholden to defending it blindly, who established truth through his independent verification (hads) and found that for the most part this truth is contained in the philosophical sciences as classified and transmitted in the Aristotelian tradition, and who is therefore in a position both to teach this more accurate version of truth - or revised Aristotelian philosophy ­ and to judge the attainment in philosophy of others. For this reason, when the Autobiography is reduced to its bare essentials, it appears as a transcript of Aristotelian studies and a model curriculum vitae in a Peripatetic program. Avicenna reports in it basically that he studied the philosophical sciences according to their classification in the Aristotelian tradition, and that he studied them in three successive stages at increasingly advanced levels. Everything else that Avicenna says he studied was studied not for its own sake but for that of the philosophical sciences, in whose terms it is to be seen. The elementary courses in the Koran, literature and arithmetic, are preparatory subjects for the philosophical sciences. Avicenna’s rejection of the IsmaTli teachings is intended to indicate that already at an early age he could refuse, through his own reflection, authoritative knowledge, taqlid. (BayhaqT’s unwarranted addition at this point, that Avicenna would read and reflect on the Rasa 'el, as already mentioned, caused many a misunderstanding and generated the myth of the IsmaTli Avicenna.15Avicenna nowhere indicates that he had any interest in the doctrine or in the kind of thinking it involved.) His study offeqh with the Hanafite jurist EsmaTl Zahed (d. 402/1012) is mentioned only to refer to the method of the discipline and to justify his knowledge of and practice in Aristotelian dialectics (as expounded in Aristotle’s Topics, Book 8). On this basis is Avicenna then able to analyze the question “What is it?” and amaze his teacher NatelT. Medicine, finally, though it belongs to the canon of sciences in the Greek Aristotelian tradition, is not a theoretical, but a practical science. Therefore it is easy, i.e., its acquisition does not require the solution of syllogisms by guessing correctly middle terms, but merely reading the texts and medical practice. Avicenna then says that he studied the philosophical sciences - logic, mathematics, physics, metaphysics, in that order - in three stages: first, initially with NatelT and finally on his own, second, entirely on his own, and third, at a research level, in the physicians’ library in the palace of the Samanid ruler.

15 See D. Gutas, “Avicenna ’s Madhab, With an Appendix on the Question of His Date of Birth,” Quaderni di Studi Arabi 5 - 6 (1987 -1988) 323 -36 (below, selection no. II).

I 6

Avicenna: Biography

The systematic nature of Avicenna’s presentation of the actual course of his studies, and its close correspondence to the theoretical classification of the philosophical sciences, raise the question whether Avicenna presented a stylized Autobiography in which the chronology of events is bent to fit the theoretical classification of the sciences. The issue cannot be resolved, [69a] insofar as the classification of the sciences in the Aristotelian tradition influenced actual educational practice which in turn is presented in an autobiographical account reproducing that very classification in order to promote it. However, the exact historical sequence of events in Avicenna’s studies is not as important as the point which the Autobiography intends to make, as discussed in the first paragraph of this section. Avicennas life. Bukhara (370/980[?]-389/999[?]). Avicenna was bom before the year 370/980, which is the traditional dating , and most likely quite a few years earlier than that. It has not yet been possible to establish the date of his birth with greater precision; there are enough inconsistencies and contradictions in the transmitted chronology, however, to make the traditional date quite untenable.16 He was bom in Afsana, a village near Bukhara. His father, who had moved in from Balk a few years previously, was the Samanid governor of nearby Karmaytan. A few years after his birth, the family moved to Bukhara. The intellectually active capital attracted scholars, and Avicenna had an excellent education. Although for the reasons mentioned in the preceding section Avicenna is reticent about his teachers, it is almost certain that he studied with more scholars than the Natell that he mentions in the Autobiography; the names of the physicians Abu Mansur Qomrl and Abu Sahl Maslhl are also mentioned among his teachers. Given the availability of teachers and libraries, his father’s high position in the Samanid administration, and his own application and precocity, Avicenna was perfectly schooled in the Greek sciences by the time he was eighteen. Avicenna began his professional career around the age of seventeen, when he was enrolled as a physician in the service of the Samanid Nuh b. Mansur (reigned 365/976-387/997) whom he was summoned to treat. After the death of his father a few years later - according to the chronological sequence of the events as described in the Autobiography, after he was twenty-one - he was also given an administrative post, perhaps a district governorship. The fact that in the Autobiography the death of his father

16 R. Sellheim in Oriens 11 (1958) 238; D. Gutas “Avicenna ’s M adhab ” Appendix.

I Avicenna: Biography

7

and his assumption of administrative duties are mentioned closely together justifies the speculation that he may have succeeded his father as governor of Karmaytan. By a relatively early age Avicenna was established, in his twin capacity as physician and political administrator, in a profession that he was to practice, in search of patronage,17 in the courts of various Iranian rulers, heads of the numerous successor states that emerged during the period of the disintegration o f cAbbasid authority. Gorganj (ca. 389/999[?]—402/1012). Avicenna remained in Bukhara until, as he puts it, “necessity called” him to leave for Gorganj in Kvarazm, where he j oined the service of the M a3munid Abu ’1-Hasan£All b. M a3mun.18 Since the latter reigned from 387/997 to 399/1009, Avicenna could have moved to Gorganj any time between these two dates; but the fact that the “necessity” to which he refers can be interpreted in any satisfactory way only in political [69b] terms suggests the following. The Samanid state was overthrown by the Turkish Qarakhanids who entered Bukhara in 389/999 and took cAbd-al-Malek II, the last amir, prisoner. Avicenna, a high functionary of the fallen state and strongly identified with the Samanid dynasty, may have found his position, to say nothing of his job, difficult to maintain. Furthermore, it may not be entirely fortuitous that soon after the fall of Bukhara to the Qarakhanids, EsmaTl Montaser, the Samanid prince, also went to Kvarazm to seek support for a political comeback. Avicenna may or may not have been involved in the undertaking, but it appears that it was the events of 389/999 and those immediately following them that generated the circumstances which made Avicenna’s departure from his home town necessary.19 From Gorganj to Jorjan (402/1012-403/1013). Avicenna left Gorganj for the same unspecified reason - “necessity called” him - and traveled south into Khorasan and then west. During the journey he passed through Nasa, Ablvard, Tus, Samangan, Jajarm, and arrived at Jorjan (Gorgan) only to find that the Ziyarid amir Qabus b. Vosmglr, his prospective patron,

17 See the details of Avicenna ’s peregrinations and political and intellectual controversies in the courts of various local rulers in D.C. Reisman, “The Life and Times of Avicenna: Patronage and Learning in Medieval Islam,” in P. Adamson, ed., Interpreting Avicenna, Cambridge, 2013, pp. 7 - 27. 18 See the article “Al-e M a ’mun ” by C.E. Bosworth in Encyclopaedia Iranica, vol. I (1985), pp. 762—4. 19 Cf. G. Liiling, “Ein anderer Avicenna. Kritik seiner Autobiographic and ihrer bisherigen Behandlung,” Zeitschrift der deutschen morgenlandischen Gesellschaft Suppl. 3/1, 1977, pp. 496 - 513, at 499.

I 8

Avicenna: Biography

had died in the meantime (winter months of 403/January-March, 1013). Avicenna’s report in the autobiography is too brief to provide any hints about the reasons behind this odyssey, although political considerations would again seem to be the only plausible answer. As for the duration of his travels, Avicenna does not mention that he stayed or worked in any of these locations, so in all likelihood he left Gorganj in 402/1012. In Jorjan (403/1013 -ca. 404/1014) Avicenna met Juzjanl. He spent little time there, apparently in the employ of Manucehr b. Qabus, and lived in the house of a private patron. Ray (ca. 404/1014^05/1015). From Jorjan Avicenna moved to Ray, where he joined the service of the Buyid Majd-al-dawla Rostam and his mother Sayyeda, the power behind the throne. Although he had with him letters of recommendation for his new employers, it appears that he gained access to the political elite of the Jebal again through his skill as a physician. He treated Majd-al-dawla who was suffering from a black bile disease (i.e. psychic disturbance). Hamadan (405/1015 -ca. 415/1024). Avicenna remained in Ray until the Buyid Sams-al-dawla, Majd -al-dawla’s brother, attacked the city after Du’l-qa'da, 405/April, 1015. Then he left for Qazvln, again for reasons unspecified, and finally arrived in Hamadan where he was summoned to treat Sams-al-dawla. Inevitably, Avicenna became also Sams-al-dawla’s vizier and acted in this capacity (with an occasional conflict with the amir’s troops) until the latter’s death in 412/1021. The new amir, Sama3-al-dawla, asked Avicenna to stay on as vizier, but “Avicenna saw fit not to remain in the same state nor to resume the same duties, and trusted that the prudent thing for him to do . . . would be to hide in anticipation of an opportunity to leave that region.”20 He secretly corresponded with the Kakuyid cAla5-aldawla in Isfahan about this [70a] matter. The Buyid court in Hamadan, and especially Taj-al-molk, the Kurdish vizier, suspected Avicenna of treachery because of these moves, and they arrested and imprisoned him in a castle outside of Hamadan called Fardajan. Avicenna remained in prison for four months until cAla5-al-dawla marched toward Hamadan and ended Sama5al-dawla’s rule there (414/1023). Released from prison in the wake of these developments, Avicenna was again offered an administrative position in Hamadan, but he declined. Some time later he decided to move to Isfahan and he left secretly Hamadan with his brother, Juzjanl, and two slaves, dressed like Sufis.

20 Juzjam, Sefa \ Madkal, p. 2, translated in Gutas Avicenna 32, §4.

I Avicenna: Biography

9

Isfahan (ca. 415/1024 -428/1037). cAla5-al-dawla received Avicenna with honors, and gave him, in JuzjanT’s words, “the respect and esteem which someone like him deserved.”21 Avicenna finally settled in Isfahan and remained in cAla3-al-dawla’s employ until his death. He accompanied his master in most of his campaigns and trips, and indeed it was during one such trip to Hamadan that he died, in 428/1037, of colic, after a protracted series of recoveries and relapses. He was buried in Hamadan. Apart from his scholarly persona, which is one of unprecedented energy and sharpness, we get almost no glimpse of Avicenna’s character. He was a self-conscious boy prodigy, professionally successful at an early age; also at an early age he became a permanent exile from a home that ceased to exist; he was forced to serve petty rulers most of whom not only did not appreciate his special genius but did not even esteem him as an intellectual; and yet he somehow stayed with these rulers and seems to have been determined to avoid, for reasons that we can only guess, the Ghaznavid court. His peregrinatory circumstances and his search for patronage, along with his radically innovative but avowedly brilliant ideas, involved him in controversies of more than academic significance.22 The combination of these factors, among many others, could produce either a hero or a villain: it does not help to speculate before all the available evidence, especially his private writings, is assessed. One thing, though, is certain, and this has again to do with his scholarly self: When it came to intellectual matters, Avicenna could accept no rival and decline no challenge. When he was slighted for his ignorance of Arabic lexicography, he answered the affront by memorizing AzharT’s Tahdib al-loga, forging three epistles in the styles of famed authors, and submitting them for identification to the person who had insulted him. That person failed to recognize the forgery. This also, however, depending on the surrounding circumstances, which we do not know, could be interpreted either as extreme arrogance, or proper estimation of self worth, or even as an exaggerated sense of humor. Avicenna the person is barely distinguishable behind the brilliance of Avicenna the mind.

21 Biography, Gohlman 65. 22 For his controversies with Abu -l­Qasim al-Kirmanl see Reisman “Patronage ” and his references there, along with the edition of some of his private writings and their study by Y. Michot, Ibn Sind. Lettre an vizir Abu Sa cd, Beirut/Paris, 2000, and his “ Le Riz trop cuit du Kirmani. Presentation, edition, traduction et lexique de VEpitre d ’Avicenne contestant Vaccusation d ’avoir pastiche le Coran,” in Melanges offerts a Hossam Elkhadem par ses amis et eleves, F. Daelemans et ah, eds, Archives et Bibliotheques de Belgique, Numero Special 83, Brussels 2007, 81 -129.

I 10

Avicenna: Biography

But for Avicenna, who saw the supreme happiness in the contact of the human intellect with the active intellect during the split-second of hitting upon the middle term, perhaps this is just the way it should be. Bibliographical Note: General bibliographical information on Avicenna can be found, in print, in the indispensable work by J. Janssens, An Annotated Bibliography on Ibn Sina (1970-1989% Leuven, 1991, and its First Supplement (1990 -1994), Louvain-la-Neuve, 1999; second supplement (2014) forthcoming. The information available on the Internet is uncontrolled and should be used, as always, with due caution. The only critical edition of the Arabic Autobiography/Biography in the independent recension (Sargodast) is that by Gohlman (above, note 5). The edition, which contains a facing English translation, needs to be done anew; [70b] see the review by M. Ullmann inDerIslam 52 (1975) 148-51, and Gutas Avicenna (above, note 8) 10-11 and note a. JuzjanT’s biography has not yet been subjected to a critical study. Particularly perceptive analyses of the autobiography, on the other hand, are offered by R. Sellheim in his review of Ergin’s Ibni Sina Bibliografyasi (1956), in Oriens 11 (1958) 231 - 39, and by G. Liiling (above, note 19). Gutas Avicenna 10-19, 169-225, contains a translation and an analysis of the Autobiography. Among scholarly translations, the earliest, and one of the most accurate, is that by P. Kraus, “Eine arabische Biographie Avicennas,” Klinische Wochenschrift 11 (1932) 1880b-1882b. The very readable, but not always as accurate, English translation by A.J. Arberry was first published in his Avicenna on Theology, London, 1951, pp. 9 - 24, and later reprinted in a number of publications (listed by Gohlman on p. 117, note 28). Among many other translations, the following may be mentioned: French, by M. Achena and H. Masse, in Le livre de science, Paris, 1955,1, pp. 6 -11; Persian, by G.H. Sadlql, Sargodast-e Ebn Sina, Tehran, 1331 S./1952; and Spanish, by M. Cruz Hernandez, La vida de Avicena como introduccion a supensamiento, Salamanca, 1997. The bibliography on the Autobiography/Biography is extensive. There is a useful collection of the traditional material in S a id NafisI, Zendagi o kar o andlsa o ruzgar-e pur -e Sina, Tehran, 1333 S./1954. A list of western European publications is given in Gohlman and supplemented by Ullmann in his review. References to eastern European publications can be found in

I Avicenna: Biography

11

B. and S. Brentjes, Ibn Sina (Avicenna), derfurstliche Meister aus Buhara, Leipzig, 1979, andA.V. Sagadeev, Ibn-Sina, Moscow, 1980. See also S.M. Afnan, Avicenna: His Life and Works, London, 1958; D. Safa, Jasn-namaye Ebn Sind I, Tehran, 1331 S./1952. Idem, TarTk-e dlum-e ‘aqli dar tamaddon-e eslam il, Tehran, 1331 S./1952, pp. 206-81.

