The Arabic, Hebrew and Latin Reception of Avicenna's Metaphysics 3110215756, 9783110215755, 9783110215762, 2011007822

Citation preview

The Arabic, Hebrew and Latin Reception of Avicenna’s Metaphysics

Scientia Graeco-Arabica herausgegeben von Marwan Rashed

Band 7

De Gruyter

The Arabic, Hebrew and Latin Reception of Avicenna’s Metaphysics edited by

Dag Nikolaus Hasse and Amos Bertolacci

De Gruyter

Gedruckt mit freundlicher Unterstützung der VolkswagenStiftung.

ISBN 978-3-11-021575-5 e-ISBN 978-3-11-021576-2 ISSN 1868-7172 Library of Congress Cataloging-in-Publication Data The Arabic, Hebrew and Latin reception of Avicenna’s Metaphysics / edited by Dag Nikolaus Hasse and Amos Bertolacci. p. cm. ⫺ (Scientia graeco-arabica) Includes bibliographical references and index. ISBN 978-3-11-021575-5 (hardcover : alk. paper) 1. Avicenna, 980⫺1037. Ilahiyat. 2. Metaphysics 3. Islamic philosophy. 4. Jewish philosophy. 5. Philosophy, Medieval. I. Hasse, Dag Nikolaus. II. Bertolacci, Amos. B751.I483A73 2011 110⫺dc22 2011007822

Bibliographic information published by the Deutsche Nationalbibliothek The Deutsche Nationalbibliothek lists this publication in the Deutsche Nationalbibliografie; detailed bibliographic data are available in the Internet at http://dnb.d-nb.de. 쑔 2012 Walter de Gruyter GmbH & Co. KG, Berlin/Boston Printing and binding: Hubert & Co. GmbH & Co. KG, Göttingen ⬁ Printed on acid-free paper Printed in Germany www.degruyter.com

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Preface The articles of this volume are presented by scholars who convened in 2008 to discuss their research on the influence of Avicenna’s metaphysics in the Villa Vigoni, Centro italo-tedesco, Menaggio, Italy. We are grateful to the participants and chairs of the congress who do not contribute to this volume: Rdiger Arnzen, Gad Freudenthal, Dimitri Gutas, Maarten Hoenen and Andreas Speer. Special thanks go to two research assistants at the University of Wrzburg: Anna-Katharina Strohschneider, who prepared the papers for type-setting, and Jon Bornholdt, who extinguished mistakes in the texts of contributors who are not native speakers of English. We gratefully acknowledge the generous and unbureaucratic funding of the conference by the VolkswagenFoundation, as part of the Lichtenberg professorship grant to Dag Nikolaus Hasse. Finally, we would like to thank Gregor Vogt-Spira and the staff of the Villa Vigoni who created an ideal atmosphere for the discussion of a research issue of considerable complexity. Dag Nikolaus Hasse Julius-Maximilians-Universitt Wrzburg

Amos Bertolacci Scuola Normale Superiore di Pisa

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Contents Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

V

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1

Jules Janssens Al-Lawkarı¯’s Reception of Ibn Sı¯na¯’s Ila¯hiyya¯t . . . . . . . . . . . . . . . . . . .

7

Robert Wisnovsky Essence and Existence in the Eleventh- and Twelfth-Century Islamic East (Masˇriq): A Sketch . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

27

Stephen Menn Fa¯ra¯bı¯ in the Reception of Avicenna’s Metaphysics: Averroes against Avicenna on Being and Unity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

51

Peter Adamson Avicenna and his Commentators on Human and Divine Self-Intellection

97

Heidrun Eichner Essence and Existence. Thirteenth-Century Perspectives in Arabic-Islamic Philosophy and Theology . . . . . . . . . . . . . . . . . . . . . . .

123

Mauro Zonta Avicenna’s Metaphysics in the Medieval Hebrew Philosophical Tradition

153

Resianne Fontaine ‘Happy is he whose children are boys’: Abraham Ibn Daud and Avicenna on Evil . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

159

Mauro Zonta Possible Hebrew Quotations of the Metaphysical Section of Avicenna’s Oriental Philosophy and Their Historical Meaning . . . . . . . . . . . . . . . . .

177

Amos Bertolacci On the Latin Reception of Avicenna’s Metaphysics before Albertus Magnus: An Attempt at Periodization . . . . . . . . . . . . . . . . . . . . . . . . .

197

Dag Nikolaus Hasse Avicenna’s ‘Giver of Forms’ in Latin Philosophy, Especially in the Works of Albertus Magnus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

225

Kara Richardson Avicenna and Aquinas on Form and Generation . . . . . . . . . . . . . . . . .

251

VIII

Contents

Pasquale Porro Immateriality and Separation in Avicenna and Thomas Aquinas . . . . .

275

Gabriele Galluzzo Two Senses of ‘Common’. Avicenna’s Doctrine of Essence and Aquinas’s View on Individuation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

309

Martin Pickav On the Latin Reception of Avicenna’s Theory of Individuation . . . . .

339

Giorgio Pini Scotus and Avicenna on What it is to Be a Thing . . . . . . . . . . . . . . . .

365

Index of Avicenna’s Works with Passages Cited . . . . . . . . . . . . . . . . . .

389

Index of Names . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

395

Introduction

˘

Many centuries passed after the composition of Aristotle’s Metaphysics before a metaphysical work of similar size and ambition was written in the Peripatetic tradition: Avicenna’s Ila-hiyya-t (Metaphysics, or: Divine Things), the fourth and last part of the summa Kita-b asˇ-Sˇifa- (Book of the Cure), dating to 1020 – 27 CE. The Ila-hiyya-t is only one of more than a dozen metaphysical works by Avicenna, but in terms of comprehensiveness, systematic effort and influence, it is his most important metaphysical text. It is rivaled only by the Kita-b al-Isˇa¯ra¯t wa-ltanbı¯ha¯t (Book of Pointers and Reminders), a late summa dating to ca. 1030 – 34 CE that contains a substantial metaphysical section of considerable influence in the Arabic tradition. The study of Avicenna’s metaphysics has made important progress in the past few years, due in part to the appearance of studies in monograph format.1 Much, however, remains to be done. Above all, critical editions of Avicenna’s metaphysical works are still lacking,2 and the study of their manuscript tradition is still at a preliminary stage.3 The present book sheds light on Avicenna’s metaphysics itself, but its proper theme is the reception of his metaphysics in three different cultures: Arabic, Hebrew and Latin. In the past few decades, it has increasingly become recognized that Avicenna’s philosophy, and in particular his metaphysics, was of overwhelming influence in the Arabic-speaking world from the eleventh to, at least, the Among recent studies, particularly important are R. Wisnovsky, Avicenna’s Metaphysics in Context, London: Duckworth, 2003, and A. Bertolacci, The Reception of Aristotle’s ‘Metaphysics’ in Avicenna’s ‘Kita¯b al-Shifa¯ ’, Leiden: Brill, 2006. A helpful tool is the collection of articles by M.E. Marmura, Probing in Islamic Philosophy, Binghamton: Global Academic Pub., 2005, as well as recent translations of the Ila-hiyya-t into English by M.E. Marmura (Avicenna, The Metaphysics of ‘The Healing’, Provo, Utah: Brigham Young University Press, 2005) and Italian by O. Lizzini with a preface by P. Porro (Avicenna, Metafisica: la scienza delle cose divine, 2nd edn, Milan: Bompiani, 2006) and A. Bertolacci (Avicenna, Il libro della guarigione: Le cose divine, Torino: Utet, 2008). See the list of Emendanda of Anawati’s Arabic edition of the Ila-hiyya-t (Avicenna, al-Sˇifa¯’, al-Ila¯hiyya¯t, Cairo: al-Hay’a al-‘a¯mma, 1960) in Bertolacci, The Reception (as in n. 1). On the Ila-hiyya-t, see A. Bertolacci, On the Manuscripts of the Ila¯hiyya¯t of Avicenna’s Kita¯b al-Sˇifa- , in A. Akasoy, W. Raven, eds, Islamic Thought in the Middle Ages. Studies in Text, Transmission and Translation, in Honour of Hans Daiber, Leiden: Brill, 2008, pp. 59 – 75, and on the Sˇifa- in general, the papers presented at the International Colloquium The Manuscript Tradition of Avicenna’s Kita-b asˇ-Sˇifa- : The Current State of Research and Future Prospects, Scuola Normale Superiore di Pisa, 22 – 24 September 2010 (proceedings forthcoming in Oriens, 40, 2012). ˘

1

˘

3

˘

2

˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

2

sixteenth century. Even those thinkers who fundamentally disagreed with Avicenna often developed their intellectual standpoint in confrontation with the Avicennian tradition. Indications of the scope of Avicenna’s influence in Arabic are the many extant commentaries on his works, the lively reception of his metaphysical theories by Islamic theologians, and the reading of his philosophy in the madrasa.4 Research on the later period of Arabic philosophy after Avicenna is still in its infancy. This is particularly true of the centuries after Fahr ˘ al-Dı¯n al-Ra¯zı¯, who died 1210 CE. Lists have been drawn up of the authors and works that should be studied and researched, but since the details of the intellectual history of the period are still unknown, the time has not yet come for generalizations. It is an open question, for example, which of Avicenna’s texts transmitted which theories. As in the case of al-Lawkarı¯, it may well be that an author read several metaphysical texts by Avicenna: Ila-hiyya-t, al-Mabda’ wa-lma‘a¯d, Isˇa¯ra¯t, Ta lı¯qa¯t, etc. (see the article by Janssens in this volume). Also, it can be shown that interpretations of Avicenna were often influenced by previous readers (see the articles by Wisnovsky and Menn). Turning to thirteenth- and fourteenth-century sources, one realizes that Avicennian theories were often transmitted through intermediate sources, such as the influential works of Fahr ˘ al-Dı¯n al-Ra¯zı¯, or through philosophical handbooks (see the articles by Adamson and Eichner). Despite the fact that so much remains unknown about this period, it is already apparent that the reception of Avicenna’s metaphysical theories in later Arabic thought gave rise to a wealth of metaphysical discussions of impressive intellectual quality. The textual transmission of Avicenna’s metaphysics in the Latin speaking world is better known. There are solid grounds for believing that Dominicus Gundisalvi, also called Gundissalinus by the scholastics, an archdeacon and canon of the cathedral of Toledo, was the translator of Avicenna’s Ila-hiyya-t from Arabic into Latin between 1150 and 1180.5 The Latin title was Liber de ˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Introduction

4

5

Recent studies: D. Gutas, The Heritage of Avicenna: The Golden Age of Arabic Philosophy, 1000 – ca. 1350, in J. Janssens, D. de Smet, eds, Avicenna and His Heritage, Leuven: Leuven University Press, 2002, pp. 81 – 97; R. Wisnovsky, The Nature and Scope of Arabic Philosophical Commentary in Post-Classical (ca. 1100 – 1900 AD) Islamic Intellectual History: Some Preliminary Observations, in P. Adamson et al., eds, Philosophy, Science and Exegesis in Greek, Arabic and Latin Commentaries, London: Institute of Classical Studies, 2004, vol. 2, pp. 149 – 91; G. Endreß, Reading Avicenna in the Madrasa: Intellectual Genealogies and Chains of Transmission of Philosophy and the Sciences in the Islamic East, in J.E. Montgomery, ed., Arabic Theology, Arabic Philosophy. From the Many to the One: Essays in Celebration of Richard M. Frank, Leuven: Peeters, 2006, pp. 371 – 422; H. Eichner, Dissolving the Unity of Metaphysics: From Fakhr al-Din al-Razi to Mulla Sadra al-Shirazi, Medioevo, 32, 2007, pp. 139 – 97. Dominicus Gundisalvi is identified as the translator in the colophon of three of the 25 manuscripts. See Avicenna, Liber de philosophia prima sive Scientia divina, I – IV, ed. S. Van Riet, Louvain/Leiden: Peeters/Brill, 1977, p. 123*, n. 2.