II

AVICENNA’S MADHAB WITH AN APPENDIX ON THE QUESTION OF HIS DATE OF BIRTH

I The question of Avicenna’ m adhab , or formal affiliation with an Islamic legal rite, has rarely been addressed with any degree of seriousness. Like a number of issues concerning his life and works — issues which, in the case of other Muslim scholars of much lesser stature, would have been settled as a matter of fact long ago — it has been treated largely in subjective and partisan ways by both Muslims and non-Muslims, and from his death onward to our very days. Discussion and exposition have frequently yielded to unsopported declarations and assertions based on false inferences, and at times even on deliberate disregard of the facts. One of the earliest such cases is a distortion perpetrated by a Muslim scholar who was destined to have a large following in posterity, Zahir al-Din al-Bayhaqi (d. 565/1169-70). Concerning the notorious incident of the discussions which Avicenna’s father and brother held with an Isma‘111 da‘i Avicenna says the following in the Autobiography: My father was among those who responded to the proselytizer o f the Egyptians and was considered one o f the Isma'ilis. He had heard from them the account about the soul and the intellect in the way in which they tell it and present it, and so had my brother. They would sometimes discuss this matter [i.e., the account about the soul and the intellect] among themselves, while I would listen to them and comprehend what they were saying but my soul would not accept it [i.e., this matter about the soul and the intellect]; and they began to summon me to it with constant talk on their tongues about philosophy, geometry, and Indian arithm etic/

BayhaqI, who recasts in his Tatimmat Siwan al-hikma the Autobiography in the third person but otherwise follows it relatively faithfully, omits this

1

Text in W.E. Gohlman, The Life oflbn Sina, Albany, 1974, p. 18.4-20.1. See the corrections and translation in D. Gutas, Avicenna and the Aristotelian Tradition, Leiden, 1988, Text 3, § 3.

II 324 entire paragraph and instead says the following: His father used to study and reflect on the Rasa ’il Jhwan al-Safi\ and he himself would also reflect on it from time to time. 2

Bayhaqi could have been following here no other source that includ-ed this detail since there is none: Avicenna nowhere says that he “would reflect on” the Rasa’il at that stage in his life, and there is no source, other than the Autobiography, that could have informed Bayhaqi of this detail. Bayhaqi’s statement, therefore, appears to be a gratuitous inference drawn from Avicenna’s account of the incident, and one that totally misrepresents it. Whatever Bayhaqi’s motives for this deliberate misrepresentation may have been 3 —he could certainly read Avicenna’s Arabic— the fact remains that it proved influential, with grave consequences for an understanding of the sources of Avicenna’s thought. As an example I may cite one of the more recent instances of the reappearance of Bayhaqi’s statement, and also one of the more prejudicial, because it occurs in a publication addressed to a large audience. In the second edition of A History of Islamic Philosophy Majid Fakhry writes the following: Apart from the works o f Aristotle and his Greek commentators, the two chief formative influences on Ibn Sina’s thought appear to have been the Epistles of the Brethren of Purity and the writings of al-Farabi. ... It is significant that the twelfth-century historian Bayhaqi makes a point o f the fact that like his father Ibn Slna was in the habit o f reading the Epistles of the Brethren of Purity . 4

Yet another inaccurate inference drawn by a Muslim scholar, also destined to have an extremely wide following down to our very days, is that of the “third” Si‘i martyr SuStari (d. 1019/1610) who argued, for the first time if I am not mistaken, that Avicenna must have been an Itna ‘a§ari §i‘i because he served a succession of Itna ‘aSari princes, i.e. the Buyids and the

2 Tatimmat Situan al-hikma, M. ShafT, ed., Lahore, 1935, p. 40.1-2; T a ’inh hukama' al-Islam, M. Kurd ‘Ali, ed., Damascus, 1365/1946, 2nd ed. 1396/1976, pp. 52-53. Both ShafT and Kurd ‘All write Risala instead of Rasa % following some manuscripts, while others have the plural. 3 For the method followed by Bayhaqi in composing his biography of Avicenna, see the article “Avicenna: Biography ” in the Encyclopcedia Iranica (D. Gutas). 4 London -New York, 1983, p. 132. In footnote 93 on that page, Fakhry refers the reader to “Al-Bayhaqi, MuntakhabSuwan [sic!] al-Hikmah, p. 40 ”. This is doubly wrong: the correct title and references are given in note 2 above. Interestingly enough, Fakhry does not seem to be aware o f ShafT’s edition; in the Bibliography (p. 386), he cites the Damascus edition and two manuscripts, Bodleian Marsh 539 and Istanbul Koprulu 902. Of these, only the latter is a manuscript o f the Tatimma (see D. Gutas in the Journal of the American Oriental Society, 102 [1982], 646b); the former is an anonymous collection o f quotations ascribed to Greek authors (see F. Rosenthal in Islamic Culture, 14 [1940], p. 396, note 1).

II 325 KakUyids 5. Given our knowledge of the liberal religious policies of the Bayids, this argument would have been judged invalid on that basis alone; but it also happens that it completely disregards the fact that Avicenna joined the service of a SIT ruler for the first time after he was about forty, viz. ca. 404/ 1014 in the court of Magd al-Dawla in Rayy 6. Nevertheless, the argument has continuously reappeared since then, and again most prejudicially, in another history of Islamic philosophy of wide diffusion, Henri Corbin’s Histoire de la philosophie islamique (Paris, 1964, pp. 238-239): [Avicenne] s ’il se deroba devant le shi’isme ismaelien, l ’accueil qu ’il re^ut aupres des princes shi’ites de Hamadan et d ’Ispahan, permet au moins d ’inferer qu ’ il ait appartenu au shi’isme duodecimain.

These two views are indicative of the superficial way in which the question of Avicenna’s madhab has been preponderantly treated, and of the tenuousness of the arguments proffered for its substantiation; they are also representative of the beliefs most widely held about Avicenna’s madhab, namely that he was an Itna ‘a&ari and/or that he had IsmaTli leanings '7. At this stage the point must be made that the question itself is not important for a substantive understanding of Avicenna’s life and works; his philosophical and scientific activities and concerns were beyond the level at which differences of madhab could play a significant role. The question nevertheless must be addressed for two reasons. First, the impression which scholars have of Avicenna’s madhab, based as it is on statements like the above, passes as fact and creates a rarely ackowledged (or even conscious) filter which distorts the way in which his life and work are understood and interpreted. I am referring here specifically to the widely held myth of Avicenna’s “mysticism” 8, nourished by the claims of his alleged SIT — IsmaTli or Itna ‘aSari — tendencies. Second, Avicenna himself, or more precisely, what he says in the Autobiography, requires that the question be ^ 6 17

8

Magalis al-mu mirtin, Teheran, 13545, II, p. 185. The periodization o f Avicenna ’s life is set forth in the article “Avicenna: Biography ” (no ­ te 3 above). See the Appendix for the arguments in favor o f an earlier birthdate for Avi­ cenna than the traditional 370/980. In Persian scholarship, the belief that he was an Itna ‘aSari is universal. See, for example, among the most recent authoritative biographies, Z. Safa, Le livre du millenaire d Avicenne (a­ daptation frangaise par...Said Naficy, [Societe Iranienne pour la Conservation des Monu ­ ments Nationaux; Collection du Millenaire d ’Avicenne, n° 27], Teheran, 1953, p. 46; A. Rizvanl, Abu AU Stm, Teheran, 13445/1965, pp. 169-176. For Avicenna ’s alleged “mysticism ” and “oriental philosophy ” see the article “Avicen ­ na: mysticism ” in the Encyclopadia Iranica (D. Gutas). The myth o f Avicenna ’ mysti­ cism has been recently also exposed by M. Turker-Kuyel in an article that builds upon and elaborates the theses o f A.-M. Goichon: “Ibn Sina ve ‘mistik ’ denen goru$ler ”, in Aydm Sayili, ed., Ibn &na. Dogumunun Bininci Ytli Armagam [Turk Tarih Kurumu Yaymlari, VII. Dizi, Sayi 80], Ankara, 1984, pp. 749-792.

II 326 answered seriously since he refers explicity to having had formal engagements with fiqh. Avicenna says in the Autobiography that as a young boy he studied fiqh with a certain Isma‘il al-Zahid. Obviously this scholar must have belonged to a madhab , and it was the teachings of this madhab that he imparted to Avicenna. Furthermore, Avicenna says that during his mid ­ teens he used to participate in disputations with legal scholars. These men all certainly belonged to a madhab, according to whose positions and rules the disputations in which Avicenna participated were conducted, and Avicenna must have accordingly been proficient in these positions and rules. Or, if it were to be assumed that these scholars belonged to different schools — an assumption which, as well shall next see, has to be discounted — then Avicenna’s contribution must have been according to the madhab in which he had been schooled, that of IsmaTl al-Zahid. Finally, and most importantly, after the fall of the Samanids, Avicenna sought refuge in Gurgang at the court of the HwarazmSah ‘Ali b. Ma’mun. There, he tells us, he earned his living practicing law: he wore a taylasan , the garment p a r excellence of fiqh scholars, and he drew a salary from the court for precisely these services. The law he practiced must have been according to a specific rite, that favored at the court of ‘Ali b. Ma’mun, and it is difficult to imagine that it could have been anything else but the madhab in which Avicenna had been trained and gained the reputation of a highly skilled practitioner. After he left Gurgang, there is no mention at all in the sources that Avicenna ever practiced law again. He served the Bayid rulers, these sources tell us, as a physician and a vizier (although there are problems even with this latter function, as Claude Cahen once perspicaciously oberved, 9 but this is another problem). The question then that has to be asked is, which was this madhab in which Avicenna was schooled, debated, and was proficient enough as to practice it professionally? The evidence, both direct and indirect, points to the fact that was Sunni Hanafi10. 9

C. Cahen, “A propos d ’Avicenne ”, La Pensee. Revue du Rationalisme Modeme, 45 (1952), p. 81. This position has already been advanced in some detail, and a substantial amount o f the evidence for it has been presented by Mustafa Gawad in 1955 during the conference held in Teheran in celebration o f Avicenna ’s millenary: “Al-laqafa al-‘aqlfya wa-l-hal aligtima'iya fi ‘asr al-Ra’Is Abi ‘All b. Sina”, Le Livre du millenaire d Avicenne (Kitdb al-Mihrigan li-Ibn Sina), vol. Ill [Societe Iranienne pour la Conservation des Monuments Nationaux; Collec­ tion du Millenaire d ’Avicenne, n° 32], Imprimerie de l ’Universite de Teheran, Teheran, 1 3 7 6 / 1 3 3 5 S /1956, pp. 248 -280. The article, however, has remained largely unheeded; thus, for example, Gohlman still maintained in 1974 that Isma‘il al-Zahid was “not listed in the standard biographical dictionaries” (Ibn Sina, p. 121, note 6), despite Gawad’s copious references (p. 269, note 1) to the Hanafi tabaqat and other works. (If I may hazard a conjecture, I believe that the reason Gohlman did not identify Isma'il al-Zahid is that he

326

II 327 II The direct evidence is unambiguous. Avicenna’s teacher of fiqh , Isma‘11 al-Zahid, was a prominent Hanafi scholar in Buhara. The earliest report about him comes from his younger contemporary, the historian of Buhara Gungar (d. ca. 412/1021 -1022), who states, in a manuscript note report­ ed by al-Hatib al-Bagdadi (d. 463/1071), that Isma‘il died on Wednesday (or rather, Tuesday), 8 5a‘ban 402/5 March 1012. The Hatlb provides further information about him. He gives his full name as Abu Muhammad Isma‘11 b. al-Husayn b. ‘All b. al-Hasan b. Harun al-Faqlh al-Zahid al-Buhari, and says that he visited Bagdad a number of times on his way to or from Makka for the Pilgrimage. In Bagdad he transmitted hadit on the authority of a number of scholars. Two of these visits, attested to by the scholars who transmitted hadit from him to the Hatlb, took place in 387/997 and 398/ 1008 n . Most of this information is repeated verbatim by Ibn Abi 1-Wafa’ al-QuraS (d. 775/ 1373), the biographer of Hanafi scholars, who only adds on his own that Isma'il was one of the leading Hanafi jurists of his time in both furu ‘ and u su l12. This is as to be expected; Isma‘11 learned his fiqh from Abu Bakr Muhammad b. al-Fadl al-Kamari (d. Friday, 24 Ramadan 381/4 December 991), 13 certainly the most distinguished Hanafi jurist in the Samanid court. The fact that Avicenna had a Hanafi education survived in the Hanafi scholarly tradition, which openly claimed him as its own and vouchsafed him a brief entry in its tabaqat and other works. Significantly, the Hanafi tradition did so independently of Avicenna’s report in the Autobiography

neglected to look for him in the Hanafi “standard biographical dictionaries ” because he assumed him to be an l£na ‘aSari). Furthermore, since Gawad’s article was not specifically on Avicenna ’s madhab, it treated the matter somewhat tangentially, without either present­ ing a sustained argument for the position or documenting it in the fullness it deserves. Gawad approached the question in an effort to explain the move o f Avicenna ’s father from Balh to Buhara; he advanced the argument that it was in order to escape from Saffi controll­ ed areas (Balh under Sebuktegin) to those under Hanafi jurisdiction (Samanids), and he accordingly discussed in this context the H anafi credentials o f Avicenna and his immediate milieu (pp. 264 - 274). Nevertheless, Gawad made the following points, from among those presented here, in support o f Avicenna ’s Hanafi affiliation: Avicenna had an aversion to Isma'ilism as he states in the Autobiography (p. 266); Avicenna grew up in a Hanafi environment (p. 268); his teacher, Isma'il al-Zahid, was a prominent Hanafi jurist (pp. 268-269); wine drinking is allowed by the Hanafiya (p. 270); Avicenna ’s neighbor, Abu Bakr al-Baraqi, was a Hanafi (p. 274); Gurgang is a Hanafi area (p. 275). 11 Ta ’nh Bagdad, Beirut, 1968, VI, pp. 310-311, n° 3355. *Imam waqtihifi l-furu* wa-l-fiqh”, in al-Gawahir al-mudx a fi tabaqat al-Hanafiya, Hyderabad ed., I, pp. 147-148. The date of Isma‘H’s first listed visit is given m Ibn Abi 1-Wafa” s text as 389 rather than 387 (tis'a for sab'a) . For further references to biographical notices on Isma'il see Gawad “ *Asr al-Ra’is”, p. 269, note 1. I3 This is reported by Ibn Abi 1-Wafa’ in the biography o f Abu Bakr (vol. II, p. 107) but not in that of Isma'il himself.

327

II 328 about Isma‘il al-Zahid, and on the basis of a different Hanafi pedigree, going back to Abu Bakr Ahmad al-Baraqi. Abu Bakr was a member o f the distinguished Baraqi family which dominated the scholarly and Hanafi legal scene in Buhara in the 4th - 5th/10th - llth centuries. His father, Abu ‘Abd Allah Muhammad al-Baraqi, a scholar of law and language, brought his family from Hwarazm to Buhara, where they setded. The Son, Abu Bakr Ahmad b. Muhammad, distinguished himself in adab, tasawwuf and kalam, and died in Muharram 376/May-June 986, at the age of sixty-three. The Grandson, AbU ‘Abd Allah Muhammad b. Abi Bakr, served successively as the qadi of Buhara, as the vizier of the Qarahanids, and Finally as their governor in Buhara. A noted hadit scholar, he taught Tirmidi’s CdmV to ‘Ali b. Hibat Allah Ibn Makula, the author of the Ikmdl (b. 422/1031). Ibn Makula, who reports this genealogy, has it, he says, from his teacher, the Grandson, and then adds the following: I saw the eRwan of [the Son] Abu Bakr Ahmad ’s poetry, most o f which was in the handwriting o f his student, the philosopher Avicenna *4.

Ibn Makula, it appears, was shown the volume by his informant, the Baraqi Grandson and the poet’s son, Abu ‘Abd Allah b. Abi Bakr. If that is the case, and there seems to be little doubt about it, then Ibn Makula must have taken the designation of Avicenna as “the student” of AbU Bakr Ahmad also from him, since Ibn Makula’s informants were for the most part from §afi‘i and Hanbali circles, not H an a fi15. Now Avicenna’s association with Abu Bakr Ahmad is attested to by Avicenna himself, who says the following in the Autobiography about him: In my neighborhood there also lived a man called Abu Bakr al-Baraqi, a Hwarazmian by birth. A prudent man, he favored jurisprudence, Qur ’anic exegesis, and asceticism, but he also had an inclination for these [i.e., the philosophical] sciences. He asked me to comment on the books [on philosophy] and so I composed al-Hisil wa-l­mahsul for him in about twenty volumes, and I also composed a book for him on ethics which I called al-Birr wa-l­itm

Apart from the question of Avicenna’s date of birth which is raised by this passage (and which is discussed in the Appendix), it is to be noted that

14 “Hakada dakara R bnu bnihi [i.e., of the Father, Aba ‘Abd Allah Muhammad] Abu Abd Allah b. Abi Bakr al-Baraqi. .. Wa-ra ’aytu diwana Si‘rihi, wa-aktaruhu bi-hatti tilmdihi Ibn Sina al-faylasuf. ” All this information is taken from Sam ‘ani ’s al-Ansab, who cites Ibn Makala: Hyderabad, 1383/1963, II, pp. 172-175, n° 454. In the Hyderabad (1381) edition o f Ibn Makala’s Ikmdl, this passage about Avicenna is absent from the entry on Baraqi (vol. I, p. 483). See the article “Ibn Makala” in E.l?, Ill, 860b (J.-C. Vadet). 16 Text in Gohlman, Ibn Sina, 38.6-40.1. My translation in Avicenna and the Aristotelian Tradi­ tion, Text 3, § 13.

II 329 Avicenna nowhere refers to Abu Bakr Ahmad as his teacher, but only as the patron of the lost al-Hasil wa-l-mahsul. This, however, may not mean much either way; as I have tried to show elsewhere, 17 Avicenna deliberately mentions in the Autobiography as few of his teachers as possible in order to present himself essentially as an autodidact. The very fact that he composed such as enormous work (twenty-one volumes) for Abu Bakr, by itself indicates that he was somehow beholden to him, and this obligation can best be explained as that of a student toward his teacher; furthermore, he identifies Aba Bakr as a scholar of the religious sciences, with fiq h enumerated first. On the other hand, Aba Bakr Ahmad need not have taught him only fiqh ; if Avicenna indeed undertook to copy Aba Bakr’s poetry in his own hand, the relationship between them may have been mostly literary rather than legal. In any case, what emerges clearly from the available evidence is that Aba Bakr Ahmad al-Baraqi in all probability did teach Avicenna something and that his field of expertise was the religious sciences, but that it is nowhere stated by Ibn Makala that he taught him fiqh. Such an assumption appears first to have been made by Ibn Abi 1-Wafa’ on the basis of Ibn Makala’s statement quoted above. After a very brief report on Avicenna’s life, Ibn Abi 1-Wafa’ continues, “He was a student of Aba Bakr Ahmad ..., with whom he studied fiqh. ...Ibn Makala said, ‘...I saw the diwan ...[as above]” 18. This identification of Aba Bakr Ahmad as Avicenna’s fiqh teacher was subsequently taken up and transmitted in the tradition 19. Tracing the transmission of this tradition has thus revealed that the specific detail of the subject(s) taught to Avicenna by Aba Bakr may not be entirely beyond doubt; nevertheless, the easy and matter-of-course way in which this has been accepted, recorded, and transmitted in Hanafi circles corroborates unambiguously the conclusion from Avicenna’s report about Isma‘U al-Zahid that he had a Hanafi legal education. The group of scholars with whom Avicenna engaged in legal disputations in his mid-teens were Hanafis, perhaps exclusively so. We have already encountered the most important among them: Aba Bakr Muham ­ mad b. al-Fadl al-Kamari, the foremost legal expert; the Baraqi family, Avi­ cenna’s neighbors; and Isma'il al-Zahid, his teacher. It is interesting to note that there were close personal and social ties among these men. Al-Kamari led the prayer at the funeral of the Son Baraqi, Aba Bakr Ahmad, who, as we

^

Avicenna and the Aristotelian Tradition, Chapter 3.4. “Wa-talmada li-l­Imam Abi Bakr Ahmad. .. wa-tafaqqaha ‘alayhi. ... Qala Ibn Makula. .. wa-ra ’dytu dxvuan.. .[as in note 14, above] ”, in al-Cawahir atmuch a, I, p. 195, s.n. al-Hasan[«c] b. ‘Abd Allah b.Sina. ' Ibn Qutlubuga, Tag al-taragim, Bagdad, 1962, p. 25, n° 66; Ibn Tagribirdi, al-Nugum al-zahira, Cairo, n.d., V, pp. 25-26, s.a. 428.

II 330 saw, was in all probability Avicenna’s teacher and, of course, his neighbor; al-Kamari was also the teacher of Isma‘il al-Zahid. They were all part of the intellectual aristocracy, the ‘ulam a’ of the Samanid dynasty, and as such offered their services in the implementation of its energetically Hanafi policies. Briefly, it may be recalled that the Samanid Mansur b. Nuh ( regn. 350/961 - 366/976) assembled a group of legal scholars and asked them to issue a fatw a on the lawfulness of translating Tabari’s Tafstr into Persian. Among them was, naturally, al-Kamari, and another scholar of some reputation, al-Halil b. Ahmad al-Sigzi, the hadzt teacher of the Son Baraqi, Abu Bakr Ahmad 20! The inner circle, it appears, was rather small. Furthermore, MansUr’s son Nuh II (regn. 366/976 - 387/ 997), the ruler treated by Avicenna himself, ordered the translation of Abu 1-Qasim Ishaq al-Samarqandi’s expose of Hanafi fiqh into Persian 21. The Samanids, it must be noted, had a stake in being Hanafls: given their Suubi policies, which may be taken to have been very high on their list of priorities, the Hanafiya was the madhab that could accommodate them best because it was the only one that allowed such translations of the Qur’an 22. It was with the scholars of this ruler’s court that Avicenna, the precocious son of a high administrator at the very court, debated “in the manner customary with these people ” 23. Avicenna left Buhara after the fall of the Samanids (389/999) and went to Gurgang, to the court of the HwarazmSah ‘Ali b. Ma’mun, apparently in search of political asylum. The choice of destination was hardly arbitrary; the Samanid ruling establishment, to which by then Avicenna belonged (after the death of his father he himself had taken an administrative post), had close ties with the Ma’munids in Hwarazm,24 and Avicenna had personal relations with Hwarazmians, notably the Baraqi family. Well recommended, Avicenna practiced law in Gurgang, for a salary “which provided enough for someone like” him 25. The Ma’munid HwarazmSahs were noted for their Sunni orthodoxy,26 and given Avicenna’s background

20 This last detail is provided by Sam 'ani, Al-Ansab, II, 173. The names o f the scholars who took part in this council are given in the introduction to the Persian translation of Tabari, listed by C.A. Storey, Persian Literature, London, 2nd ed., 1 9 7 0 ,1, i, p. 2, note 1. For the argument of the jurists and its connection with the “linguistic Su'ubiya” policies of the Samanids see L. Richter-Bernburg, “Linguistic Shu'ubiya and Early Neo -Persian Prose ”, Journal of the American Oriental Society, 94 (1974), pp. 62-63. 2* Richter-Bernburg, “Linguistic Shu'abiya”, 56 and note 13. For further references to Abu 1Qasim Ishaq al-Samarqandi see W. Barthold, Turkestan down to the Mongol Invasion, Lon ­ don, 3rd ed., 1968, p. 267 and note 5. 22 Richter-Bernburg, “Linguistic Shu'ublya”, 62-63. 23 Autobiography, Gohlman Ibn Sina, 20.6-7. 24 See the article “Al-e Ma’mun ” in the Encyclopcedia Iranica, I, 762-764 (C.E. Bosworth). 26 Autobiography, Gohlman Ibn Sina, 40.7-8. 26 See the article “Khwarazm” in E .lJ, IV, 1063a (C.E. Bosworth).

II 331 in Hanafi law, it is hardly likely that he practiced anything else. The same argument applies to his subsequent move, after a considerable odyssey, to 6urgan, to the court of Qabus b. VuSmagir (d. 403/1013), who was also known for his aggressive Sunni policies to the point of persecuting 5i‘is and Mu‘tazilis 2

... (= Miskawayh, Kitab as-Sa ada, Cairo 1335/1917, p. 37. 4-6). The scribe of the Taymur manuscript ignored the insert sign ( ^) after al-igtihad, and instead of writing al-igtihadat, he added the marginal correction -ati to the end of the top line, reading (p. 58), J&j

... [szc!] cjlUj

...

P) The note, added above the title on f. 142r of Abd-ar-Razzaq’s manuscript, by astudious reader/owner about the meaning of the word Utulugiya (see note24) is duly reproduced by the scribe of the Taymur manuscript, also as a note, in his margin (p. 111).

Now the contents of the entire 'Abd -ar-Razzaq manuscript were copied by Muhammad Ibrahim for the Dar al-Kutub; it is quite unlikely that the very first selection of the manuscript, and one of the most extensive, should have been omitted (see the list of contents in Section IV below, No. 1). M. Ibrahim ’s copies of all the other parts of the manuscript are duly accounted for by the volumes Hikma 209 and Hikma 213 -23; only the first selection, Avicenna ’s Tallqat, is missing. I therefore suspect some confusion with a presumed or another Istanbul volume, also entitled Ta liqat, but nothing definite can be said before an actual inspection of the volume(s) concerned.