3

Introduction

6 7

8

˘

˘ ˘

philosophia prima sive scientia divina. This text gradually found its readers in the Latin West (see the article by Bertolacci) and reached the high point of its influence in the period from Thomas Aquinas to John Duns Scotus (as evidenced in the articles of Richardson, Porro, Galluzzo, Pickav and Pini). The manuscript transmission of the Philosophia prima thins out considerably after 1400 CE: 15 manuscripts are extant from the thirteenth century, 7 from the fourteenth, 3 from the fifteenth.6 But Avicenna’s doctrines continued to be discussed in the Renaissance (see the article by Hasse). The Latin reception of Avicenna’s metaphysics was also influenced by the ˙ aza-lı¯’s Maqa-sid al-fala-sifa (Intentions of the Philosophers) into translation of al-G ˙ degree an intelligent reworking of Avicenna’s Latin. The Maqasid are to a large ˙ Persian Da-nesˇna-me-ye Ala-¯ı (Philosophy for Ala- -al-Dawla) and thus exhibit basic teachings of Avicenna, though not always faithfully. Since, until the early ˙ aza-lı¯ known in the Latin West, fourteenth century, this was the only text by al-G ˙ the scholastics read al-Gazalı¯ as a sequax Avicennae. Another influence on the Western reading of Avicenna’s metaphysics was Averroes’ Long commentary on Aristotle’s Metaphysics, which contains several passages that engage in criticism of Avicenna’s metaphysical theories. There exist a good number of in-depth studies on the reception of Avicenna by individual scholastic authors and on the reception of certain Avicennian theories, e. g. on the primary notions or the subject matter of metaphysics,7 but scholarship has not yet arrived at a comprehensive picture of Avicenna’s influence on Latin metaphysics. The present volume is meant as a contribution to such a picture. The importance of the issue is widely recognized, in view of the pivotal significance of Avicenna for the formation of metaphysical thought in high scholasticism. The Ila-hiyya-t of The Cure was not translated into Hebrew, but Avicenna’s metaphysics nevertheless influenced medieval Jewish thought.8 This influence ˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Avicenna, ibid., p. 127*. To mention only a few studies: J. Aertsen, Avicenna’s Doctrine of the Primary Notions and its Impact on Medieval Philosophy, in Akasoy, Raven (as in n. 3), pp. 21 – 42; A. Zimmermann, Ontologie oder Metaphysik? Die Diskussion ber den Gegenstand der Metaphysik im 13. und 14. Jahrhundert; Texte und Untersuchungen, Leuven: Peeters, 1998; J.F. Wippel, The Latin Avicenna as a Source for Thomas Aquinas’s Metaphysics, Freiburger Zeitschrift fr Philosophie und Theologie, 37, 1990, pp. 65 – 72. See the bibliography in P. Porro’s preface to Lizzini’s Italian translation of the Ila-hiyya-t (as in n. 1) and J. Janssens’ An Annotated Bibliography on Ibn Sı¯na-, Leuven: Leuven University Press, 1991, with its First Supplement, Louvain-la-Neuve: FIDEM, 1999. See also the collection of articles on the Latin transmission of Avicenna’s works by M.-Th. d’Alverny, Avicenne en occident, Paris: Vrin, 1993. S. Harvey, Avicenna’s Influence on Jewish Thought: Some Reflections, in Y.T. Langermann, ed., Avicenna and his Legacy. A Golden Age of Science and Philosophy, Turnhout: Brepols, 2009, pp. 327 – 40.

4 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Introduction

occurred through Jewish philosophers reading Avicenna in Arabic, the partial translation of Avicenna’s Kita-b an-Nag˘a-t (Book of Salvation) into Hebrew, and – ˙ aza-lı¯, in particular the indirectly – the Hebrew translations of Averroes and al-G ˙ translation of al-Gazalı¯’s Intentions of the Philosophers (see the articles by Zonta). Some topics of this reception have received detailed attention in scholarship, such as Maimonides’ usage of several Avicennian doctrines: the proof of God’s existence, the concept of the necessary being, and the distinction between essence and existence.9 Other issues remain to be settled, such as the identification of the Avicennian texts read and quoted by Jews directly from the Arabic, for example by Abraham ibn Daud, the Andalusian philosopher of the twelfth century (see the article by Fontaine), or the transportation of ˙ aza-lı¯’s Intentions of the Avicennian ideas through Hebrew commentaries on al-G Philosophers. An important question is whether there is evidence for a current of ‘Jewish Avicennism’ in the fourteenth century rivaling the dominant philosophical current influenced by Averroes, as Mauro Zonta has suggested.10 It is clear, however, that Avicenna’s overall influence in Hebrew philosophy was never on the same scale as that of Averroes. Some articles of the present volume are devoted to the textual transmission of Avicenna’s metaphysics in different cultures (Janssens, Zonta, Bertolacci). Others study the reception of several prominent doctrines of Avicenna: the distinction between essence and existence (Wisnovsky, Eichner, Porro); the doctrine of primary notions (Menn, Pini); the theory of individuation and universals (Galluzzo, Pickav); emanation theory and the related topics of the necessary being, the emanation of the intelligences, the nature and function of the giver of forms, and the origin of evil (Adamson, Hasse, Richardson, Fontaine). It is the explicit aim of the present volume to facilitate the comparison between the reception processes in distinct cultures and times, thus contributing to our knowledge both of Avicenna’s metaphysics itself, through the lenses of his medieval readers, and of its culturally complex reception history.

9 Among the many studies on this topic, only two shall be mentioned: M. Zonta, Maimonides’ Knowledge of Avicenna. Some Tentative Conclusions About a Debated Question, in G. Tamer, ed., The Trias of Maimonides/Die Trias des Maimonides. Jewish, Arabic, and Ancient Culture of Knowledge/Jdische, arabische und antike Wissenskultur, Berlin/New York: de Gruyter, 2005, pp. 211 – 22; and the chapter on the influence of Avicenna’s metaphysical proof of God in the classic study by H.A. Davidson, Proofs for Eternity, Creation and the Existence of God in Medieval Islamic and Jewish Philosophy, New York/Oxford: Oxford University Press, 1987, pp. 378 – 406. 10 M. Zonta, The Role of Avicenna and of Islamic ‘Avicennism’ in the 14th-Century Jewish Debate Around Philosophy and Religion, Oriente moderno, 80, n.s. 19, 2000, pp. 647 – 60, and M. Zonta, Avicenna in Medieval Jewish Philosophy, in Janssens, De Smet (as in n. 4), pp. 267 – 79.

Introduction 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

5

The Avicennian doctrines studied in this volume are important, but they also represent a somewhat accidental choice, since they reflect what several contemporary scholars in the field are currently working on. Equally important issues of Avicennian metaphysics (for example, the subject-matter of metaphysics, the proof of God’s existence, the metaphysically grounded prophetology, and the theory of substance and accident) are only treated cursorily. In this connection it is important to remember that Avicenna’s metaphysics is very rich in content and coherent in structure and ought not to be reduced to a sample of famous doctrines. There are many chapters in the Ila-hiyya-t and the Isˇa¯ra¯t that still await more detailed analysis – not to speak of studies on the reception of these chapters. On the other hand, the reception history shows that two doctrines in particular were extremely successful in all three cultures considered here and in both philosophical and theological milieus: the distinction between essence and existence, and the concept of the necessary existent by itself. The medieval readers of Avicenna thus testify to the remarkable philosophical originality of these doctrines.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Al-Lawkarı¯’s Reception of Ibn Sı¯na¯’s Ila¯hiyya¯t Jules Janssens Al-Lawkarı¯, who was born at Lawkar near Merw (Iran) at an unknown date and who died most likely at the beginning of the twelfth century (in 1123 at the latest), is presented by al-Bayhaqı¯ as a disciple of Bahmanya¯r, a first-generation student of Ibn Sı¯na¯1. If this is correct, he is a second-generation student of the Sˇayh al-ra’ı¯s. However, if he really died in 1123, it would be surprising that he had˘ been a direct disciple of Bahmanya¯r, who died in 1066 at the latest. Whatever the case, he is known as the author of a fihrist, i. e. a list, of Ibn Sı¯na¯’s Ta lı¯qa¯t, but, above all, as the author of a major philosophical encyclopedia, entitled Baya¯n al-haqq bi-dima¯n al-sidq 2. This latter encompasses three major ˙ ˙ ˙ parts: logic, physics and metaphysics. The latter divides into two major parts, i. e., al-‘Ilm al-kullı¯ (Universal Science), and Rubu¯biyya¯t (Lordly Things), followed by al-Risa¯la l-mulhaqa bi-‘ilmi al-rubu¯biyyati, the appended exposition of the ˙ lordly science, and al-Fusu¯l al-muntahabat min ‘ilm al-ahla¯q, the selected ˙ ˘ division into ‘universal ˘ chapters of the moral science. The basic science’ and ‘lordly science’ is, as such, not mentioned by Ibn Sı¯na¯ in the Ila¯hiyya¯t of his Sˇifa¯’, but might ultimately stem from his al-Masˇriqiyyu¯n 3. As to the appellation ‘lordly science’, it probably derives from the prologue to the Kita¯b al-Mabda’ wa-l-ma‘a¯d 4. In any case, al-Lawkarı¯’s division differs substantially from that of Bahmanya¯r, which distinguishes between metaphysics, ma¯ ba‘d al-tabı¯‘a, and ˙ science of the states of the essences of the existing beings, al-‘ilm bi-ahwa¯l a‘ya¯n ˙ ˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

1 2 3

4

See al-Lawkarı¯, Baya¯n. Mantiq. al-Madhal. Muqaddima, pp. 57 – 71. For Bayhaqı¯’s af˙ ¯h, pp. 145 ˘ – 6. firmation, see al-Bayhaqı¯, Tarı ˘ For a more detailed survey of al-Lawkarı¯’s life and works, see Marcotte, Life and Work of al-Lawkarı¯, pp. 134 – 57. See Ibn Sı¯na¯, al-Mantiq al-masˇriqiyyı¯n, pp. 8, 9 – 10 (but note that Ibn Sı¯na¯ uses the expression ‘ilm ila¯hı¯, ˙ not rubu¯biyya¯t). Bertolacci, Reception of Aristotle’s ‘Metaphysics’, pp. 188 – 9, stresses that the bipartition in question breaks with the unitary view of the Sˇifa¯’, but can be regarded as the expression of the more original and independent way of exposition that Ibn Sı¯na¯ avowedly follows in the Masˇriqiyyu¯n. Regarding the names of metaphysics as a discipline, as well of its parts, see ibid., pp. 599 – 605. See Ibn Sı¯na¯, Kita¯b al-Mabda’ wa-l-ma‘a¯d, pp. 1, 7 – 8, where it is stated that ‘the fruit of the branch of metaphysics is Theology, which treats [the subjects of ] lordship, i. e., the First Principle, and the relationship which beings bear to it according to their rank’ (translation, slightly modified, taken from Gutas, Avicenna, p. 31).