III 14

Texts from Avicenna s Library y) On p. 126 of the Taymur manuscript, at the beginning of a new piece, the scribe wrote the customary formula after the basmala, wa-bihi tiqati, but then he crossed it out and wrote instead the unusual one of Abd-arRazzaq, < bihi> atiqu wa-'alayhi atawakkalu (see Section III.2 above). In addition to the works copied from Abd-ar-Razzaq’s manuscript, the Taymur manuscript contains also the following: i) Pp. 1-54. Himam-ad-Dln (?). Sarh Tahmis flib] al-Qasida alAymya al-musamma bi-Kitab al-Kasf wa-l-bayan f i Aim ma Tifat alinsan . A commentary on Tilimsanl’s commentary on the Aymya (Mahdavlp. 196, no. 7). Pp. 55-6. Blank. ii) P. 57. Ps.-Ibn Sina. Al-Qasida al- Aymya (Mahdavi No. 99, Anawatl No. 93). iii-ix) Pp. 58-176. See nos. 13, 14, 15, 4i-iii, and 5 in Section IV below.

e) O ther MSS. It is extremely likely that other known manuscripts are also dependent, in some cases perhaps through a number of intermediaries, on 'Abd -ar-Razzaq’s manuscript. The probability is particularly high with manuscripts which contain works that have a tenuous record of transmission, like the fragments from Avicenna’s Insaf. I have examined some of these manuscripts, but the cursory inspection which I have been able to make of them for the present purposes has not provided conclusive evidence of their dependence. This is a task that the future editor of these works will have to perform. Nevertheless, in order to remind the reader of this probability, I have mentioned these manuscripts in the appropriate place in the list of contents of 'Abd -ar-Razzaq’s manuscript in Section IV below. IV. Contents and Publication Record of the M anuscript Each entry in the following list of contents contains: a) the title of the selection as found in the manuscript, followed by references, in the case of works by Avicenna, to their serial numbers in the Inventory of his works in my book Avicenna and the Aristotelian Tradition (22014, note 1 above), where the references to Mahdavl’s and Anawati’s bibliographies can also be found. Since the works are sufficiently

III Texts from Avicenna s Library

15

identified there as well as by the incipits and explicits provided by BadawT Aristu (44)-(51), which should be consulted, no further details are given; P) a listing of manuscripts directly transcribed from Abd-ar-Razzaq’s manuscript, as described in Section III.5c-d; manuscripts Dar al-Kutub 2694W and 2972W are not listed separately since they apply to all entries; y) a listing of manuscripts probably dependent on Abd-ar-Razzaq’s manuscript (see Section III.5e above); 8) information about the record of publication of each selection, references directly relating to the text in question and not isted by the bibliographies in (a), and other comments. 1. Avicenna, Notes (GS 12a). a) Ff. lv-68r. At-Ta liqdt. The work starts immediately after the basmala, the pious formula (tiqatl bi-llahi wahdahu; different from the usual one of Abd-ar-Razzaq: see Section III.2), and the hamdala. The title is given at the end of the work, on f. 68r only: ahir al- mawgud min hadihi t-ta liqdt, wa-li-llah al-hamd wa-l- [12a] minna; tamma kitab at-Ta liqat, wakatabahu Abd-ar-Razzaq.... p) Cairo, Dar al-Kutub, Hikma 207 (?).28 y) Cairo, Dar al-Kutub, Taymur Hikma 67, pp. 1-242. Copied in Muharram 1057/February 1647. 8) Published by Abd-ar-Rahman BadawT from this manuscript, Ibn Sina, at-Taliqat, Cairo 1973. 2. Avicenna, Discussions (GS 14). a) Ff. 68v-116r. Kitab al-Mubahatat An as-Sayh ar-Ra Ts ... Ibn Sina. P) Cairo, Dar al-Kutub, Hikma 209. 8) Published by Abd-ar-Rahman BadawT from this manuscript in his Aristu 119-249. The order in which the material is found in the manuscript was not always followed by BadawT, who also omitted arbitrarily certain passages. The following table lists in order the contents of this work in the manuscript with the corresponding pages from BadawT’s book (see

See the preceding note.

III Texts from Avicenna s Library

16

also David C. Reisman, The Making o f the Avicennan Tradition: The Transmission, Contents, and Structure o f Ibn Slnas al-Mubahatat (The Discussions), Leiden: Brill, 2002, Appendix A, pp. 265-73). MS ff. 68v-69v = Badawl pp. 119-122.8. Letter to Kiya, otherwise known as Ihtilaf an-nas f i amr an-nafs wa-amr al- ‘aql (GP 13). See Gutas Avicenna T12 for a translation and further references. Ff. 69v-104r5 = pp. 122.9-222.20. Ff. 104r5-105r6, Badawl omits. The text of paragraphs 146-66, according to BadawT’s enumeration, is repeated here. Badawl does not collate the two versions (see p. 222, note 1). Ff. 105r7-107r penult. = pp. 222.21-229.11. Ff. 107r penult. - 107vll, Badawl omits. This is a doublet of par. 94; see Badawl p. 229, note 1. F. 107vll- 14, Badawl omits. This is a doublet of pars. 105-6; see Badawl p. 229, note 1. Ff. 107vl4 -109rl0 = pp. 229.12-233.17. Ff. 109rl0-110v21 = pp. 240-44. Letter (.Kitab) = Mubahata I (cf. Reisman Avicennan Tradition 205-7. Ff. 110v21—11 lvl 1. Letter {Kitab) = Mubahata II. See Gutas Avicenna T il for a partial translation and further references. Badawl published only the following ff.: ff. 110v21—l l l r l l = p. 245 ff. 11 lrl 1—111v4, Badawl omits; it is a doublet of 458: see Badawl p. 245, note 5. f. Illv 4 —11 = p. 246. Ff. Illvll - 112v4 = pp. 247-9. Nushat ahdin cahida li-nafsihi (GPP le).

F. 112v4 - ll, Badawl omits. Letter {Kitab) to Abu-Tahir b. Hassul (GPW 4c). See Badawl p. 249, note 1. Ff. 112vl 1—113vl 5, Badawl omits. Badawl p. 249, note 1, claims, without specifying, that the paragraphs omitted here are doublets. Ff. 113vl5 -115vl7 = pp. 233.18-239.16. [12b]

III Texts from Avicenna s Library

17

Ff. 115vl7 -116rl, BadawT omits. It is a doublet of par. 152; see BadawT, p. 239, note 1. F. 116r = 239.17-22. 3. Avicenna, The Easterners (GS 8). a) Ff. 116v-138r. Min Kitab al-Masriqlyln. See Gutas, Avicenna, W9. P) Cairo, Dar al-Kutub, Hikma 213. 5) Published from this manuscript in Mantiq al-Masriqlyln, Cairo 1910. The publisher does not mention the number of the manuscript but only that it is preserved in the Khedivial Library (p. 83). The colophon from f. 138r which he provides, however, is sufficient to identify his source as Abd ar-Razzaq’s manuscript. For reprints of the Cairo edition and further references see also Gutas Avicenna T8. 4. Avicenna, Fair Judgment (GS lla - b). i) Commentary on [Metaphysics\ Lambda. a) Ff. 138v-142rl4. Min Kitab al-Insaf. Sarh harf Kitab [sic] al-Ldm li-sSayh ar-Ra 'Is ... Ibn Sina. See Gutas, Avicenna, WlOc.i. P) Cairo, Dar al-Kutub, Hikma 216, pp. 1-22. Cairo, Dar al-Kutub, Taymur Hikma 86, pp. 97-111. 5) Published from this manuscript by BadawT Aristu 22-33. A critical edition on the basis of this and the independent Bursa manuscript (discovered by Y. Michot, Huseyin Qelebi 1194) and indirect sources, together with a French translation and notes, was published by M. Geoffroy, J. Janssens, and M. Sebti, Commentaire sur le Livre Lambda de la Metaphysique dAristote (Chapitres 6-10) par Ibn Sina (Avicenne), Paris 2014. ii) Commentary on the Theologia [Aristotelis]. a) Ff. 142rl5-146r6. FI sarh Utulugiya min Kitab al-Insaf An as-Sayh ar-Ra Is ... Ibn Sina. See Gutas Avicenna WlOc.ii. P) Cairo, Dar al-Kutub, Hikma 216, pp. 22-43. Cairo, Dar al-Kutub, Taymur Hikma 86, pp. 111-26.

III 18

Texts from Avicenna s Library 5) Only that part of this work contained on ff. 143vl9-146r6 was published by Badawl Aristu 59.13-66.4.29 Ff. 142rl5 -143vl9 were omitted by Badawl because they offer a different recension of some of the text included in the manuscript under the title TafsTr Utulugiya (no. 4iii below). The text omitted by Badawl overlaps with the following sections of the TafsTr published in Aristu: 37.14-38.10; 46.3-15; 47.1-6; 49.3-50.10; 50.10-18; 53.9-19; 56.14-57.8; 58.5-59.11. Badawl did not collate the two recensions.

iii) Exegesis o f the Book Theologia [Aristotelis]. a) Ff. 146r7-153v. TafsTr Kitab Utulugiya min al-Insaf An as-Sayh arRa I s ... Ibn STna.30 See Gutas Avicenna WlOc.ii. P) Cairo, Dar al-Kutub, Hikma 215. Cairo, Dar al-Kutub, Taymur Hikma 86, pp. 126-57. y) Cairo, Dar al-Kutub, Taymur Hikma 102, pp. 2-20; copied end of Muharram 1095/mid-January 1684 in Awrang. The title of this work is worded differently: Ta iTqat ra Is al-hukama5al-muta AllihTn Ala Utulugiya (p. 2).31 Oxford, Bodleian, Marsh 536, ff. 69v-84v. Ankara, ismail Saib 4605, copied 696H, referred to in M.T. Danespajuh, Fehrest-e MTkrufilm-ha-ye Ketabhane-ye MarkazT-ye Danesgah-e Tehran, Tehran 1348S, 1,449, no. 6 (no ff. references given). 8) Published from this manuscript by Badawl Aristu as follows: [13a] ff. 146r7-151vl5 = Badawl pp. 37.1 - 59.11 ff. 151vl5-20 = 59 note 7 ff. 151v20-153v6 = 66.5-73.14 The Exegesis was translated into French from BadawT’s text by G. Vajda, “Les notes d’Avicenne sur la ‘Theologie d’Aristote’”, Revue Thomiste 51 (1951) 346-406, and studied, on the basis of this translation, by L. Gardet, “En l’honneur du millenaire d’Avicenne. L’importance d’un texte nouvellement traduit: les gloses d’Avicenne sur la pseudo-Theologie 29 The folio number given by Badawi Aristu 65.20 as “ 164a” is a misprint for 146a. 30 The number of the first manuscript listed under this title by Sayyid Ibn STna 14 should be, 6 Hikma M. Most of the manuscripts containing this work are also listed by Kraus “Plotin ” 274, note 2. 31 Sayyid Ibn STna 13 cites the title inaccurately and does not list it under the TafsTr on p. 14. Kraus, ibid., is also guilty of the first oversight.

III Texts from Avicenna s Library

19

d’Aristote,” Revue Thomiste 51 (1951) 333-45, reprinted in his Etudes de philosophic et de mystique comparees, Paris 1972, 135-46.

4.1. “Plato”, Plato’s Arguments for the Immortality o f the Soul a) F. 153v6-18. Hugag Aflatun Ala baqa5an-nafs. 5) Published from this manuscript by BadawT Aristu 73.15-74. Translation and study by Ahmad Hasnawi, “Deux textes en arabe sur les preuves platoniciennes de l’immortalite de Tame,” Medioevo 23 (1997) 395-408, according to whom, the arguments in Arabic, which go back to the Republic X, 608c9-611a3, the Phaedo 105b5-107al, and Proclus’s Elements o f Theology, prop. 186-7, may derive from the translation of an otherwise lost essay by Proclus.

5. Avicenna, Marginal Glosses on Aristotle’s De Anima (GS 11c). a) Ff. 154r-168rl5. At-Ta liqat Ala hawasl Kitab an-Nafs li-AristdtalTs min kalam as-Sayh ar-Ra Is ... Ibn Sina. See Gutas Avicenna WlOc.iii. P) Cairo, Dar al-Kutub, Hikma 214.32 Cairo, Dar al-Kutub, Taymur Hikma 86, pp. 157-76 (incomplete; see Section III.5d above). 5) Published from this manuscript by BadawT Aristu 75-116.16. The fragments of Ishaq b. Hunayn’s translation of De Anima contained in this work were identified and edited separately, on the basis of BadawT’s text, by R. M. Frank, “Some Fragments of Ishaq’s translation of the De Anim af Cahiers de Byrsa 8 (1958-59) 231-51, reprinted in his Philosophy; Theology and Mysticism in Medieval Islam. Texts and Studies on the Development and History o f Kalam, /, ed. by D. Gutas, Aldershot: Ashgate, 2005, no. II. A study was published by D. Gutas, “Avicenna’s Marginal Glosses on De anima and the Greek Commentatorial Tradition,” in Philosophy Science & Exegesis in Greek, Arabic & Latin Commentaries, P. Adamson, H. Baltussen, and M.W.F. Stone, eds (Bulletin of the Institute of Classical Studies Supplement 83.2), London 2004, vol. II, pp. 77-88.

32

The number of the manuscript is wrongly given by Sayyid Ibn Sina 13 as “214M.”

III 20

Texts from Avicenna s Library

5a. Avicenna (?). A note on the numbers as principles. a) F. 168r 16-21. The note bears no title but simply begins, Inda katir min al-awa HI anna l-a ‘dad mabadi5li-l­ Alam. P) Cairo, Dar al-Kutub, Hikma 214, p. 81. 5) Published from this manuscript by Badawl Aristu 116.17-25. Since the note bears no ascription, it is reasonable to assume, on the basis of its position in the manuscript, that Abd-ar-Razzaq found it either at the end of Avicenna’s copy of De Anima from which he was transcribing the Marginal Glosses, or as a stray note in that copy. Frank “Ishaq’s Translation” 239, note 7, suggests that it refers to Aristotle’s De Anima 404b 19ff. 6. Avicenna, Explanation o f the Modal Propositions (GL 10). a) Ff. 168v-187r4. Bayan dawat al-giha An as-Sayh ar-Ra Is ... Ibn Sina. P) Cairo, Dar al-Kutub, Hikma 217. 5) The work remains unpublished.

7. Avicenna, [Answers to] Twenty Questions [on Logic] (GL 15). a)Ff. 187r5-193r. Lsruna mas Ala sa Ala anhd s-Sayh ar-Ra Is Abu All Ibn Sina ahl al- Asr.33 P) Cairo, Dar al-Kutub, Hikma 222. 5) Published from three Istanbul manuscripts by M. T. Danespajuh, “AlMasa’il al-gariba al- isrlnlya li- Abi All Ibn Sina,” in Collected Texts and Papers on Logic and Language, eds M. Mohaghegh and T. Izutsu [Wisdom of Persia Series VIII], Tehran 1353S/1974, [13b] pp. 15-16 (introduction), 79-106 (text). In his introduction (p. 16), Danespajuh merely mentions Abd-ar-Razzaq’s manuscript on the basis of Badawl’s introduction (Aristu [47]). See the study of this text, with full listing of the questions, by T. Street, “Avicenna’s Twenty Questions on Logic:

33 The reference to the old Khedivial catalogue of Cairo manuscripts, vol. VI, p. 104, given by Mahdavl p. 17 (work no. 8), and, on that basis, by Gutas, Avicenna (20142), 440, is to 'Abd - ar-Razzaq ’s manuscript.