8 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Jules Janssens

al-mawg˘uda¯t 5. However, both Ibn Sı¯na¯’s Ila¯hiyya¯t of the Sˇifa¯’ and Bahmanya¯r’s Kita¯b al-Tahs¯ıl are largely used as sources in these basic parts, as is already shown ˙˙ his edition of the Baya¯n. In the part entitled Universal Science, they by Dı¯ba¯g˘¯ı in even constitute the only sources. He reproduces verbatim entire chapters or, at least, large parts of them of both works. With respect to the Ila¯hiyya¯t of the Sˇifa¯’, the following correspondences come to the fore: Baya¯n c. 1 – 4 c. 14 c. 18 – 21 c. 24 c. 29 – 37 c. 40 – 41

Ila¯hiyya¯t of Sˇifa¯’ I, 1 – 4 (metaphysics as science) III, 1 (accidental categories in general) III, 7 – 10 (quality and relation) IV, 3 (complete/incomplete) V, 5 – 6, 8 – 9; VI, 1 – 4 (species, differentia and definition)6 VII, 2 – 3 (refutation of Platonic Ideas).

As to Bahmanya¯r’s Kita¯b al-Tahs¯ıl, one may point out these derivations: ˙˙ Baya¯n c. 6 – 13 c. 15 – 17 c. 22 – 3 c. 25 – 8 c. 38 c. 39

Tahs¯ıl ˙˙ II, 1, 3 – 4, 7 – 8, 11 – 13 (truth, substance, matter and form)7 II, 2, 3 – 5 (quantity) II, 3, 1 – 2 (anteriority/posteriority; potency/act) II, 4, 2 – 5 (universal/particular; genus/matter) II, 5, 3 (chance/fortune/final cause) II, 6, 1 (unity/multiplicity).

It has to be noted that many of these chapters drawing on Bahmanya¯r are clearly influenced also by Ibn Sı¯na¯’s Ila¯hiyya¯t of the Sˇifa¯’, as indicated by al-Dı¯ba¯g˘¯ı. Some sources have escaped the editor: regarding chapter 7 (Tahs¯ıl, II, 1, 4) there ˙˙ exists, besides an inspiration from Ila¯hiyya¯t, II, 1, a strong influence from the ˇ Maqu¯la¯t of the Sifa¯’ (I, 4). On the other hand, one looks in vain for any 5 6

7

See Bahmanya¯r, Tahs¯ıl, pp. 277 and 567. Regarding the significant deviation from ˙˙ Avicenna that is involved in this distinction, see Janssens, Faithful Disciple, pp. 188 – 9. In his Baya¯n, al-Lawkarı¯ has surprisingly reversed the order of the chapters 6 and 8 of the Ila¯hiyya¯t. Moreover, he has divided the text of c. 3 (on the compatibility between the efficient causes and their effects) over two chapters, presenting the second part as a detailing (tafs¯ıl) of the basic affirmation of the first. ˙ that the title of c. 6 of the Baya¯n is derived from the Ila¯hiyya¯t of his Sˇifa¯’, Despite the fact I, 8, the text of the chapter reproduces Tahs¯ıl, II, 1, 2, pp. 291,12 – 293,5. C. 13 corresponds to c. II, 1, 13 of the Tahs¯ıl, not 3˙˙as indicated in al-Lawkarı¯, Baya¯n. al-‘Ilm ˙ observed that c. II, 1, 11 of the Tahs¯ıl (corporeal al-ila¯hı¯, p. 70, n. 1. Finally, it has to˙be ˙˙ for its end matter is not devoid of form) has been copied in c. 10 of the Baya¯n, except (see Bahmanya¯r, Tahs¯ıl, p. 336,1–12 – on the natural form), a literal copy of which constitutes c. 11. ˙˙

Al-Lawkarı¯’s Reception of Ibn Sı¯na¯’s Ila¯hiyya¯t 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

9

reference to Ila¯hiyya¯t, II, 4, with respect to chapter 12 (Tahs¯ıl, II, 1, 12), or to V, ˙ 6, 8 and 9 regarding chapter 28 (Tahs¯ıl, II, 4, 5). Finally,˙chapter 38 (Tahs¯ıl, II, ˙˙ ˙ ¯‘ı¯ of 5, 3) has not only been inspired by chapter 14 of book 1 of al-Sama¯‘ al-t˙abı ˙ 8 ˇ the Sifa¯’, but also by chapter 13 of the same book . As to chapter 5, it is also based on both works, but it is distinctive insofar as it first (pp. 27,2 – 28,9) quotes the very beginnings of I, 5 (regarding the primary concepts and principles) of the Ila¯hiyya¯t of the Sˇifa¯’ (Sˇifa¯’, Ila¯hiyya¯t, pp. 29,3 – 30,4), and then continues (pp. 28,9 – 32) with the reproduction of chapters from Bahmanya¯r’s Kita¯b al-Tahs¯ıl: the entire chapter II, 1, 2 (on ˇsay’, ˙˙ thing, as distinguished from being and non-being) and the beginning of chapter II, 1, 3 (on possibility, impossibility and necessity; Tahs¯ıl, pp. 290,14 – 291,11)9. ˙˙ direct links with Sˇifa¯’, In Bahmanya¯r’s wording one easily detects several Ila¯hiyya¯t, I, 5. Surveying the totality of these derivations, one sees that al-Lawkarı¯ covers almost the entirety of books 1 – 7 of the Ila¯hiyya¯t of the Sˇifa¯’. Only four chapters seem to have been completely omitted, i. e., I, 6 – 7 (on the Necessary Being); III, 3 (on unity and multiplicity, and the accidentality of number) and V, 4 (what among the things contained in the genus render it a species or not a species?). The omission of the former two is certainly not by chance, since a treatment of the very idea of the Necessary Existent evidently has its natural place in the section on lordly things, not in that of the general ontology of the universal science. As to chapter III, 3, it does not so much offer new ideas as rather study in depth the difficult issues of unity and accidentality of number. In fact, it deals with the aporetic nature of the notions of unity and multiplicity and moreover proves the inseparability of unity and substance (despite the fact that unity does not constitute the quiddity of the substance) by way of diaeresis and reductio ad absurdum10. Similarly, chapter V, 4 can be considered as dealing with a particular examination of a difficult topic, i. e., the relation between a genus and its differentiae, both notions being dealt with in other chapters. In ignoring them, al-Lawkarı¯ wanted perhaps to avoid too technical questions. For 8 I have limited myself here to mentioning only the most important omissions or mistakes by al-Dı¯ba¯g˘¯ı. For more details regarding Bahmanya¯r’s derivations from Avicennian texts regarding the chapters II, 1, 4 – 13 and II, 2, 3 – 5, see Janssens, Revision, pp. 99 – 117 (102 – 4 and 110). In c. 28 (Tahs¯ıl, II, 4, 5), one detects an inspiration of Ila¯hiyya¯t, V, 8; V, 6 and V, 9 on pp. 185 – 7 in˙˙the order as indicated here. As to c. 38 (Tahs¯ıl, II, 5, 3), the inspiration of elements of c. 13 of al-Sama¯‘ al-tabı¯‘ı¯, see Ibn Sı¯na¯, Sˇifa¯’,˙˙al-Sama¯‘ altabı¯‘ı¯, pp. 63 – 6 (detectable on pp. 239 – 41 of the˙ Baya¯n). It has to be noted moreover ˙that the opening paragraph is also identical with the final paragraph of Ila¯hiyya¯t, VI, 4, see Ibn Sı¯na¯, Sˇifa¯’, Ila¯hiyya¯t, p. 283, 4 – 9. 9 The rest of c. II, 1, 3, which deals with truth and nullity, will be reproduced in c. 6 of the Baya¯n. 10 Lizzini rightly stresses this double method of analysis in the given context, see Ibn Sı¯na¯, Metafisica, pp. 207 – 9.

10 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Jules Janssens

the very same reason, he might have dismissed the last part of VI, 3 (Sˇifa¯’, Ila¯hiyya¯t, pp. 274,5 – 278,8), where a rather technical objection against the theory that the recipient of an act cannot be equal to its agent – one is more burnt when one puts one’s hand in molten metal than in fire, hence the molten metal is hotter than fire, although it became hot by the fire – is discussed. However, in this case the omission can be explained (perhaps primarily) by Ibn Sı¯na¯’s remark that this discussion more properly belongs to the art of physics (Sˇifa¯’, Ila¯hiyya¯t, pp. 275,18 – 276,1)11. But why has he preferred on occasion reproducing Bahmanya¯r’s rewording to giving the very text of the related chapters of the Ila¯hiyya¯t of the Sˇifa¯’? One could be tempted to answer: in order to avoid being shown up as a plagiarist. But why has he then not modified Ibn Sı¯na¯’s text more significantly? Moreover, one may not forget that in his lifetime authors’ rights did not exist, and scholars were constantly copying large extracts from their predecessors’ works. Certainly, al-Lawkarı¯ has taken the practice to an excess. So, could it be that he, in acting this way, wanted to partake in what I have qualified elsewhere as a revision of Ibn Sı¯na¯’s metaphysics? The inclusion of two chapters, one of which is linked with the Maqu¯la¯t, the other with al-Sama¯‘ al-tabı¯‘ı¯ of the Sˇifa¯’, as indicated ˙ case. Indeed, it seems to blur above, might at first sight suggest that this is the the limits between metaphysics, on the one hand, and logic and physics, on the other, just as Bahmanya¯r had done. However, as soon as one looks more carefully, it appears immediately that this is anything but evident. Indeed, when Bahmanya¯r in his Tahs¯ıl, II, 1, 4, ˙ used rather Maqu¯la¯t, I, 4 than Ila¯hiyya¯t, I, 2 of the Sˇifa¯’, he simply ˙followed Ibn ˇ Sı¯na¯’s own indications. The Sayh al-ra’ı¯s, in the latter text says: ‘This (i. e., to ˘ a substance and an accident with respect to claim that something can be both two things) is a grave error. We have discussed it fully in the first parts of Logic. For, even though it was not the [proper] place [for discussing it], it was there that they committed this error’12. The reference is clearly to the Maqu¯la¯t, which is the second book of the logical section of the Sˇifa¯’ 13. The exact identification of the chapter referred to is not easy, but Bahmanya¯r obviously has identified it as I, 4 – and in contemporary scholarship M. Marmura fully agrees with him14. 11 The omission of a great part of Sˇifa¯’, Ila¯hiyya¯t, VI, 5 (pp. 288,12 – 300,9) is undoubtedly due to the reliance on Bahmanya¯r, but can as well be explained in a similar line. 12 Ibn Sı¯na¯, Sˇifa¯’, Ila¯hiyya¯t, II, 1, p. 58,14 – 16; English translation of Marmura (Ibn Sı¯na¯, The Metaphysics of ‘The Healing’, p. 46,29 – 31). 13 The explicit reference to the first parts, seems to have misled Van Riet (Ibn Sı¯na¯, Liber de philosophia prima, V–X, p. 66, n. 37 – 8). She offers what is in my view a mistaken reference to Logyca. Prima pars (i. e., Isagog), Venice 1508, fol. 4r. This reference has been taken over by Lizzini in Ibn Sı¯na¯, Metafisica, p. 1080, n. 11. 14 See Marmura (Ibn Sı¯na¯, The Metaphysics of ‘The Healing’, II, 1, p. 389, n. 4). Contrary to him, Horten (Ibn Sı¯na¯, Metaphysik, p. 91, n. 6), Anawati (Ibn Sı¯na¯, La mtaphysique du

Al-Lawkarı¯’s Reception of Ibn Sı¯na¯’s Ila¯hiyya¯t 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

11

As to the introduction of two chapters of al-Sama¯‘ al-tabı¯‘ı¯, it once more has an ˙ explicit basis in the Ila¯hiyya¯t, this time VI, 5, where one reads: ‘Regarding coincidence and its being an end, we have finished [discussing] it in the Physics.’15 This time, there is no doubt about the locus referentiae, i. e. I, 13, to be complemented naturally with I, 14, insofar as this latter chapter refutes mistaken conceptions regarding luck and coincidence that had been exposed in the former16. So, in both cases the introduction of textual materials that are foreign to the Ila¯hiyya¯t of the Sˇifa¯’ are supported by Ibn Sı¯na¯’s own words. Moreover, it has to be noted that al-Lawkarı¯ never quotes a chapter of Bahmanya¯r’s Kita¯b al-Tahs¯ıl, where this latter deviates perceptibly from Ibn ˙˙ Sı¯na¯’s basic perspective. In this sense, it is most relevant that he, contrary to Bahmanya¯r, does not mention in the present context any of the other categories than the four (i. e., substance, quality, quantity and relation) treated by Ibn Sı¯na¯ himself in the context of metaphysics. Herein I detect a fidelity to the original metaphysical project of the venerable Master, i. e., Ibn Sı¯na¯, and, at the same time, a rejection of the revised version elaborated by one of the latter’s most important immediate disciples, i. e., Bahmanya¯r. However, thus far I have only dealt with the part on universal science. Hence, one may wonder whether the above conclusion remains also valid for that on the lordly things. It is immediately striking that in this part, one finds, with the exception of two small fragments, no reference to Bahmanya¯r’s Kita¯b al-Tahs¯ıl. The vast majority of the chapters (which are enumerated anew from 1) ˙˙ are once again a literal, or, at least, almost literal quotation of chapters of the Ila¯hiyya¯t of the Sˇifa¯’: Baya¯n c. 1 – 2 c. 3 – 7 c. 8

Ila¯hiyya¯t of Sˇifa¯’ VIII, 1, 3 (finitude of causes) VIII, 4; I, 7; VIII, 5 – 7 (Necessary Existent and attributes) IX, 1 (activity of God)

Sˇifa¯’, I–V, p. 334, n. 58,14) (based on the commentary of Mulla¯ Sadra¯ al-Sˇ¯ıra¯zı¯) and Bertolacci (Ibn Sı¯na¯, Libro della guarigione, p. 231, n. 11), refer to˙ I, 6 (Horten adds moreover II, 1 – 3). Based solely on its title, this latter chapter appears to be the best candidate, but if one looks at what is really at issue I, 4 may be preferable. To settle this delicate issue clearly exceeds the limits of the present paper. 15 Ibn Sı¯na¯, Sˇifa¯’, Ila¯hiyya¯t, VI, 5, p. 284,8 – 9; English translation of Marmura (Ibn Sı¯na¯, The Metaphysics of ‘The Healing’, p. 220,28 – 9). 16 Among contemporary scholars, Lizzini (Ibn Sı¯na¯, Metafisica, p. 1187, n. 178), and Bertolacci (Ibn Sı¯na¯, Libro della guarigione, p. 549, n. 162) refer to both chapters, whereas Marmura (Ibn Sı¯na¯, The Metaphysics of ‘The Healing’, VI, 5, p. 410, n. 3) and Van Riet (Ibn Sı¯na¯, Liber de philosophia prima, p. 327, n. 76) refer to the sole c. 13. As to Horten (Ibn Sı¯na¯, Metaphysik, p. 416, n. 3), he refers to chapters 12 – 13. Anawati offers no reference.

12 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

c. 9 c. 14 c. 17 (2. part) c. 20 c. 21 c. 28

Jules Janssens

IX, 2 (pp. 381,15 – 384,12 and 386,7 – 17) (soul as proximate mover of the heavens) IX, 2 (pp. 392,7 – 393,10) (one single mover for the whole universe, but each sphere has a specific mover) IX, 5 (pp. 313,7 – 414,13) (generation of elements) IX, 6 (pp. 415,8 – 418,12) (evil) IX, 7 (return, ma‘a¯d) X, 3 (acts of worship).

A few remarks have to be made: 1. Inside chapter 1 (pp. 272,6 – 273,2), a fragment is inserted into the text ˇ of Sifa¯’, Ila¯hiyya¯t, VIII, 117. It starts with the words ‘according to another way of consideration, he says’. This is surprising, insofar as before one looks in vain for a previous occurrence of ‘he says’, let alone an explicit mention of a particular author. But given that what precedes is a literal quotation of Ibn Sı¯na¯’s Ila¯hiyya¯t of the Sˇifa¯’, one suspects that the reference is to this latter. This turns out to be correct, since the fragment reproduces – once again almost verbatim – part of a text of the Sˇayh al-ra’ı¯s, taken not however from the Sˇifa¯’, but from al-Isˇa¯ra¯t wa˘ precisely, it concerns the chapters 12 – 15 of namat 4 of part l-tanbı¯ha¯t. More ˙ II18. 2. At the end of chapter 7, two passages of Bahmanya¯r’s Tahs¯ıl, i. e., one ˙˙ (Baya¯n, pp. 316,10 – 317,9) covering III, 1, 1, pp. 576,12 – 577,16 (on absolute perfection), another (Baya¯n, pp. 317,10 – 320,2) covering II, 6, 3, pp. 559,10 – 561,14 (on pain and pleasure, sensible and intellectual), have been added19. They seem to have been inspired by Sˇifa¯’, Ila¯hiyya¯t, VIII, 7, pp. 368,13 – 369,10. 3. Chapter 9 opens with a proof that a celestial motion cannot be by force. This proof is explicitly linked with Aristotle’s On the Heavens, designated by its common Arabic title (fı¯) al-Sama¯’ wa-l-‘a¯lam (On Heaven and Earth). The passage (Baya¯n, p. 333,5 – 14) is not a literal quotation, in spite of its opening word qa¯la (‘he has said’), but offers a paraphrase of a fragment of the latter 17 The quotation of Ibn Sı¯na¯, Sˇifa¯’, Ila¯hiyya¯t, VIII, 1 ends at Baya¯n, p. 272,2 with the reproduction of line 2 of p. 329 and reopens at p. 273,2 (starting with fa-qad) with that of line 3 of the very same page (not at p. 273, 4 with line 7, as indicated by al-Dı¯ba¯g˘¯ı, see al-Lawkarı¯, Baya¯n, p. 273, n. 4). 18 The former of these chapters is characterized as a ˇsarh, the remaining ones are designated ˘ as isˇa¯ra, see Ibn Sı¯na¯, al-Isˇa¯rat, III, pp. 23 – 7. 19 Al-Dı¯ba¯g˘¯ı seems to have forgotten to indicate the beginnings of this second fragment on p. 310,10 of his edition of al-Lawkarı¯’s Baya¯n (at p. 320, n. 5 he remarks: ‘this is the end of what has been transmitted from the Tahs¯ıl ’, but this cannot be the end of the former ˙˙ fragment, the only one to which he is referring to on p. 316, n. 14).

Al-Lawkarı¯’s Reception of Ibn Sı¯na¯’s Ila¯hiyya¯t 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

13

work, i. e., I, 2, 269a9 – 18. Al-Lawkarı¯ agrees with this, but adds that Ibn Sı¯na¯ has surpassed the Philosopher in adding a proof that such motion is not natural either. Somewhat later in the chapter (pp. 335,9 – 336,4), al-Lawkarı¯ introduces a saying of (pseudo-)Ptolemy’s Kita¯b al-Tamara (Book of the Result), which states that there is no difference between what¯ chooses the best and the natural. The saying is followed by an explanation due to a certain Abu¯ al-‘Abba¯s Ahmad ibn ‘Alı¯ al-Isfaha¯nı¯20. In doing this, al-Lawkarı¯ wants to stress that Ibn Sı¯na¯˙was right ˙ of refuting radically any natural circular motion – to accept that – in spite somehow circular motion may be called natural, namely insofar as what moves a body in a circular motion is not alien to that body (Sˇifa¯’, Ila¯hiyya¯t, IX, 2, p. 383,1 – 3). 4. The second part of chapter 17 corresponds to pages 362,8 – 364,9 of the Baya¯n. The first part (p. 362,1 – 8) shows affinities with Sˇifa¯’, Ila¯hiyya¯t, IX, 5, p. 413,1 – 7, but stems ultimately from another source as immediately afterwards will be indicated. All the remaining chapters of the part on lordly things, except for two of them, have their ultimate source in another Avicennian text, i. e., his Kita¯b alMabda’ wa-l-ma‘a¯d: Baya¯n c. 10 – 13 c. 16 c. 17 (1. part) c. 18 – 19 c. 23 – 7

al-Mabda’ wa-l-ma‘a¯d II, 1 – 4 (emanation from the One; ibda¯‘; first caused is one and is intellect; multiplicity out of first intellect) II, 5 (generation of what is beneath the spheres) II, 6 (evocation of a certain theory on the generation of elements) II, 7 – 8 (providence, especially regarding beings of generation and corruption) III, 16 – 20 (prophecy, and related issues).