III Texts from Avicenna ’s Library

21

Preliminary Notes for Further Work,” Documenti e studi sulla tradizione filosofica medievale 21 (2010) 97-111. 8. Aristotle, Metaphysics A. a) Ff. 193v-195v8. Fas If i harf al-Lam min Kitab Ma ba da t-Tabid liAristutalis al-faylasuf The translator is not named. p) Cairo, Dar al-Kutub, Hikma 220. 8) This slightly paraphrastic version of Metaphysics Lambda, chapters 6-10 (1071b3-1076a4) was first published by Abu-1-Ala Afifi, “Targama arabiya qadlma li-maqalat al-Lam min Kitab Ma ba da t - Tabla li-Aristu,” Bulletin o f the Faculty o f Arts o f the University of Egypt 5,1 (1937) 89-138, in parallel columns with his own Arabic version of the 1928 English translation by D. Ross. The text was published anew from the same manuscript by BadawT Aristu 3-11, and described as “an otherwise unknown shortened paraphrase of Metaphysics A, chapters 6 - 10” by R. Walzer, “New Light on the Arabic Translations of Aristotle,” Oriens 6 (1953) 92 [reprinted in his Greek into Arabic, Cambridge, Mass. 1962, p. 61]. For the relation of this translation to others, and further references, see A. Neuwirth, 'Abd al-Latif al-Bagdadi’s Bearbeitung von Buch Lambda der aristotelischen Metaphysik, Wiesbaden 1976, pp. 166-8, and A. Bertolacci, The Reception o f Aristotle’s Metaphysics in Avicenna’s Kitab al-Sifa5, Leiden 2006, pp. 16-17. 9. Avicenna, On the Definition o f Body (GP 1). a) Ff. 195v9-206r6. Kalam f i hadd al-gism 'an as-Sayh ar-Ra Is ... Ibn Sina. P) Cairo, Dar al-Kutub, Hikma 218. 8) The work remains unpublished. 10. Ibn-Zayla, Compilation on Metaphysics. a) F. 206r7-9. Min gumlat al-magmu 'fi l-ilahiyat 'an as-Sayh Abi Mansur b. Zayla. P) Cairo, Dar al-Kutub, Hikma 221.

III 22

Texts from Avicenna s Library 5) The manuscript contains only the title in the middle of the page, followed by one line of the hamdala. Then follows the next treatise. It would seem that either Abd-ar-Razzaq had no text to go along with the title and neglected to cross the latter out, or, more plausibly, IbnZayla’s Compilation consisted, in effect, of the next two selections in the manuscript.

11. Aristotle, Metaphysics a. a) Ff. 206rl0 - 206vl5. Min harf al-Alif as-Sugra min Kitab Ma ba da t-Tabl a li-AristutalTs al-faylasuf The translator is not named. P) Cairo, Dar al-Kutub, Hikma 221. 8) This is an abridged version of Ishaq’s translation of the Metaphysics a, chapters 1-2, 993a30-994b31, pp. 3-41 in the edition of Bouyges, Averroes, Tafsir ma b a d at-tabi'at, Beirut 1938. Despite the claims of Badawl Aristu (49), the text preserved here contains a number of readings better than those in the Leiden [14a] Averroes manuscript used by Bouyges, and it should be consulted in a future edition. For the Arabic translations of this book of the Metaphysics see Bertolacci Reception 14-15. 12. Themistios, Commentary on Aristotle’s Metaphysics A. a) Ff. 206vl6-210r7. Min Sarh Tamistiyus li-harf al-Ldm. The translator is not named. P) Cairo, Dar al-Kutub, Hikma 221. 8) Published from this manuscript in Badawl Aristu 12-21. See R. Walzer “New Light” 92-3 and A. Neuwirth A bd al-Latif 172, and especially R.M. Frank, “Some Textual Notes on the Oriental Versions of Themistius’ Paraphrase of Book A of the Metaphysics,” Cahiers de Byrsa 8 (1958— 59) 215-30, reprinted in his Texts and Studies on the Development and History ofKalam, /, no. III. There is a French translation by Remi Brague, Themistius. Paraphrase de la Metaphysique d ’Aristote (Livre Lambda); traduit de Lhebreu et de Larabe, introduction, notes et indices, Paris: Vrin, 1999.

III Texts from Avicenna s Library

23

13. Miskawayh, On Happiness. a) Ff. 210r8-217v. Kitab as-Sa Ada li-s-Sayh Abi All b. Miskawayh. P) Cairo, Dar al-Kutub, Hikma 223, ff. lv-19r. Cairo, Dar al-Kutub, Taymur Hikma 86, pp. 58-88. 8) Published from this manuscript by All at-Tobgl as-Suyutl, Kitab asSa Ada li-Ibn Miskawayh, Cairo 1335/1917, reprinted Cairo 1928. The publisher refers to his source manuscript as Hikma 6 only (p. 74), omitting the mim for the Mustafa Fadil collection. For the work, other manuscripts, and references see Gutas “Paul the Persian” (note 1 above), pp. 231-2. 14. Avicenna, Letter to Abu-SaTd Ibn-Abl-l-Hayr on the Cause of the Efficacy of Prayer and Visitation of Holy Sites (GM -Ps 5). a) Ff. 218r-219rl3. Kitab as-Sayh as-sa Id Abl Sa Id b. Abl l-Hayr ... ila s-Sayh ar-Ra 7s Abl All Ibn Sina (fi sabab igabat ad-du A 5wa-kayfiyat az-ziyara). P) Cairo, Dar al-Kutub, Hikma 223, ff. 19r-21v. Cairo, Dar al-Kutub, Taymur Hikma 86, pp. 89-92. 8) First published from other manuscripts by M.A.F. Mehren, Traites mystiques d ’Abou All al-Hosain b. Abdallah b. Sina ou d ’Avicenne, IIIe Fascicule, Leiden 1894, pp. 44-8; and more recently, among others, by Hasan cAsl, At-TafsTr al-Qur Am wa-l-luga as-sufiya fifalsafat Ibn Sina, Beirut 1403/1983, pp. 283-8. The fact that this letter is written in a hand that may be possibly different from that of Abd-ar-Razzaq (see Section III.2 above) raises the question whether it was, in fact, among Avicenna’s texts transmitted by Abd-ar-Razzaq, and ultimately of its authenticity; see Gutas Avicenna (20142), pp. 490-91. 15. Ps-Aristotle, Prayer. a) F.219rl4-18. Du A 5Aristatalls. P) Cairo, Dar al-Kutub, Hikma 223, f. 21v. Cairo, Dar al-Kutub, Taymur Hikma 86, p. 93. 8) Published from this manuscript by BadawT Aristu (50). 12-16. The same question raised in the preceding entry, paragraph (8), applies to this entry also. Ff. 219v-221r. Blank. [14b]

III Texts from Avicenna s Library

24

16. Avicenna, Hayy b. Yaqzan (GM 7), with the commentary by Ibn-Zayla. a) Ff. 221v-239r. Qissat Hayy b. Yaqzan al-MaqdisT. P) Cairo, Dar al-Kutub, Hikma 219. 5) None of the available editions of the text has drawn upon this manuscript. The first, and only critical edition by Mehren was apparently copied in Muhylddln Sabri al-Kurdl’s Garnical-bada 7 c, Cairo 1335/1917, pp. 91-113, and both together were used by Yaltkaya, who also added arbitrarily his own corrections. Corbin used principally Mehren’s text for his own edition and translation, and so did Goichon for her translation and study. Subsequent oriental printings of the work that I have seen offer no discernible improvements. References: M.A.F. Mehren, Traites mystiques (Ier Fascicule, Leiden 1889); M. $erefeddin Yaltkaya, “Hayy ibni Yakzan li-ibni Sina,” in Buyuk Tiirk filozof ve Tib Ustadi Ibni Sina. gasiyeti [sic] ve eserleri hakkmda tetkikler, Istanbul 1937, pp. 1-8 of that particular fascicle; H. Corbin, Avicenne et le recit visionnaire, Tehran 1954 (two volumes, text and translation), Paris 19792 (translation volume only); A.-M. Goichon, Le recit de Hayy ibn Yaqzan commente par des textes d Avicenne, Paris 1959. The different recension of the work {Hayy b. Yaqzan dla bayan ahar) preserved in the Istanbul manuscripts Aya Sofya 4829 and Nuruosmaniye 4894 (Mahdavl p. 96) has now been published by Asl, Falsafat Ibn STna (as in No. 14 above), pp. 321-35. The final three pieces in the manuscript are not in the hand of 'Abd-arRazzaq (see Section III.2) and accordingly do not have a common source with the other texts, i.e., Avicenna’s library. They would appear to be notes by an owner. 17. F. 239vl - 8. An anonymous fragment in Persian on Greek philosophers, described as the seven pillars: Thales, Anaxagoras, Anaximenes, Empedokles, Pythagoras, Sokrates, Plato; then the six great philosophers: Plutarch, Demokritos, Aristotle, Xenokrates, Porphyry, Alexander Du 1-Qamayn [sic for Alexander of Aphrodisias?]. This list appears to have been taken from Sahrastanl’s al-Milal wa-n-nihal.

III Texts from Avicenna s Library

25

18. Ff. 239v9 -240r. A fragment of an unidentified commentary on pseudo-Avicenna’s Ayniya (GP-Ps 4). Cf. the list of such commentaries in Mahdavl 99, pp. 196-7. 19. F. 240v. An anonymous fragment on the laylat al-mi Tag.

IV

AVICENNA’S MARGINAL GLOSSES ON AND THE GREEK COMMENTATORIAL TRADITION To the memory of H. J. Blumenthal The commentatorial activity proper of Avicenna (d. 1037 AD) on Aristotelian texts was directed, by his own admission, toward the composition of a voluminous work which he called al - Insaf, or Fair Judgment. In an extant letter addressed to Kiya, one of his disciples, he provides valuable information about the conception, composition, and fate of the book. He says: I had composed a book which I called Fair Judgment. I divided scholars into two groups, the Westerners and the Easterners, and I had the Easterners argue against the Westerners until I intervened to judge fairly when there was a real point of dispute between them. This book had contained approximately tw enty - eight thousand questions. I commented clearly on the difficult passages in the essential parts1 [of Aristotle’s books], up to the end of the Theology o f Aristotle, despite the fact that the Theology is somewhat suspect.2 I talked about the oversights of the commentators. I wrote it in a short period of time - [a work] which, had it been transcribed clearly, would have comprised twenty volumes. Then it was lost in the course of some rout, since there was only the first draft. Investigating it and these controversies was a pastime; after completing something I am working on [at present], I will occupy myself with rewriting it, although even thinking about rewriting is oppressive... At the present moment it is impossible for me [to rewrite it]: I do not have the free time for it, but am occupied with men like Alexander [of Aphrodisias], Themistius, John Philoponus, and their likes.3 The military setback to which Avicenna is here referring took place in 1030 AD, or seven years before his death.4 He was at that time active in the court of the local ruler of Isfahan, cAlaDal-Dawla. Not all of the Fair Judgment was lost, however. W hat survived of it, apparently on account of transcripts prepared by students or companions,5 are a commentary on Book 1Arabic fusus meaning here ‘essential parts’, as established by Bertolacci 2001, 261-265 and 270. 2 For a study of the Theology , an Arabic paraphrase of Plotinus’ Enneads IV-VI (whose authenticity was clearly suspected by Avicenna), see now Adamson 2002. 3 Text in BadawT 1947, 121-122; BTdarfar 1992, 375, #1161. Translated in Gutas 1988, 63-64. 4 Gutas 1988, 134-135. See now the extended discussion and revised chronology of the plunder and loss of Avicenna’s books offered by Reisman 2003.

IV 78

Lambda of the M etaphysics, and two recensions of a commentary on the Theology o f Aristotle - one described as a commentary and the other as an exegesis. These texts have survived in what appears to be a unique manuscript, Cairo Hikma 6M, dating from the first half of the 12th century, ie. a century after Avicenna ’s death. The manuscript is unique in the sense that the other copies of these texts that are extant, some demonstrably and some most probably, all derive from this manuscript.6 The commentary on De anima did not form part of the original draft of the Fair Judgment that was lost. Its title in the Cairo manuscript, where it follows the other three surviving sections of Fair Judgment, reads as follows: ‘Glosses in the Margins of Aristotle ’s De anima, from what Avicenna said ’ (al - tacllqat cala hawashl Kitab al-nafs liAristatalis min kalam al-shaykh al-ra Js Abi cAlT Ibn Sina). Unlike the titles in the three previous cases, this title does not mention the Fair Judgment at all. It is thus clear that Avicenna wrote these comments in the margins of his own copy of De anima and that this copy eventually came into the hands of the scribe of the manuscript who transcribed these notes clearly and consecutively, omitting the Aristotelian text. On the other hand, there is little doubt that in conception and execution these G losses follow the procedure of the Fair Judgment as described by Avicenna in his letter to Kiya: by ‘Easterners ’ Avicenna referred to his own philosophy and by ‘W esterners ’ to that of the Baghdad Aristotelians, in particular Abu ’1-Faraj Ibn al- Tayyib and especially Abu ’1-Qasim al- Kirmam, with whom he had acrimonious philosophical disputes.7 Accordingly, the Glosses ‘were written either immediately before the Fair Judgm ent... and directly occasioned [it] by whetting the appetite of Avicenna’s students for a similar and more extensive composition, or immediately afterwards, in compensation for its loss’, most likely the former.8 The entire text of the Glosses, 29 compact pages in the manuscript, 42 in the edition,9 consists of 133 passages of varying length which comment on the Aristotelian text. These passages are numbered in the manuscript, but the numbering is not continuous: 23 times the numbering breaks off and starts all over again. Some passages are not preceded by a number at all but by a sign, such as a crux. These numbers and signs cannot be arbitrary, merely serving to indicate the exact place in the Aristotelian text to which the marginal gloss is referring, for then there would be no purpose in breaking off the numbering and re - starting it. It rather seems that Avicenna began the numbering anew for each new segment or section of the Aristotelian text.10 Aristotle’s De anima was traditionally divided into three books totalling thirty chapters; but Avicenna is not following this division. The explanation of his procedure must rather be sought in the practice of the Neoplatonic commentators of Aristotle of dividing their commentaries into tmemata or sections, and these in turn into praxeis, or lessons.11 Of the extant Greek commentaries on De anim a, only that on the third book by Pseudo - Philoponus (ie., Stephanus of

5 It would appear that one of these companions, if not the only companion, responsible for these transcripts was Ibn Zayla; see now Reisman 2002, 203 and note 125. 6 See Gutas 1987. 7For Abu ’1-Faraj see Gutas 1988, 67-68; for al-Kirmanl see Reisman 2002, 166-185 and Michot 2000. 8 See Gutas 1988, 130-140 for a full discussion. 9 Cairo MS, Dar al-Kutub Hikma 6M, ff. 154r-168r; ed. in Badawl 1947, 75-116. I am preparing a revised edition with annotated English translation. 10 Avicenna follows the same procedure of numbering his individual comments in his commentary on the Theology o f Aristotle, where the numbering starts anew with each new chapter of the book, or mimar. See Vaida

1951,349, note 1. 11 See Festugiere 1963, 77-80 for the methods of Olympiodorus, which were adopted by later Neoplatonists.

IV AVICENNA ON DE ANIMA

79

Alexandria) is explicitly divided into tmemata and praxeis.12 As we shall presently see, there are some indications that the text of Stephanus may have been known in some form to Avicenna in Arabic translation, and although we do not have evidence to posit a direct connection, it is interesting to note that at least two of Stephanus ’ praxeis correspond with two segments in Avicenna’s division.13 Furthermore, in one instance, Avicenna refers to the part of De anima which he is studying as hadha ’l­fasl ‘this section’.14 This certainly cannot mean ‘this chapter ’ according to the traditional division of De anima because, judging by the contents of his comment, the discussion concerns a portion only of III.4.15 To Avicenna, therefore, such passages seemed, or were presented in his translated sources, as fu s u l, or sections, which would seem to stand for the smaller unit of praxis rather than the larger tmema, as one might expect.16 It accordingly seems relatively clear that Avicenna was, in fact, using a division of the text into smaller and, from a pedagogical point of view, more readily intelligible units, along the lines of the Neoplatonic commentators. We are not yet in a position to specify his immediate sources in this regard, however, despite the possible link with Stephanus. Avicenna used two translations of De anima, one by Ishaq b. Hunayn and an older one.17 It may be that either of these translations, especially if it was later corrected, as we have reason to believe, by Yahya b. cAdi,18 was so divided: we must not forget that in identifying and establishing a Greek text to be translated into Arabic, or in correcting it, translators used not only Greek manuscripts containing the text independently but also the lemmata of that text as it appeared in commentaries.19 Or it may be that the recension of one of the commentaries which Avicenna was using might have contained such a division; or, finally, Avicenna himself might have introduced the divisions for De anima, having encountered such a practice in other commentaries on Aristotle. Avicenna also used another m ethodological convention of the N eoplatonic commentators. The praxeis of a commentary were divided into two parts: the first was the thedria, in which the subject-matter of the praxis was given in a brief form, and the second was the lexis, in which the individual sentences or words of the text were commented upon in greater detail. Sometimes the thedria was quite simply reproduced,

12The authorship of this commentary on the third book was attributed to Stephanus already by the first editor of the text, Hayduck 1897, page v; the recent translator into English, W. Charlton, reviews the debates in the intervening century and offers corroborating evidence to support Hayduck’s conjecture: Charlton 2000a and 2000b, 1-10 in both volumes. Ps.-Philoponus (Stephanus) divides the third book into three tmemata and twenty praxeis, clearly marked by Charlton in his translation. Charlton translates tmema as ‘division’ and praxis as ‘lecture’. In what follows I number the praxeis consecutively while Charlton starts the numbering anew with each tmema, division. 13 Hayduck 446ff. {praxis 1) = Badawi 1947, 96.17ff.; Hayduck 506ff. {praxis 9) = Badawi 1947, 97.18ff. See also the second next note. 14Badawi 1947, 103.14 and 103.22. 15 Avicenna comments in this passage (Badawi 1947, 103.13-104.16) onDeanima III.4, 429bl0-430a9. Although the three paragraphs that form Avicenna’s comments are numberedas 4, 5, and 9 in the manuscript (three paragraphs would appear to be missing) and hence are part of a larger section, the text that is referred to as hadha ’l­fasl in these paragraphs {De anima 429bl0 -430a9) corresponds exactly to praxis 11 of Stephanus’ commentary (Hayduck 525.3-534.15). 16 Cf. Avicenna’s use of the word/as/ in his own Kitab al-nafs of the Shifcd (edited in Rahman 1959): the book is divided into five maqdlat {partes in the medieval Latin translation), each of which is subdivided into a number of fusul {capitula). See also the edition of the Latin text in van Riet 1968-72. 17Frank 1958-59, 231-234; Gatje 1971. 18 Gatje 1971,41-42. 19Flligel 1871-72,1.249.17-21. See also the other references in Peters 1968, 62 note 30.

IV 80

rather redundantly, in the lexis part.20 In the passage I just mentioned, where Avicenna refers to the part he is studying as ‘this section’, he follows this pattern quite closely. After a brief thedria which, in fact, starts with the words, wa bi ’l-jumla ( ‘in short’), he proceeds to a detailed discussion of the text under consideration and ends by repeating part of the thedria with which he had started.21 This does not occur very often in a pure and easily recognizable form, but traces of the thedria -lexis structure are present in the Glosses. The thedria form is especially apparent in five lengthy passages22 - excursuses one might call them - in which Avicenna goes beyond the requirements of the text immediately to be commented upon into a detailed exposition of the topic he is investigating or an extensive account of various interpretations of the text. In form at least, these lengthy passages are highly reminiscent of the theoriai introducing the praxeis of Stephanus’ commentary.23 Coming next to the literary form of the Glosses, we are confronted with an extremely variegated picture. For the nature of the glosses in this work ranges from the purely 24 25 26 27 28 lexical to the textual and grammatical, from the logical and geometrical to the 29 30 illustrative and, finally, from the paraphrastic to the interpretive and dogmatic. 31 We thus find Avicenna using now one literary form, now another, as the occasion may require. When he paraphrases, he follows the method of Themistius and sometimes even his text, instead of that of Aristotle.32 When his comments are more extensive, by contrast,

20 Festugiere 1963, 81. 21 Badawl 1947, 103.22-104.16 on De anima 429b22-430a9. 22 Badawl 1947, 77.16-79.14; 79.20-83.11; 84.12-89.20; 91.17-94.11; 106.15-108.3. 23 Badawl 1947, 91.17-94.11 is especially to be compared, in form only, with Hayduck 534.16-539.12. 24 Eg., Badawi 1947, 114.17, on De anima 433b24, explaining the words sakinan (ppepeT) and mutaharrikan (KiveiTdi) in the old translation he was using (Badawi 1954, 83 last line). 25 Eg., Badawl 1947, 112.12-16, on De anima 433al4 (vouc 8e o evem tou Xoyi£6p.evoc m i o TTpaKTimc). The translation which Avicenna was using misinterpreted this statement as referring to two intellects, a goal oriented intellect and a practical one (wa ’l ­ caqlu caqlanl caqlun mufakkirun li - cillatin wa min ajli shay3in wa caqlun camaliyyun (read thus the last word; in Badawl 1954, 82.6, it reads ‘an ta ’ammulin, based on a corruption in the manuscript). Avicenna immediately noticed the discrepancy - the practical intellect is the one which reasons for the sake of something - and assumed that there was a textual corruption (sahw) in his manuscript copy. He explained the corruption as being due to a transposition (taqdlm ) of caqlun mufakkirun in the place of caqlun camaliyyun and thus proposed to associate the phrase li - cillatin (sic; la callahu at Badawi 1947, 112.13 is wrong) wa min ajli shay3in with caqlun camaliyyun. In this he was most likely guided by Themistius; see Lyons 1973, 218.16: a cni ’I ­ caqla T -amaliyya alladhi bi -sababi shay3in yufakkiru (tov TTpuKTiKov Xeyto vovv Kdi evem tou XoyiCopievov, 118.31-32 Heinze). 26 Eg., Badawi 1947, 98.17-22, on De anima 429al0-12. See the discussion in Gutas 1986, 122-123. 27 Eg., Badawl 1947, 85.14-20, on De anima 408M8-20, where Avicenna constructs a syllogism to explicate the compressed Aristotelian text. For another example, analyzed in detail, see the Appendix. 28 Eg., Badawi 1947, 83.5-10 (ka-ma ya fa lu h u T -muhandisun), on De anima 407a6ff. The same geometrical demonstration of the nature of imagination occurs in the De anima part of the Shifa3 (Rahman 1959, 189 = Van Riet 1968, 46), and in the Najat (Cairo 1331, 281; English translation in Rahman 1952, 42). For an analysis of this argument see Kemal 1991, Appendix 5, 269-273. 29 Eg., Badawl 1947, 112.10, on De anima 433al citing as an illustration for eTepov organs (following a tradition represented by Simplicius, In de an. Hayduck 295.25-26: and the salivary glands.

tl to

|iopiov the sexual yevvr|TiKd piopia)

30 Eg., Badawi 1947, 97.18-98.2, paraphrasing De anima 428bl0-17. 31 Eg., Badawi 1947, 75.16-21, on De anima 402a23-25. See Avicenna’s independent treatment of this subject, in much the same terms, in the De anima part of the Shifa3 10.16ff. Rahman = 26.24ff Van Riet. 32 See eg. Badawi 1947, 75.4-7 on De anima 402al-4, where Avicenna, as already noted by Frank 1958-59, 240 note 1, is echoing Themistius’ paraphrase (Lyons 1973, 1.10-12). Also cf. the other references given by Frank 236, note 5 of the preceding page. In one instance, the verbal correspondence between Avicenna (Badawi 1947, 89.13-18, on De anima 408b25-29) and Themistius (Lyons 1973, 22.2-5) is striking.

IV AVICENNA ON

ANIMA

81

he often follows very closely the expository method of Philoponus33 and on occasion his contents as w ell.34 In this regard, there is very little, if anything, that formally distinguishes such passages from the great commentaries of Averroes.35 The rationale behind this uneven procedure of Avicenna’s can again be understood through reference to his letter to Kiya. Avicenna is not interested in writing one kind of commentary on De anima, an epitome, paraphrase, long commentary, etc., but rather in commenting ‘clearly on the difficult passages in the essential parts [of A ristotle’s books] ’. Although the Glosses are not part of the Fair Judgment, as we have seen above, the purpose is nevertheless the same as that expressed in the letter with regard to the Fair Judgment. In Avicenna in this case, literary form is subservient to purpose, not coextensive with it as in the Greek commentators and in Averroes. The extent to which and the way in which Avicenna uses the ideas of the Greek commentators also merit our attention. In the letter to Kiya Avicenna mentions that at that moment he was occupied with Alexander, Themistius, Philoponus, and others like them. In the Glosses he refers to the first two by name. The connection between Avicenna and the Greek commentatorial tradition is thus established by Avicenna himself. What is the nature of this connection in all its aspects, independently of the ambivalent attitude of Avicenna toward the commentators, whom he now charges with carelessness, as in the letter to Kiya, and now praises, as in certain passages in the Glossesl To begin with Themistius, Avicenna refers to him by name only twice - in both cases to criticize him - but his use of Themistius ’ paraphrase is much more extensive than these two references would indicate. In addition to following Themistius ’ method and often his wording, Avicenna also seems to be drawn to commenting on certain passages of D e anima simply because Themistius elaborated on them, not because the Aristotelian text may be particularly difficult or, in Avicenna’s eyes, erroneous. In such passages Avicenna develops arguments against Themistius and carries on a dialogue with him, not Aristotle. For example: In Book 1.4 of De anima (408a34ff.) Aristotle discusses the question whether the soul is in motion. In the course of his arguments he mentions that even if certain affections like grief, joy, and anger may seem to be movements, this would still not imply that the soul is moved, since it can be said that the movement in such cases is due to the soul, not that the soul itself is moved. In this argument Aristotle mentions anger only as an example and very briefly; it is not his intention to discuss anger as such. Themistius, however, while paraphrasing this Aristotelian passage, goes into a rather lengthy discussion of anger itself. Avicenna does not find this discussion satisfactory. He says (using the Easterners as his mouthpiece), ‘this statement is not conclusive’, and then embarks on an analysis of anger, concluding with the words (again attributed to the Easterners), ‘this subject, therefore, should not be investigated in such a fashion, and its investigation should not be restricted to the likes of such a discussion ’. These words seem to imply that it is not the Aristotelian text, but rather that of Themistius, that is commented upon here. Aristotle neither conducts an investigation upon nor even discusses the subject of anger; it is Themistius who does this. Furthermore, neither one of the other two extant Greek commentaries on De anima, those of Simplicius and Philoponus, discusses anger in this context. Barring then the possibility that Avicenna is arguing against some passage in Alexander’s lost commentary, he would clearly seem to be criticizing Themistius here.

33 Cf., eg., the similar structure and contents in Avicenna, Badawi 1947, 85.2-8, on De anima 408bl5-18 and Philoponus, Hayduck 157.23-158.34. 34 See Gutas 1986, 128-129. 35 Cf. Averroes’ commentary on the passage referred to in the second preceding note (408M5-18), in Crawford 1953, 85.27-86.58.