Very close resemblances with the Ila¯hiyya¯t of the Sˇifa¯’ are present in chapters 10 (pp. 339,4 – 340,4; Sˇifa¯’, Ila¯hiyya¯t, IX, 3, pp. 402,6 – 403,13); 12 (pp. 345,3 – 346,5; Sˇifa¯’, Ila¯hiyya¯t, IX, 4, pp. 403,13 – 404,8), 13 (pp. 347,6 – 353,13; Sˇifa¯’, Ila¯hiyya¯t, IX, 4, 405,10 – 409,20), 16 (pp. 359,3 – 361,6; Sˇifa¯’, Ila¯hiyya¯t, IX, 5, pp. 410,4 – 12 and 411,16 – 412,15) and 17, as has already been noted21. With respect to chapters 23 – 7, a doctrinal similarity shows up – at least, in a broad sense – with Sˇifa¯’, Ila¯hiyya¯t, X, 1 – 2, and this in spite of the absence of any literal correspondence. It looks as if al-Lawkarı¯ has opted for the version of the Kita¯b al-Mabda’ wa l-ma‘a¯d because of its being more detailed (e. g., the three 20 I was unable to identify this scholar. 21 Dı¯ba¯g˘¯ı (see al-Lawkarı¯, Baya¯n, p. 343, n. 1; p. 365, n. 2 and p. 391, n. 2) refers also – in my view in an unjustified way – for the chapters 11, 18 and 23 to parallels in the Ila¯hiyya¯t of the Sˇifa¯’.

14 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Jules Janssens

properties of prophecy are presented in an elaborated way, not just briefly referred to, as was the case in Sˇifa¯’, Ila¯hiyya¯t, X, 1, p. 435,14 – 15). In a similar vein, one can explain the use of the Kita¯b al-Mabda’ wa-l-ma‘a¯d in chapters 18 – 9 as a more detailed exposition of the definition of providence given at the beginnings (pp. 414,17 – 415,7) of Sˇifa¯’, Ila¯hiyya¯t, IX, 6 (including a specification with regard to corruptible beings). With respect to chapter 15, I looked in vain for a precise source. The chapter contains a question: ‘does the soul’s desire cause only the motion of the spheres, nothing else?’, and offers a mistaken answer (‘motion originates in the spheres when its soul desires to assimilate with the intellect in perfection, in the same way as it originates in us, humans, when we desire our beloved, hence as a kind of upsurge’) as well as the correct one (‘the soul’s desire to imitate the intellect makes the motion the perfection itself, insofar as it brings into act what was in potency in a continuous, eternal manner’). One easily detects elements of resemblance with Sˇifa¯’, Ila¯hiyya¯t, IX, 2, pp. 389,4 – 392,6, but there is no wordfor-word correspondence. The formulation might therefore be al-Lawkarı¯’s own, although I do not exclude that it copies a fragment of still another – Avicennian or Avicennian inspired – work. The probability that this is the case significantly increases as soon as one realizes that a passage of Ibn Sı¯na¯’s Commentary on the Theology constitutes the source of chapter 22. It has to be noted that al-Lawkarı¯, contrary to his usual habits, avows (Baya¯n, p. 388,3 – 4) that he is quoting Ibn Sı¯na¯, and more precisely the latter’s Kita¯b al-Insa¯f. As is well known, this work had been lost during Ibn Sı¯na¯’s own lifetime, ˙but fragments of it survived, among which the Commentary on the Theology22. Al-Lawkarı¯ quotes the beginning of Ibn Sı¯na¯’s commentary on Mı¯mar 2, where it is stressed that the soul, after its separation from the body, understands all things, both universals and particulars, all at once, since it grasps the particulars from their causes, i. e., as universals23. This way of intellection in a universal way is strongly stressed and sharply contrasted with the bodily mediated perception of (material) particular things in the ten last lines of the chapter (pp. 389,14 – 390,7), which, as far as I can see, are not present in the edited version of Badawı¯24. Whatever be the case, they are fully 22 With respect to the Kita¯b al-Insa¯f, see for more details Gutas, Avicenna, pp. 130 – 40. ˙ Regarding the Theology, I may refer to Adamson, Arabic Plotinus, passim. ˇ 23 See Ibn Sı¯na¯, Sarh Kita¯b Utu¯lu¯g˘iya¯, pp. 35 – 74 (47,8 – 49,2). Despite the explicit ˙ ¯ b al-Insa¯¯f, al-Dı¯ba¯g˘¯ı has not identified any specific passage as a reference to the Kita ˙ is the affirmation by Marcotte, Life and Work of alsource. Even more surprisingly Lawkarı¯, p. 146 that ‘it is unclear what may have been Lawkarı¯’s source for the passage from al-Insa¯f to which he refers and quotes (in the Metaphysics) on the posthumous life ˙ of the soul’. 24 Although one cannot a priori exclude the possibility that al-Lawkarı¯ has himself added these lines, it looks nevertheless more probable that he quotes a longer version than the

Al-Lawkarı¯’s Reception of Ibn Sı¯na¯’s Ila¯hiyya¯t 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

15

congruent with what precedes. They reflect moreover a genuinely Avicennian spirit. In this respect, one may point to the Sˇayh al-ra’ı¯s’ Risa¯la l-adhawiyya fı¯ l˙˙ ˘ ma‘a¯d, where the universal character of the perception of the rational soul, as well as the specification of the object of its perception in terms of the stable intentions (al-ma‘a¯nı¯ l-tha¯bita) are both clearly expressed25. However, there remains one fundamental question: why has al-Lawkarı¯ inserted this passage of the Commentary on the Theology? As far as I can see, he considered it in all likelihood a natural complement to the preceding chapter – and hence to Sˇifa¯’, Ila¯hiyya¯t, IX, 7. Indeed, the passage specifies the true happiness of the fully accomplished soul in the hereafter. But, more importantly, he himself stresses in the introduction to the section on divine science (pp. 3 – 4) that he wants to limit his exposition to the basic principles, except for one topic, i. e., the nature of the knowledge of the soul in the hereafter. Al-Lawkarı¯ emphasizes that he does this because the usual manuals do not pay any attention to this topic (in fact, they are always restricted to the issues of punishment and reward, pain and joy). But this does not mean that he includes among these manuals Ibn Sı¯na¯’s works. On the contrary, al-Lawkarı¯ is aware that the latter, in his writings, has dealt with it in a significant manner and, hence, makes clear that he owes this idea to his ultimate master by the explicit mention of the latter’s name, followed by a formula of great blessing: ‘May the spirit of God cover him and sanctify his soul (or: him)’ (‘ru¯h Alla¯h ramasahu wa-qaddusa nafasahu’; p. 388,3). ˙ Surveying both basic parts, one must admit that there is no real originality in al-Lawkarı¯’s work. It is more than just inspired by the Ila¯hiyya¯t of the Sˇifa¯’; in fact, it quotes verbatim many chapters, or parts of them, of the latter. On occasion, it replaces some of them by other fragments, taken from elsewhere in Ibn Sı¯na¯’s œuvre, or in the Kita¯b al-Tahs¯ıl of the latter’s disciple, Bahmanya¯r – ˙˙ once again by way of literal copying. But even then, there is no appearance of a single rupture with the exposition of the Metaphysics of Ibn Sı¯na¯’s major encyclopedia. It is therefore surprising that al-Lawkarı¯ (p. 3) qualifies this part of his work as being both in the way of a talh¯ıs, concise exposition, and a ˇsarh, ˙ ˙ commentary, since there is no real additional˘ input of himself. As indicated at the beginning, the division of the divine science into ‘universal science’ and ‘lordly science’ might reveal a personal accent, although possibly still inspired by Ibn Sı¯na¯. It is interesting to see how al-Lawkarı¯, in his introduction (p. 4), articulates this division. First of all, he specifies that the ‘universal science’ deals with the principles of all the sciences. As to the other part, it exposes the intentions of the Lordly Book, called Utu¯lu¯g˘iya¯ (Theology), – clearly a reference ¯ to the famous pseudepigrapic work called Theologia Aristotelis. A little later, the one edited by Badawı¯. It might be present in the so-called longer version, which till now has not yet been edited. 25 See Ibn Sı¯na¯, Risa¯la l-adhawiyya fı¯ l-ma‘a¯d, p. 197,5 – 9. ˙˙

16 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Jules Janssens

proper object of this part is more precisely articulated as the establishment of the First Principle and Its attributes, and of the spiritual, angelic separated beings. In between, al-Lawkarı¯ has noted that the name ‘meta-physics’ (ma¯ ba‘d at-tabı¯‘a) applies in common understanding to both parts. In the introduction to˙ ˙the part on the lordly things (p. 265), he repeats that the lordly science is usually considered as a part of metaphysics, but now he adds that this is not out of necessity, but for the sake of brevity (li-l-ihtisa¯r). In this sense, the ˘ ˙ establishment of the First Principle is part of the science of meta-physics, the subject of which is ‘being qua being’ and which therefore can be called ‘general divine science’ (al-‘ilm al-ila¯hı¯ al-‘a¯mm). From all this, one gets the impression that al-Lawkarı¯ does not doubt the authenticity of the Theologia, and, moreover, seems to consider the inclusion of a theology in the Metaphysics, i. e., in lambda, as more a matter of convenience than of necessity. This might appear to sharply contradict Ibn Sı¯na¯’s basic dmarche of an integrated project, as present in the Ila¯hiyya¯t of the Sˇifa¯’, but this is not necessarily the case, insofar as Ibn Sı¯na¯ elsewhere, as indicated above, may have open the way to this kind of loosened link between metaphysics proper and theology. Al-Lawkarı¯, in the first introduction (p. 4), mentions also that he will attach to the section of the Lordly Things selected aphorisms (fusu¯l muntaza‘a) on ethics, especially on the virtues of the soul and making firm˙ the perfect state, and that he will conclude the book with questions and remarks (masa¯’il wanukat) regarding the secrets of meta-physics, i. e., the divine attributes, the agent intellects and the celestial souls. It strikes one immediately that the addition of an ethical section fully corresponds with the structure of the Ila¯hiyya¯t of the Sˇifa¯’, where X, 4 – 5 treat of the morals of the city, the household and the individual, although very briefly. But has al-Lawkarı¯ effectively quoted these chapters? Al-Dı¯ba¯g˘¯ı has not included the above-mentioned additional parts in his edition. They have been preserved in a single manuscript, i. e., Tehran University, Central Library 250 (= 108)26. Unfortunately, this manuscript is frequently damaged in the margins, especially near the end of the text. In spite of this damage, I succeeded in identifying the source of the vast majority of the passages. Before presenting the results of my research, I want to stress that the actual order is not that indicated by al-Lawkarı¯ in his introduction. In fact, one first finds the appendix regarding the secrets of metaphysics; it is divided into 26 Many thanks to Frank Griffel and Meryem Sebti, who each kindly provided me with a separate CD-Rom of the manuscript, thus giving me two scanned versions. For the analysis of the appendices, I have used both CD-Roms, since in Griffel’s copy folios are lacking, whereas in that of Sebti the order of the different parts of the work is changed. Moreover, there is no numbering on the folios, but based on the information given by Nag˘g˘a¯r in al-Fa¯ra¯bı¯, Fusu¯l muntaza‘a, p. 18, I fixed the folio numbering. However, even ˙ if it would be found incorrect, I am confident that the indications I offer will permit the reader to identify beyond any doubt the passage in the manuscript.

Al-Lawkarı¯’s Reception of Ibn Sı¯na¯’s Ila¯hiyya¯t 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

17

four chapters (abwa¯b): the attributes of the Creator, intellects, souls, sanctity and prophecy (wila¯ya wa-nubuwwa) respectively. Only afterwards does the ethical section follow. This, in its turn, contains five chapters, i. e., health and sickness of soul; social relations (mu‘a¯ˇsira); politics (siya¯sa); theoretical and practical intellect; and classes of the virtuous city27. Let us now concentrate on the first appendix. Its first chapter (fols 212rv 214 ) starts with a rather long discursus on the divine attributes of knowledge, will, providence, power, wisdom and liberality. It is a literal reproduction of the discursus on these subjects in the Ta‘lı¯qa¯t, which offers what I have characterized as the Arabic original of the corresponding chapters in the Da¯nesˇ-Na¯meh28. The three following subdivisions (fusu¯l), while focusing respectively on divine ˙ goodness, on the essential (not temporal) priority of God towards His action, and on the divine Light, reproduce three fragments derived from Ibn Sı¯na¯’s Commentary on the Theology29. Finally, three subdivisions have their source in Ibn Sı¯na¯’s Muba¯hata¯t30 : the first deals with God’s being necessary (wa¯g˘ibiyya), ˙ ¯ the second insists that from the One only one can follow (yalzimu) and the third denies any multiplication in the divine essence. The divine tawh¯ıd, unity and ˙ unicity, plays a central role in this chapter. In spite of a possible (logical) distinction between several attributes in the divine essence, and in spite of God’s acting resulting in a creation outside Him, He has a unique, indivisible fullness of being, more precisely of being necessary. This idea is certainly present in the Ila¯hiyya¯t of the Sˇifa¯’, but is now stressed in a more pronounced way, based on texts of the Sˇayh al-ra’ı¯s himself! The second ˘chapter (fols 214v-216r) contains additional information on the higher intellects. Its first two subdivisions relate to the highest of them. This latter is designated as the simple intellect (al-‘aql al-bası¯t) and both its essence ˙ and its action are discussed, once again based on the Muba¯hata¯t (pp. 301 – 2, ˙ ¯ § 844, respectively p. 302, §§ 845,1 – 2 and 846). As to the third subdivision, which contains five proofs (i. e., necessity of essential unity of first emanated being; impossibility of existence of celestial bodies; essential separate nature of human soul; actualisation of human soul; eternal motion of universe), in order to justify the existence of the agent intellects, it copies verbatim an entire chapter 27 The list of these headings is present in al-Lawkarı¯, Baya¯n. Mantiq. al-Madhal, pp. 90 – ˙ ˘ 91. 28 See Ibn Sı¯na¯, Ta‘lı¯qa¯t, pp. 13,4 – 22,9. Regarding the link with the Da¯nesˇ-Na¯meh, see Janssens, Le Da¯nesh-Na¯meh, pp. 163 – 5. 29 See Ibn Sı¯na¯, Sˇarh Kita¯b Utu¯lu¯g˘iya¯, pp. 46,4 – 15; 47,1 – 5 and 56,14 – 57,8. ˙ ¯ hata¯t,¯ respectively pp. 140 – 41, §§ 386 – 90; p. 226, § 674; 30 See Ibn Sı¯na¯, Muba ˙ ¯ § 261,1 – 3 (al-Lawkarı¯ introduces the two last fragments with p. 271 – 2, § 787; p. 112, aydan, ‘also’) and p. 366, § 1141. Unless otherwise indicated, page and paragraph ˙ references are in what follows always to the edition by Bı¯da¯rfar.

18

of the Risa¯la mara¯tib al-mawg˘u¯da¯t, also known as Risa¯la fı¯ itba¯t al-mufa¯raqa¯t31. ¯ As to the fourth subdivision, it specifies that the higher intelligences are only able to grasp the First Principle thanks to a divine illumination (tag˘allı¯) by quoting the Commentary on the Theology (pp. 49,6 – 50,10). The fifth subdivision emphasizes that these intellects do not act in view of what is beneath them and reproduces Ta‘lı¯qa¯t (p. 49,16 – 18 and 11 – 12). Their possible nature is examined in the two following subdivisions, which have their source in the same work (p. 52,9 – 10, respectively p. 54,7 – 14)32. Eventually, the unavoidable presence of a multiplicity in their intellection is affirmed, once more based on the Ta‘lı¯qa¯t (p. 62,20 – 24). In this chapter, the mediating – but indispensable – role of the higher intellects occupies a central place. Also this time, the idea itself is not foreign to the Ila¯hiyya¯t of the Sˇifa¯’, and, again, one has to do with a clarification expressed in Ibn Sı¯na¯’s own words. As to the third chapter (fol. 216r-v), it concerns the soul. The Ta‘lı¯qa¯t forms its unique source. Only the first subdivision pays attention to the celestial souls and their role in the different circular motions of the spheres (Ta‘lı¯qa¯t, p. 54,18 – 25). The seven remaining subdivisions all deal with the human soul: the identity in it between active and final cause; the fact that the goal of its motion is nothing outside itself; that its perception is not in view of what is perceived; that the perfection of the vegetative soul is not the real end of man; the goal-directedness of the soul as basis for the difference between voluntary and natural motions; perception (idra¯k), not sensation, as activity proper to the soul; and the impossibility for an embodied soul to perceive itself directly as a separate being (Ta‘lı¯qa¯t, pp. 63,20 – 8; 63,3 – 7; 63,8 – 9; 63,10 – 19; 53,20 – 25; 23,1 – 19 and 23,23 – 8). Except for the discussion of the celestial souls, the metaphysical relevance of this chapter is less evident. Nevertheless, one can make a link between the emphasis on the perfection of the human soul in its quality as separate substance, on the one hand, and the evocation of the ma‘a¯d in Sˇifa¯’, Ila¯hiyya¯t, X, 3, on the other. Regarding the fourth chapter (216v–218r), it opens with the idea that a human soul has to prepare itself adequately in order to achieve its perfection, reproducing once again Ta‘lı¯qa¯t (p. 37,22 – 4). The second subdivision is rather long. It spotlights the pure soul (al-nafs al-zakiyya), the significance of prayer

˘

31 Not only the title, but also the authorship of the treatise is doubtful: although other names are given as well, the best candidates for the latter are undoubtedly Bahmanya¯r and al-Fa¯ra¯bı¯. This problem certainly deserves a profound analysis that cannot be offered in the present paper. Let me simply note that al-Lawkarı¯ copies c. 3, which corresponds to Bahmanya¯r (?), Risa¯la mara¯tib al-mawg˘u¯da¯t, pp. 63,8 – 64,15. 32 In the former of the two fragments, al-Lawkarı¯ adds to the affirmation of the Ta lı¯qa¯t that this possibility is not like that of the generable beings. This addition was maybe already present in the copy of the Ta lı¯qa¯t he had at his disposal. A systematic study of this latter work, and its different redactions, remains a major desideratum. ˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Jules Janssens

Al-Lawkarı¯’s Reception of Ibn Sı¯na¯’s Ila¯hiyya¯t

and extraordinary events such as magic, talismans and miracles. It copies Ta‘lı¯qa¯t (pp. 47,20 – 48,12), but before quoting the final sentence (lines 10 – 12) introduces a large section of Namat 10 of part II of the Isˇa¯ra¯t33. The subdivisions 3 – 6 copy once more ˙ verbatim passages from the Ta‘lı¯qa¯t (pp. 69,25 – 70,3; 77,3 – 5; 79,27 – 80,17; 82,22 – 5; 81,18 – 22; 81,26 – 8; 82,1 – 5; 82,17 – 18; and 193,25 – 194,5). Here, the soul is presented as a stable entity, as having a separate essence and as being self-perceptive. It is affirmed, moreover, that the body is a condition only with respect to the existence, not the survival of the soul; that the soul cannot reach anything of the Malaku¯t, i. e., the intelligible celestial world, unless it is entirely spiritual; and that the simple intellectual representation (al-tasawwur al-bası¯t al-‘aqlı¯), which is a gift of the Giver of Forms (Wa¯hib al-suwa¯˙r), brings our ˙intellects from potency into act. ˙ The latest of these affirmations is supplemented with two additional remarks: intelligible things are devoid of change, and hence of individuality in the way sensible ones are; human beings cannot know intelligible things unless through a conjunction with the agent intellect. Regarding both, I looked in vain for a source. However, these additions might have been present in the copy of the Ta lı¯qa¯t he had at his disposal34. In the following subdivisions, i. e., 7 – 11, alLawkarı¯ extensively deals with the (self-)knowledge of the soul, paying special attention to its way of understanding after the separation of the body and to the way the soul links with the agent intellect. This time, he combines different passages taken from the Muba¯hata¯t (pp. 155 – 6, §§ 427 – 8; p. 316, § 888; ˙ ¯ § 893; and p. 318, § 892). The last pp. 87 – 8, §§ 150 – 5435 ; p. 318, subdivision insists that the perfect human soul enjoys after its separation from the body a purely intellectual life; it offers a literal quotation of the final part of the Risa¯la mara¯tib al-mawg˘u¯da¯t36. All in all, the chapter concentrates on the human soul and its ultimate perfection. Given its title, i. e., On Sanctity and Prophecy, this is somewhat surprising. Certainly the passage, which mentions a reaching of the Malaku¯t, deals with the mode of the prophet’s receiving revelation37, but generally speaking almost no attention is paid to the specific issue of prophecy (or sanctity). On the sole basis of the title, one might suspect that the chapter was meant as somehow supplementing Sˇifa¯’, Ila¯hiyya¯t, X, 1 – 2. Based on its actual wording, this is, however, far from being evident. Let us now turn to the second section, on ethics. It was already noticed by Fawzı¯ M. Nag˘g˘a¯r that this section in our manuscript corresponds to the Fusu¯l ˙ ˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

19

33 Ibn Sı¯na¯, al-Isˇa¯rat, IV, pp. 153,2 – 159,6. 34 See above, n. 32. 35 The beginning of the quotation corresponds to the version of the fragment as published by Badawı¯, p. 227, § 457. 36 See Bahmanya¯r (?), Risa¯la mara¯tib al-mawg˘u¯da¯t, pp. 65,19 – 66,7: the actual wording of the text corresponds to the one given as variant p. 65, n. 14. 37 See Michot, Destine de l’homme, p. 127.