IV 82

As for Alexander of Aphrodisias, Finnegan amply demonstrated already half a century ago the influence Alexander’s De intellectu had on certain theories which Avicenna attempts to refute in the Glosses?6 Among the four times Avicenna refers to Alexander by name, one is of particular importance. In this passage, Avicenna comments on III.4 of De anima (429a24 -25), where Aristotle says about the intellect that ‘it is reasonable that it should not be mixed with the body ’ (Hamlyn). After elaborating on this subject, Avicenna concludes with the following remark: I am unable to understand how Alexander allowed it to be attributed to this man [scil. Aristotle] that he said that the material intellect, which is this predisposed faculty (al quwwa al - isticdadiyya), is hylic and material, and that the soul which possesses this faculty is also hylic and material.37 Now from what we know of Alexander, he nowhere states that the material intellect is matter itself; on the contrary, in De intellectu he explicitly states that the material intellect is not matter but that it is called material because it is sheer potentiality: One type [of intellect] is material intellect, but by ‘material’ I do not mean that it is a substrate (h u p o k e im e n o n ) like matter (for I call matter a substrate capable of becoming a particular thing (tode ti) through the presence of a form), but since what it is for matter to be mattter lies in its capacity [to become] all things, then that in which this capacity and potentiality itself lies is, in so far as it is potential, material.38 This is quite conclusive, and although we do not possess Alexander’s commentary on De anima, it is inconceivable that he should have changed his opinion on so basic a subject. On the other hand, in the third book of his commentary on De anima, extant in a Latin translation, Philoponus informs us that Alexander reported that Xenarchus, the first century B.C. Peripatetic, had misunderstood Aristotle and thought that the intellect was prime matter: ‘Xenarchus (according to Alexander) was misled into making the ill-judged conjecture that Aristotle is saying the intellect is prime matter’.39 In all likelihood it was this text of Alexander’s which misled Stephanus into making the following claim about Alexander: Alexander says th a t... intellect is potential, and if it is potential by virtue of suitability 0epitedeiotes),40 clearly it does not yet have the form, but is only matter. For everything that is potential in the first way is in matter. Intellect, then, is matter and does not have form, lest being muddied by form it should incline to the cognition of its own form and should not know forms contrary to it.41

36 Finnegan, 191-197. For Alexander’s De intellectu see Sharpies 1987, 1189 and 1202-1214. 37 Badawi 1947, 101.17-19. 38 Translated in Schroeder and Todd 1990, 46, on the basis of the Greek text edited in Bruns 1887, with this passage at 106.19-23. 39 Translated in Charlton 1991, 40, from the text edited in Verbeke 1966, with this passage at 15.68-69. Since Alexander nowhere mentions Xenarchus in the extant works, Paul Moraux suggests that this passage must derive from Alexander’s lost commentary on De cinimci: Moraux 1942, 211, note to paragraph e. 40 Or rather, ‘predisposition, readiness’, this word clearly being reflected in the term istVdad used by Avicenna in his description of this faculty {quwwa isticdadiyya) in the quotation cited above from the Glosses. 41 Translated by Charlton 2000a, 97, from the text in Hayduck 519.23-28.

IV AVICENNA ON DE ANIMA

83

This, as far as is known, is the only passage which attributes to Alexander the opinion that the intellect is matter.42 Since Avicenna maintains the same thing about Alexander - and indeed, he seems to be echoing the above passage - we are faced with the question of whether he in fact knew the text of Stephanus. This, however, has to be discussed in connection with another, more important problem: did Avicenna have the commentary of Philoponus available in an Arabic translation, despite the fact that such a translation is not mentioned at all by the Arab bibliographers and no manuscript of it has survived? There is weighty evidence, quite apart from Avicenna’s own statements, that he did, and that he actually used it. His interpretation of the Aristotelian aporia in III.4 (429all - 12), whether the rational part of the soul ‘is separable or not with respect to magnitude (kata m e g eth o sY , is totally dependent on the interpretation of Philoponus, as I discussed elsewhere 43 Another passage where Avicenna’s dependence on Philoponus cannot be seriously doubted is his comment on De anima 1.4 (408b 15-18), where Aristotle mentions that recollection is a movement which starts from the soul and ends in the sense -organs. Philoponus introduces in the process of recollection the concept of the spirit (pneuma) as the medium which contains the images (en toi pneumati enginetai ta phantasmata) which are nothing but remnants (enkataleimmata) of previous sensory experience - and transmits the mood associated with these images to the body.44 Avicenna interprets the ambiguous Aristotelian passage in the same way and describes the spirit in much the same words as Philoponus: ‘the spirit which contains the image ’ (al-ruh allatlfi - ha ’l-khayal).45 The concept of pneuma is important in the psychology of both Philoponus and Avicenna, and it seems, on the basis of such indications, that Philoponus should be considered seriously as a source for major aspects of the psychology of Avicenna. To the indications presented here for the existence of an Arabic translation of Philoponus ’ commentary we also have to add the evidence adduced by Jolivet with regard to al - Kindi’s use of the same commentary in his treatise on the intellect, and by Arnzen with regard to the anonymous Arabic paraphrase of De anima.46 The problem, then, with regard to this commentary is not any more whether it was available in Arabic translation but rather in what form or recension it was available. For, as we saw above, we find in the Arabic tradition traces of both the recension extant in the Latin translation of the third book and also the one circulating under the name of Stephanus. W hether these two recensions were available separately or together, the names under which they circulated in Arabic, and the reasons for the silence of the Arab bibliographers regarding them are questions that still have to be investigated. In his Glosses Avicenna continues the work of the Greek commentators. He uses their literary forms, their division of the text into praxeis, their techniques, and often their ideas. He discusses with them and against them the passages of the Aristotelian text which they considered important, but he also argues against them independently of the Aristotelian text under discussion. In all this he is firmly rooted in the commentatorial tradition. But there is one major difference. To all this tradition, within whose thoughtworld he works and with which he has to contend, he opposes another system, his own, under the guise of the Eastern Philosophy. The Easterners are mentioned forty -two times 42 Moraux, who also finds this passage not very trustworthy, instead of blaming Stephanus for the misinformation tends to consider this alleged tenet of Alexander as representing an earlier and transitory stage in his thinking: Moraux 2001, 348. 43 This point is discussed in detail in Gutas 1986. 44 Philoponus in De an., 158 Hayduck. 45 Badawi 1947, 85.6 46 Jolivet 1971; Arnzen 1998.

IV 84

in the Glosses, and in each case they are represented as the possessors of the true doctrine. Whenever this happens, one will find the very same issue that is raised in the Glosses discussed extensively and elaborated upon, in the manner suggested in the Glosses, in the corresponding part of Avicenna’s magnum opus, the Book o f the Cure (al - Shifa\ the Sufficientia of the Latins).47 By taking over the Aristotelian commentatorial tradition and using it in manifold ways, Avicenna clearly established his philosophical pedigree. By opposing to it another system, his own, and by writing about it in philosophical summae that dispense with the commentary format, he asserted his independence and founded a tradition that dominated philosophical thinking for centuries, East and West.

47 For examples see the Appendix and passage 12 in Gutas 2001, 11-12.

IV AVICENNA ON

ANIMA

85

Appendix: Avicenna ’s Method in Logical Analysis The following passage from Avicenna’s M arginal Glosses on De anima is a good example of Avicenna’s logical analysis of Aristotelian texts, a procedure which he followed ever since he was a young boy studying philosophy on his own, and which he described as follows in his Autobiography: The next year and a half [at age sixteen] I devoted myself entirely to reading philosophy: I read logic and all the parts of philosophy once again. During this time I did not sleep completely through a single night, or occupy myself with anything else by day. I compiled a set of files for myself, and for each argument that I examined, I recorded the syllogistic premisses it contained, the way in which they were composed [ie., which syllogistic figures were used], and the conclusions which they might yield, and I would also take into account the conditions of its premisses [ie., their modalities] until I had ascertained that particular problem. 48 The precise way in which Avicenna examined an argument, recorded the syllogistic premisses, and evaluated the conclusion can be seen in his comment on De anima 4 0 3 a ll 12, the second of two conditional sentences in a sequence. The first sentence in Aristotle says (4 0 3 al0 - ll), ‘If there is any of the functions (epycop) or affections (TTa0r||Jid.Twv) of the soul which is proper to it, it will be possible for it to be separable (xwpiCeaOai)’. The second one is the one which Avicenna analyzes here. It reads, ‘But if there is nothing proper to it, it will not be separable ’ (el 8e px|0ev eo riv t8iov auTfjg, ouk du elr\ X(jopiaTf|). On this second conditional Avicenna says the following: The Easterners said: [Aristotle’s] statement, ‘if there is nothing proper to it, it will not be separable ’ is based on [the assumption] that something does not exist unless it acts and is acted upon, otherwise it would be idle (m u cattal). This is an argument of the kind that is generally accepted (m a s h h u r), which is inappropriate for the [demonstrative philosophical] sciences. For something may possess in itself a perfection that extends neither to others nor to itself from others, and hence it would not be a priori (fi badihat al - caql) necessary that it should not exist. As for their49 statement, ‘because it would be idle’, it is a statement which, when used, resorts to a petitio principii,50 because it is as if51 one were saying, ‘otherwise, it would be something neither acting at all nor being acted upon at all ’, which is a tautology (wa-huwa nafs al - dacwa). [This minor premiss] needs to be followed by the major premiss, namely, ‘and whatever is like that, it does not exist;’ at which point the objection will be raised, ‘from where does he know that? ’ Avicenna here reconstructs as a categorical syllogism the hypothetical statement of Aristotle and the commentators, analyzes it, and refutes it. His reconstruction is as follows. In Aristotle’s statement, ‘But if there is nothing proper to it, it will not be 48 Translation in Gutas 1988, 27 with analysis of the passage at 177-181. 49 Ie., the Aristotelian commentators, specifically Themistius against whom this paragraph seems to be directed. Cf. Heinze 1899, 6.32-33. Also cf. Philoponus in De an. Hayduck 46.«/r.-47.1 50 For the Arabic term used by Avicenna, al-musadara cald ’l-matlub al-awwal, see Goichon 1938, §357.2 and Goichon 1951, 240 and note 2. In the latter passage, from Avicenna’s Ishdrdt, he explains that a petitio principii occurs ‘when two terms of a syllogism are two names for the same idea, although they ought to have a different meaning’. When this happens, the conclusion necessarily becomes one of the premisses. For the syllogism in question here, see what follows. 51 Perhaps the reading ought to bq, fa -innahu ka-annahu yaqulu instead offa -innahu kana yaqulu.

IV 86

separable’, which is the object of Avicenna’s analysis, the word ‘nothing ’ refers to the words ‘function ’ and ‘affection ’ in the immediately preceding sentence; the statement can be rephrased, then, as, ‘If there is no function or affection proper to the soul, it will not be separable ’. This is a hypothetical syllogism whose terms need to be checked for validity. Recast in assertoric terms, this statement becomes, ‘Whatever has no proper function or affection is not separable’, or, ‘cannot exist separately ’. In logical terms, the subject of this sentence is the minor term (A), and the predicate the major term (C). If this statement is to be logically demonstrated, it needs to be seen as the conclusion of a categorical syllogism which will require the addition of a middle term giving the reason why the statement is correct. In his paraphrase, Themistius supplies the middle term (B)52: ‘Because (or otherwise) the soul would be idle (apyov = m uca tta l)\ and ‘nothing’, Themistius adds, ‘comes into existence through nature without a function’, which could be seen as an extension of the well- known Aristotelian axiom that ‘nature does nothing in vain’.53 Avicenna rephrases as statements the major and minor terms, as follows: The minor term (A), ‘If there is no function or affection proper to the soul ’, becomes, ‘Whatever neither acts nor is acted upon at all’, and the major term (C), ‘it is not separable ’ becomes, ‘does not exist’, while the middle term (B) is, of course, ‘idle’. Then Avicenna reconstructs the syllogism implied by Aristotle and Themistius: A is B: Whatever neither acts nor is acted upon at all is idle; B is C: Whatever is idle does not exist. Conclusion (A is C): Whatever neither acts nor is acted upon at all does not exist. This is the inference made by Aristotle and Themistius, namely, that if the soul has no action or affection that is proper to it, it does not exist separately. Avicenna disagreed with this statement (or this conclusion, if one takes into consideration that Themistius, by providing the middle term, attempted to make this into a valid syllogism). Avicenna’s objection to the sense of the statement lies in the fact that it would make the soul, which, according to Avicenna is an immaterial substance that survives the death of the body, a mortal entity, since after the death of the body the surviving soul would neither affect nor be affected by anything. As he says, something may be in a state of perfection (ie., completely actualized) without having to be in communication with anything else. He therefore has to maintain the separate existence of the soul independently of its actions and affections. To do this he must first refute the syllogism which claims that, ie., Aristotle’s statement and Themistius ’ defense of it. In order to disprove the statement, Avicenna brings forward two arguments, one procedural the other substantive. His procedural objection is to the form of the syllogism, as implied by Themistius and reconstructed by Avicenna. He says that the syllogism begs the question because the middle term (B) of the syllogism, ie., ‘idle’, is nothing else but a synonym of the minor term (A), ‘whatever neither acts nor is acted upon at all’. The minor premiss (A is B) thus becomes, ‘Something that neither acts nor is acted upon is something neither acting at all nor being acted upon at all’, which is indeed a tautology. This causes the major premiss (B is C) to be, ‘whatever neither acts nor is acted upon at all does not exist’, which is identical with the conclusion (A is C), and therefore a petitio principii. Avicenna’s substantive objection is to the validity of the major premiss itself (B is C), ‘Whatever is idle (ie., neither acts nor is acted upon) does not exist’. Themistius calls this

52 Heinze 6.10, 6.33. 5 3 M r |8 e v

fic n riv

tto k e T

f) 4 > v ( J ig , 4 3 2 b 2 1 , 4 3 4 a 3 1 .

IV AVICENNA ON

ANIMA

87

statement a ‘self-evident axiom’,54 but Avicenna disputes this and says ‘where does Themistius know that from ? ’ meaning that this assertion itself needs to be proven. Avicenna further explains that this axiom is ‘inappropriate for the [demonstrative philosophical] sciences’, being dialectical (mashhur = evSo^ov) in nature, and hence far from having the axiomatic self-evidence that Themistius contends.