20 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Jules Janssens

muntaza‘a, i. e. §§ 1 – 61 (except §§ 3, 15, 23 and 40) commonly attributed to al-Fa¯ra¯bı¯38. Nag˘g˘a¯r also observed that a large passage (from fol. 219r to the beginning – in fact line 3 – of fol. 219v) has no direct counterpart in the actual text of the Fusu¯l39. However, he mentions another Iranian manuscript, i. e., ˙ Tehran University, Faculty of Theology, 695d, which contains the part of the Fusu¯l preserved in al-Lawkarı¯’s Baya¯n. It omits the same paragraphs and also has ˙ a large – even larger than in our manuscript – section that does not derive from the Fusu¯l 40. Also more important is the fact that it does not stop, as does our ˙ manuscript, in the midst of § 61 (p. 70,10), but at § 62 (still incomplete, since the last words correspond to p. 73,2). In none of the two manuscripts does there seem to be a closing formula, so that one suspects that al-Lawkarı¯’s original text was longer and probably included still further aphorisms41. Anyway, for our present study the way in which he introduces this ethical section is more worthy of attention. Indeed, al-Lawkarı¯ affirms42 : These are the sentences and aphorisms chosen from the science of morals. They comprise: acquiring the virtues of the human soul, avoiding its vices; moving the human being from his bad habits to fine habits; making firm the virtuous city; making firm the household, i. e., the rulership over its members. They are all brought together in this note [in five chapters].

It is obvious that this is not a neutral statement. On the contrary, an outspoken interest in human virtue manifests itself. Certainly, attention is also paid to the 38 See al-Fa¯ra¯bı¯, Fusu¯l muntaza‘a, p. 18, where it is given as one of the manuscripts used to establish the text.˙ It has to be noted that the omitted paragraphs are exactly the same as the ones not present in Dunlop’s edition (see al-Fa¯ra¯bı¯, Fusu¯l al-madanı¯). It should be ˙ to §§ 6 and 26 is exactly noted moreover that the wording of the passages corresponding the same as in Dunlop’s edition, hence omitting the last sentence in both cases. 39 In fact, the last two lines of fol. 218v are already not present in the Fusu¯l, but they most naturally continue the discursus on the nutriment of the body. This˙ suggests that alLawkarı¯ had at his disposal another version than the one actually available to us. It has to be noted moreover that this addition shows certain similarities with § 10 of the Fusu¯l. Nag˘g˘a¯r obviously forgot to mention that al-Lawkarı¯’s text, in turn, does not quote ˙the paragraphs 8 – 11 of the Fusu¯l. ˙ to this manuscript and Nag˘g˘a¯r’s information remains very 40 Unfortunately, I had no access limited in this respect (see al-Fa¯ra¯bı¯, Fusu¯l muntaza‘a, pp. 18 – 19). ˙ one to decide up to which point al-Lawkarı¯ 41 The actual state of affairs does not permit quoted the Fusu¯l. In MS Tehran University, Central Library 250 (= 108), it ends in the ˙ whereas in MS Tehran University, Faculty of Theology, 695d, it ends in middle of § 61, the middle of § 62 (see Nag˘g˘a¯r’s introduction to al-Fa¯ra¯bı¯, Fusu¯l muntaza‘a, pp. 18 – 19). ˙ In view of the title of the section one might guess that al-Lawkarı ¯ continued till § 67, and not further, since afterwards other issues come to the fore. 42 For the Arabic text, see al-Fa¯ra¯bı¯, Fusu¯l muntaza‘a, pp. 18 – 19, as well as p. 23, n. 2; I ˙ modified) of Butterworth (al-Fa¯ra¯bı¯, Political quote the English translation (slightly Writings, p. 6). Note that the addition ‘in five chapters’ occurs only in MS Tehran University, Faculty of Theology, 695d.

Al-Lawkarı¯’s Reception of Ibn Sı¯na¯’s Ila¯hiyya¯t 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

21

social and political order, but this is not presented as the exclusive or primary object of the exposition that will follow. In this respect, it is most relevant that the long addition mainly deals with the issue of the possibility of changing one’s moral conduct, i. e., one’s habits. At the end, it is even stressed that the governance of the soul (read: of oneself ) has precedence over all other kinds of governance and that one has to make a jiha¯d in order to perfect one’s habits. Before, a brief remark has been offered on the phenomenon of magic and other extraordinary arts. In my view, this fits better an Avicennian than a Farabian perspective43. This impression is only reinforced when one looks at the division into chapters. The first (fols 218r–220r), containing fifteen subdivision (or aphorisms?), discusses the health and sickness of the soul in parallelism with the health and sickness of the body. The use of a medical metaphor is rather unusual – although not completely lacking44 – in al-Fa¯ra¯bı¯, while it is quite natural in Ibn Sı¯na¯. Suffice to say that the latter calls his major philosophical encyclopedia Book of Healing – his ‘healing’ being that of the soul. In the second chapter (fols 220r–221r), which has seven parts, the medical metaphor is maintained in the explanation of the social relations. One detects moreover a strong emphasis on the necessity of being virtuous for each individual person. As to the third (fol. 221r), it briefly – in only three subdivisions – deals with the notion of malik, king. Unsurprisingly, the kingly craft is compared to the medical craft of the physician. It may be noted that the largest of the three subdivisions offers a survey of different historical opinions about the goal intended in kingship. This kind of historical doxography is once again more typical of Ibn Sı¯na¯ than of alFa¯ra¯bı¯45. The fourth chapter (fols 221r–222v) deals with both theoretical and practical intellection and is divided into no less than twenty subdivisions46. In this section, there is no room for any medical metaphor, but it is striking that in one passage (fol. 222r, lines 3 – 8; Fusu¯l, § 41) the idea is evoked that a sick ˙ person imagine what is sweet bitter, and vice-versa, an idea one also encounters 47 in Ibn Sı¯na¯’s Kita¯b al-Nafs of the Sˇifa¯’ . Regarding chapter five, which might be 43 Regarding the primacy of the governance of one own’s soul, see e. g. Ibn Sı¯na¯, Fı¯ l-siya¯sa l-manziliyya, pp. 232 – 60 (240). With respect to magic and other occult phenomena, see for example Ibn Sı¯na¯, al-Isˇa¯ra¯t, IV, pp. 158 – 9, and compare moreover above, p. 19. The same basic idea is also present in the Kita¯b al-nafs of the Sˇifa¯’ (especially IV, 4). 44 See al-Fa¯ra¯bı¯, Ihsa¯’ al-‘ulu¯m, pp. 67 – 76 (71), where the need of the king in his political ˙˙ practice for experience is compared to that of the physician in his medical practice; see Janssens, Experience, pp. 45 – 62 (49). Charles Burnett kindly informed me that this kind of metaphor is the subject of a Ph.D. thesis by Badr el Fekkak. 45 See Janssens, Ibn Sı¯na¯, pp. 83 – 93. 46 Due to heavy damage, the title of the chapter is not readable, but in view of the space seems to have been ta‘aqqul, ‘intellection’. 47 See Ibn Sı¯na¯, ‘De Anima’, p. 62,16 – 17.

22 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Jules Janssens

incomplete as indicated above, it might be worthwhile to note that it seems to involve in its second subdivision (Fusu¯l, § 57) a conception of jiha¯d that ˙ substantially differs from that of al-Fa¯ra¯bı¯’s al-Madı¯na al-fa¯dila48. ˙ Certainly, none of these elements definitely excludes a Farabian origin. However, if the work is authentically Farabian, then Ibn Sı¯na¯ has largely used it in his early work on ethics, i. e., al-Birr wa-l-itm (Piety and Sin)49. Indeed, a part ¯ of the Risa¯la l-birr wa-l-itm, which seems to be a part of the otherwise lost work, ¯ shows a very close similarity with Fusu¯l, §§ 1 – 19, respectively 33 – 56, while ˙ omitting as al-Lawkarı¯ §§ 3, 15 and 4050. If this were correct, the young Ibn Sı¯na¯ would have been profoundly indebted to his great predecessor. However, one may wonder then why he has chosen a rather small, and, moreover, not very typical work of the Second Master? Or did there exist a nucleus of aphorisms assembled by al-Fa¯ra¯bı¯ and has Ibn Sı¯na¯ enlarged it? But the attribution of the Fusu¯l to al-Fa¯ra¯bı¯ may be mistaken, since it is, after all, only explicitly present ˙ with the title in one single manuscript, namely Bodleian, Hunt 30751. A thorough examination is needed in order to settle this complex issue. It clearly exceeds the limits of the present paper. But whatever hypothesis one favours, it is clear that for al-Lawkarı¯ the Fusu¯l was sufficiently Avicennian in contents to ˙ of Avicennian texts. Even more strikingly be included in a kind of florilegium Avicennian is his decision to make this ethical tract an appendix of Divine Science. This corresponds to the fact that Ibn Sı¯na¯ himself, in his al-Birr wa-litm, probably treated ethics in the context of what D. Gutas has labelled ¯ ‘metaphysics of the rational soul’52. Looking over the whole, one has to admit that al-Lawkarı¯ gives an overwhelming impression of a desire to reproduce to the utmost possible degree an encompassing synthetic survey of Ibn Sı¯na¯’s Ila¯hiyya¯t. Still, one may remain somewhat hesitant, insofar as al-Lawkarı¯, contrary to Ibn Sı¯na¯ in the Sˇifa¯’, seems to dissolve the unity of metaphysics in favour of a sharp bipartition. The intimate link between universal Science and theology appears to have been dramatically loosened. But this might have already happened in Ibn Sı¯na¯, as indicated at the beginning53. Moreover, one must not forget that the latter, in 48 See Dunlop’s introduction to al-Fa¯ra¯bı¯, Fusu¯l al-madanı¯, p. 13. ˙ work when he was still in Buha¯ra¯, just after 49 According to his autobiography, he wrote the having finished at the age of twenty-one the Compilation for ‘Aru¯d¯ı (see ˘Ibn Sı¯na¯, The ˙ Life of Ibn Sina, pp. 38 – 41) This information permits one to date al-Birr wa l-itm ¯ around the year 1000. 50 See Ibn Sı¯na¯, Risa¯la l-birr wa l-itm, pp. 360 – 68. It has to be observed that the wording ¯ in this Risa¯la is rather of a summarizing nature, but that a close link between both texts is beyond any doubt. I hope to publish later a more detailed comparison. 51 See al-Fa¯ra¯bı¯, Political Writings, p. 5. 52 See Gutas, Avicenna, pp. 95 and 254 – 61. 53 See above, p. 7.