Bibliography Adamson 2002 = Adamson, P. The Arabic Plotinus: a Philosophical Study o f the ‘Theology o f Aristotle ’ (London 2002). Amzen 1998 = Arnzen, R. Aristoteles ’ De Anima, eine verlorene spdtantike Paraphrase in arabischer und persischer Uberlieferung (Leiden 1998). Badawi 1947 = Badawi, CA. Aristu cinda ’l ­ cArab (Cairo 1947). Badawi 1954 = Badawi, CA. Aristutalis f i ’l-nafs (Cairo 1954). Bertolacci 2001 = Bertolacci, A. ‘From al-Kindl to al-Farabl: Avicenna’s progressive knowledge of Aristotle’s Metaphysics according to his autobiography’, Arabic Sciences and Philosophy 11 (2001) 257-295. Bldarfar 1992 = Bldarfar, M. (ed.), al-Mubahathat (Qum 1371/1413/1992). Bruns 1887 = Bruns, I. Alexandri Aphrodensiensis... De anima Liber cum Mantissa (Berlin 1887). Charlton 1991 = Charlton, W. Philoponus on Aristotle on the Intellect (London - Ithaca NY 1991). Charlton 2000a = Charlton, W. ‘Philoponus ’ On Aristotle ’s On the Soul 3.1-8 (London ­ Ithaca, NY, 2000). Charlton 2000b = Charlton, W. ‘Philoponus ’ On Aristotle ’s On the Soul 3.9-13 with Stephanus On Aristotle ’s On Interpretation (London ­ Ithaca, NY, 2000). Crawford 1953 = Crawford, F. S. Averrois Cordubensis commentarium magnum in Aristotelis De anima libros (Cambridge MA 1953). Festugiere 1963 = Festugiere, A.-J. ‘Modes de composition des Commentaires de Proclus’, Museum Helveticum 20(1963) 77-100. Finnegan 1956 = Finnegan, J. ‘Avicenna’s refutation of Porphyrius’, Avicenna Commemoration Volume (Calcutta 1956), 187-203. Fliigel 1871-72 = Fliigel, G. (ed.), Ibn al-Nadlm, Kitab al-Fihrist, 2 vols (Leipzig 1871-72). Frank 1958-59 = R. M. ‘Some fragments of Ishaq’s translation of the De anima’, Cahiers de Byrsa 8 (1958-59) 231-251. Gatje 1971 = Gatje, H. ‘Die vorhandenen Ubersetzungen der Schrift liber die Seele, Avicennas Kitab al- insdf und die Autoren der Ubersetzungen’, in his Studien zur Uberlieferung der aristotelischen Psychologie im Islam (Heidelberg 1971), 38-44. Goichon 1938 = Goichon, A.-M. Lexique de la langue philosophique d ’lbn Sina (Paris 1938). Goichon 1951 = Goichon, A.-M. Ibn Sina, Livre des directives et remarques (Paris 1951). Gutas 1986 = Gutas, D. ‘Philoponos and Avicenna on the separability of the intellect’, Greek Orthodox Theological Review 31 (1986), pages 121-129. Rpt. in D. Gutas, Greek Philosophers in the Arabic Tradition (Aldershot 2000), no. XI. Gutas 1987 = Gutas, D. ‘notes and texts from Cairo manuscripts, II: texts from Avicenna’s library in a copy by cAbd-ar-Razzaq as-Signahf, Manuscripts o f the Middle East 2 (1987) 8-17. Gutas 1988 = Gutas, D. Avicenna and the Aristotelian Tradition (Leiden 1988). Gutas 2001 = Gutas, D. ‘Intuition and thinking’, in Aspects o f Avicenna, ed. R. Wisnovsky (Princeton 2001), 1­ 38. Hayduck 1897 = Hayduck, M. Ioannis Philoponi in Aristotelis De anima libros commentaria (Berlin 1897). Heinze 1899 = Heinze, R. Themistii in libros Aristotelis De anima paraphrasis (Berlin 1899). Jolivet 1971 = Jolivet, J. L ’intellect selon Kindi (Leiden 1971).

54 ’Euapycg d^LW|ia, Heinze 6.32.

IV 88

Kemal 1991 = Kemal, S. The Poetics o f Alfarabi and Avicenna (Leiden 1991). Lyons 1973 = Lyons, M. C. An Arabic Translation o f Themistius Commentary on Aristoteles De anima (London 1973). Michot 2000 = Michot, Y. Ibn Sina. Lettre au Vizir Abu Sacd (Beyrouth 2000). Moraux 1942 = Moraux, P. Alexandre dAphrodise. Exegete de la noetique d ’Aristote (Liege ­ Paris 1942). Moraux 2001 = Moraux, P. Die aristotelismus bei den Griechen, vol. Ill (Berlin ­ New York, 2001). Peters 1968 = Peters, F. E. Aristoteles Arabus (Leiden 1968). Rahman 1952 = Rahman, F. Avicenna ’s Psychology (London 1952). Rahman 1959 = Rahman, F. Avicenna ’s De anima (London 1959). Reisman 2002 = Reisman, D. C. The Making o f the Avicennan Tradition (Leiden 2002). Reisman 2003 = Reisman, D. C. ‘Stealing Avicenna’s books: a study of the historical sources for the life and times of Avicenna’, in Before and After Avicenna, ed. D. C. Reisman (Leiden 2003) 91-126. van Riet 1968-72 = van Riet, S. Liber de Anima, 2 vols (Louvain/Leiden 1968-1972). Schroeder and Todd 1990 = Schroeder, F. M. and Todd, R. B. Two Greek Aristotelian Commentators on the Intellect (Toronto 1990). Sharpies 1987 = Sharpies, R. W. ‘Alexander of Aphrodisias: scholasticism and innovation’, ANRW 36.2 (Berlin - New York 1987), 1176-1243. Vajda 1951= Vajda, G. ‘Les notes d ’Avicenne sur la “Theologie d ’Aristote”,’ Revue Thomiste 51 (1951) 346­ 406. Verbeke 1966 = Verbeke, G. Jean Philopon. Commentaire sur le De anima d ’Aristote (Louvain ­ Paris 1966).

V

AVICENNA’S EASTERN (“ORIENTAL”) PHILOSOPHY NATURE, CONTENTS, TRANSMISSION I. INTRODUCTION

Avicenna’s Eastern philosophy, what has been predominantly called “O riental” in the secondary literature,1 became some­ thing of a minor cause celebre among Avicennisants in the cen­ tury th at is just ending because it was assumed to present an esoteric, mystical aspect of Avicenna th at is not apparent in his other works. After a review of all the evidence th a t was avail­ able to me in 1988, it became reasonably clear th a t Avicenna’s book on Eastern philosophy differed from the rest of his works, and especially from al-Sifa’ ( The Cure), only i n substance. As Avicenna himself claimed in the prologue of the S ifa ’ (text quoted at the beginning of the next section), his pur­ pose in his book on Eastern philosophy was to present philoso­ phy as such, i.e., in a systematic and not historical way, to avoid references to views of other philosophers, and to use a direct form of exposition even if this m eant contradicting some cher­ ished and traditional views (and thus ruffling some feathers). He also intended, as he added in his introduction to the book on Eastern philosophy,2 to be selective in his treatm ent: his pur­ pose was to include only those subjects in the fields of logic, metaphysics, physics, and ethics th at were controversial; the 1 It is now best to discontinue the use of “Oriental” as translation of the Arabic m asriqiyya. Although literally correct (oriens is Latin for East), the term is currently loaded with cultural concepts, peculiar to the late tw entieth century, that should not be transferred to Avicenna ’s use of m asriqiyya, by which he merely intended to refer to the Islamic East, i.e., Hurasan. It is more advisable to use the neutral English term “Eastern. ” See the discussion of this point in D. Gutas, Avicenna and the A ristotelian Tradition. Introduction to R eadin g A vicenna ’s Philosophical Works, Islamic Theology and Philosophy, 4 (Leiden, 1988), p. 127 and note 26, with additional docu ­ m entation in D. Gutas, “Ibn Tufayl on Ibn Slna ’s Eastern Philosophy, ” Oriens, 34 (1994): 222 - 41, p. 223 note 2. 2 See the text in Gutas, Avicenna, p. 120, § 17.

V 160

rest, about which there was no disagreement, could be readily studied in extant treatises and needed no further comment.3 In the decade since th at review was conducted I have col­ lected some additional evidence th at corroborates this view; I have also not seen any studies th at challenge it.4 The purpose of this paper is to present this evidence as a further installm ent in the efforts to round out the account about Avicenna’s Eastern philosophy. II. THE MYTHS OF THE “MYSTICAL” AND “ILLUMINATIONIST” EASTERN PHILOSOPHY

The reasons why Avicenna’s book on Eastern philosophy should be assumed in the first place to be esoteric or mystical without any, or even against, prim a facie evidence - since for most of this century what was known of it was only the section on logic5 - are complicated and have a long history. It is well worth retelling the highlights of the story, which continues to this day, if only as a brief exercise in the sociology of knowledge.6 In the introduction to his Hayy b. Yaqzan, Ibn Tufayl creates the fiction of an esoteric and exoteric Avicenna through deliber­ ate m isinterpretation and m isrepresentation of Avicenna’s Prologue to the S ifa ’.In th at Prologue Avicenna says t lowing, comparing the S ifa ’ and his book on Eastern philosophy (al-falsafa al-m asriqiya): I also wrote a book other than these two [the Sifa ’ and the Lawahiq (Appendices)], in which I presented philosophy as it is naturally [perceived]7 and as required by an unbiased view which neither takes into account in

3 See the discussion in Gutas, Avicenna, pp. 115 -30. 4 The more recent mantric statem ents by someone like S.H. N asr will be discussed in the following section. 6 The extant section on logic from Avicenna ’s work on Eastern philosophy was first published under the artificial title M antiq (Cairo, 1910) reprinted in a photographic reproduction in Tehran by the Maktabat al- Ga ‘fari alTabrizi (no date given, but ca. 1970) and in Qum in 1985, and, in a new typeset, in Beirut in 1982, w ith an introduction by Sukri al-Naggar. See now H. Daiber, B ibliography o f Islam ic Philosophy (Leiden, 1999), vol. I, p. 477, no. 4607. 6 1 have mentioned some of these reasons in a number of places; see Gutas, Avicenna, pp. 129 -30; D. Gutas, “Avicenna: M ysticism , ” in Encyclopaedia Iranica, III, 82a-b; and Gutas “Ibn Tufayl, ” pp. 231 ­4. ‘aid m a hiya fi a l - tab ‘,“as it is in nature. ” In my Avicenna book I had translated it “as it is in itself, ” taking the word ta b ‘, nature, to refer to the nature

V AVICENNA’S EASTERN (“ORIENTAL”) PHILOSOPHY

161

[this book on Eastern philosophy] the views of colleagues in the discipline, nor takes precautions here against creating schisms among them as is done elsewhere; this is my book on Eastern philosophy. But as for the present book [the Sifa9], it is more elaborate and more accommodating to my Peripatetic colleagues. Whoever wants the truth [stated] without indirection, he should seek the former book [on Eastern philosophy]; whoever wants the truth [stated] in a way which is somewhat conciliatory to colleagues, elabo­ rates a lot, and alludes [talwih] to things which, had they been perceived, there would have been no need for the other book, then he should read the present book [the Sifa9].8

The distinction in style between the two books to which Avicenna is referring here Ibn Tufayl m isinterprets as a differ­ ence in doctrine: he says, echoing the words of Avicenna, As for the books of Aristotle, Avicenna undertook in the Sifa9 to interpret their contents, proceeding according to Aristotle’s doctrine and following the method of his philosophy. But in the beginning of the book, Avicenna stated explicitly that in his opinion the truth is something else [al-haqq ‘indahu gayru dalika], that he wrote the Sifa9 according to the doctrine of the Peripatetics only, and that “whoever wants the truth without indirection should seek” his book on Eastern philosophy.9

As is evident even from the quotations above, Avicenna nowhere states “explicitly th at in his opinion the tru th is some­ thing else” than what he has in the Sifa'; he is not talking about a difference in doctrine but one in style.10 Ibn Tufayl created of philosophy. In his review of my book, M. Marmura ( “Plotting the course of Avicenna ’s thought, ” Journal o f the Am erican Oriental Society, 111 [1991]: 333 -42, pp. 339 -40) suggested that ta b ‘ refers to the “natural cognitive activity ” of the person doing philosophy. The parallels which Marmura adduces from Avicenna all have the expression bi - al - tab ‘, “by nature, ” and not fi a l - ta b \ “in nature, ” and thus are not precisely relevant; it is only al-Gazali who uses fi al - tab c (p. 337a), which is a differ ­ ent matter. I am much more swayed to Marmura ’s interpretation by the words alr a ’y al - sarih ( “unbiased view ”) in the text of Avicenna immediately following. They clearly refer to the view of the person studying philosophy, and as such they can be taken to be parallel to the word ta b ‘, which would also refer to the nature, i.e., nat ­ ural intelligence, of the same person. In either case, however, the meaning would be roughly the same: a person who has an unbiased view, i.e., who is not misled by h is ­ torical or accidental misinterpretations in his study of philosophy, can have a clear view of philosophy as it is in itself by letting his natural intelligence guide him. 8 Translation in Gutas, Avicenna, pp. 52 -3, §4. Text in Ibn Sina, a l - S ifa ’, alM antiq, al - M adhal, ed. El-Ahwani, Anawati, El-Khodeiri (Cairo, 1952), p. 10.11 -17. 9 Translation in Gutas, “Ibn Tufayl, ” p. 226a; text in L. Gauthier, Hayy Ben Yaqdhan. Rom an philosophique d lb n Thofail (Beirut, 1936), pp. 14 -15. 10 Ibn Tufayl is defended against my thesis by A. Elamrani-Jamal “Experience de la vision contemplative et forme du recit chez Ibn Tufayl, ” in M.A. Amir - Moezzi (ed.), Le voyage initiatique en terre d ’Islam (Louvain - Paris, 1996), p. 165, note 25. He says

V 162

this fiction in order to foist upon Avicenna, and thereby gain authority for his own epistemology in which the mystical vision plays a leading role.11 In addition to this passage, which is quite explicit, it is also clear from the whole tenor of Ibn Tufayl’s introduction th at he m eant to generate the impression th at Avicenna’s Eastern philosophy has somehow to do with mysti­ cism. The fact th at Ibn Tufayl had no access to a book by Avicenna by this title made the fiction all the more plausible as the actual book was not available for verification. Given a free hand, Ibn Tufayl also added to his Hayy b. Yaqzan a subtitle th at went farther along in this direction: he called it fi asrar althat Ibn Tufayl did not oppose the S ifa ’ to the Eastern philosophy in term s of doc ­ trine. This objection, however, overlooks Ibn Tufayl’s unambiguous statem ent that Avicenna claimed that the truth was som ething else than what was contained in the S ifa ’ (