Al-Lawkarı¯’s Reception of Ibn Sı¯na¯’s Ila¯hiyya¯t 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

23

the general prologue to the work, has characterized the Sˇifa¯’ as ‘accommodating his Peripatetic colleagues’54. On the other hand, al-Lawkarı¯’s work clearly lacks originality. Most of the time, not to say always, he copies verbatim large passages, and even entire chapters of different works of Ibn Sı¯na¯, supplemented with fragments taken from the Kita¯b al-Tahs¯ıl of the latter’s immediate disciple, Bahmanya¯r and of ˙˙ be a work of the latter’s master, al-Farabı¯ (if it is the Fusu¯l, which might ¯¯ ˙ not an Avicennian text, which, in my view, in the actual state of definitely affairs, cannot be proven). Anyhow, in all cases, the quotations are so literal that al-Lawkarı¯’s text may be used as an independent testimony, besides available manuscripts, for the establishment of the critical edition of the respective works. In this respect, it is worthwhile to note that the oldest known manuscript of the Baya¯n, i. e., Tehran University, Central Library 250 (= 108), is explicitly dated 601 H. (beginnings twelfth century) and is hence very old55. A first rapid survey has permitted me to detect a large number of interesting variants in comparison with the existing editions of the works in question. Of course, a thorough investigation is needed for a final judgement. But just to show how really interesting some of these variants are, I will quote three cases by way of example. They all are related to the Ila¯hiyya¯t of the Sˇifa¯’. The first occurs at p. 171. Having affirmed that the relationships in irrational roots and in numerical relations are easily accessible to the soul, Ibn Sı¯na¯ continues to state that ‘it does not follow that the soul in one state would intellectually apprehend (an taku¯na … tu‘qila) all of these’, adding a little later ‘within its proximate power to intellectually apprehend (an tu‘qila) this’ (Sˇifa¯’, Ila¯hiyya¯t, p. 211,4 – 5). In al-Lawkarı¯ ‘intellectually apprehend’ is twice replaced by ‘do’ (an yuf‘ala). Even if the context favours rather the reading of the edition of the Ila¯hiyya¯t of the Sˇifa¯’, the variant of al-Lawkarı¯, which seems not to occur anywhere else, is not totally devoid of sense. Moreover, it sheds light on why the Latin translation has ‘agat’, or ‘agere’; a rendering that corresponds perfectly to al-Lawkarı¯’s variant, and hence might not constitute a free rendering of the Arabic verb ‘aqala (normally translated as intelligere)56. The second concerns the possibility of an intellectual knowledge of an individual entity, namely when it is unique in its species. According to the 54 Ibn Sı¯na¯, Sˇifa¯’, al-Madhal, p. 10,14. ˘ 55 It is almost as old as several of the rather old known manuscripts of the Ila¯hiyya¯t of the Sˇifa¯’, see Bertolacci, Reception of Aristotle’s ‘Metaphysics’, p. 486, although approximately a century later than the oldest one, i. e., Tehran, Malik 1085, see ibid., p. 486, n. 21. 56 See Van Riet (Ibn Sı¯na¯, Liber de philosophia prima, V-X, p. 243,58 – 9). Van Riet limits to indicate in the second apparatus (Latin-Arabic) that agat/agere is here to be understood in the sense of intelligat/intelligere, suggesting that the Latin translator has used a not very literal, but nevertheless acceptable translation of the Arabic verb ‘aqala, there being no variant present in the apparatus of the Cairo edition.

24 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Jules Janssens

Cairo-edition of the Ila¯hiyya¯t – once again without any variant attested – Ibn Sı¯na¯ affirms that ‘if the mind intellectually apprehends that species through its individual instance (bi-sˇahsihi), it will have knowledge of it’ (Sˇifa¯’, Ila¯hiyya¯t, ˘ ˙¯’s version (Baya¯n, p. 195,3), this becomes: ‘if the p. 247,1 – 2). In al-Lawkarı mind intellectually apprehends that species and represents it as individual (watasˇahissu), it will have knowledge of it’. In this case, this variant seems to be ˘ ˙˙ proper to al-Lawkarı¯, although the Latin translation ‘et eius individuum’ entirely (Philosophia prima, p. 277,12) also presupposes the presence in the Arabic (at least, in the manuscript on which it was based) of the conjunction wa-. Finally, al-Lawkarı¯ (Baya¯n, p. 261,8) confirms the reading fa¯’ilin instead of ka¯milin (Sˇifa¯’, Ila¯hiyya¯t, p. 318,12) in accordance with many other testimonies57. I am aware that I do not do full justice to the relevance of al-Lawkarı¯’s text, since it contains hundreds of variants. Its significance is undoubtedly still much higher with respect to Ibn Sı¯na¯’s Commentary on the Theology, since for that text thus far only two manuscripts are known, i. e., Cairo, Hikma 6M, and Bursa, ˙ ¯ n may help us in Hseyin C ¸ elebi, 119458. I therefore believe that the Baya editing better, and thus in understanding better a large variety of Avicennian, or related, fragments of texts. On the doctrinal level, its value is much more limited. Nevertheless, it draws our attention to the delicate way in which Ibn Sı¯na¯ seems to have conceived the relationship between the different parts of the science of metaphysics. Bibliography P. Adamson, The Arabic Plotinus. A Philosophical Study of the ‘Theology of Aristotle’, London: Duckworth, 2002. al-Bayhaqı¯, Tarı¯h hukama¯’ al-Isla¯m, ed. Mamdu¯h Hasan Muhammad, Cairo: Maktaba ˙ ˙ ˙ ˙ ˘ ¯niyya, al-taqafa al-dı 1996. ¯ al-Fa¯ra¯bı¯, Fusu¯l al-madanı¯. Aphorisms of the Statesman, transl. and ed. M. Dunlop, ˙ Cambridge University Press, 1961. Cambridge: –– , Fusu¯l muntaza‘a (Selected Aphorisms), ed. Fawzı¯ M. Nag˘g˘a¯r, Beirut: Da¯r al-Masˇriq, ˙ 1971. –– , Ihsa¯’ al-‘ulu¯m, in Kita¯b al-Milla wa-nusu¯s uhra¯, ed. M. Mahdi, 2nd edn, Beirut: Da¯r ˙ ˙ ˇriq, 1991, pp. 67 – 76. ˙ ˘ al-Mas –– , The Political Writings. ‘Selected Aphorisms’ and Other Texts, transl. and ed. Ch. Butterworth, Ithaca, New York/London: Cornell University Press, 2001. al-Lawkarı¯, Baya¯n al-haqq bi-dima¯ni al-sidq. al-‘ilm al-ila¯hı¯, ed. I. Dı¯ba¯g˘¯ı, Tehran: ˙ ˙ ISTAC, 1995. ˙ 57 For more details, see Bertolacci, Reception of Aristotle’s ‘Metaphysics’, p. 529. It is worthwhile to note that many of Bertolacci’s proposals of correction to the Cairo edition are confirmed by the Baya¯n, although not all of them. But it has to be stressed that Bertolacci has formulated his proposals of correction with due care. 58 See Michot, Recueil avicennien, p. 125.

Al-Lawkarı¯’s Reception of Ibn Sı¯na¯’s Ila¯hiyya¯t

–– , Baya¯n al-haqq bi-dima¯ni al-sidq. Mantiq. al-Madhal, ed. I. Dı¯ba¯g˘¯ı, Tehran: ˙ ˇa¯ra¯t Amı ˙ ¯r Kabı¯r,˙ 1986. ˙ ˘ Mu’assasat Intis Bahmanya¯r ibn Marzuba¯n, al-Tahs¯ıl, ed. M. Mutahharrı¯, 2nd edn, Tehran: Mu’assasa-ye ˙ ˙ intisˇa¯ra¯t ve-cha¯p da¯nesˇhah-e˙Tehran, 1375 H. ˘ Bahmanya¯r ibn Marzuba¯n (?), Risa¯la mara¯tib al-mawg˘u¯da¯t, ed. I. Java¯dı¯, Sophia Perennis, 32, 1977, pp. 56 – 73. A. Bertolacci, The Reception of Aristotle’s ‘Metaphysics’ in Avicenna’s ‘Kita¯b al-Sˇifa¯’’. A Milestone of Western Metaphysics, Leiden/Boston: Brill, 2006. D. Gutas, Avicenna and the Aristotelian Tradition, Leiden: Brill, 1988. Ibn Sı¯na¯, al-Isˇa¯ra¯t wa-l-tanbı¯ha¯t ma’a ˇsarh Nas¯ır al-Dı¯n al-Tu¯sı¯, ed. S. Dunya, 2nd edn, ˙ 4 vols, Cairo: Da¯r al-ma’a¯rif, 1968. ˙ ˙ –– , al-Mantiq al-masˇriqiyyı¯n, Cairo: Matba‘a l-mu’ayyid, 1910. ˙¯hata¯t, in Aristu¯ ‘inda l-‘Arab,˙ ed. A. Badawı¯, Cairo: Maktaba al-nahda al–– , al-Muba ¯ ˙ – 239. ˙ misriyya, ˙1947, pp. 122 ˙ ˇ –– , al-Muba¯hata¯t, ed. M. Bı¯da¯rfar, Qom: Intisa¯ra¯t Bı¯da¯r, 1992. ˙ ¯ ¯ hiyya¯t, eds G.C. Anawati, S. Zayed, M. Musa and S. Dunya, Cairo: al–– , al-Sˇifa¯’, al-Ila Hay’a al-‘a¯mma li-sˇu‘u¯n al-mata¯bi‘ al-amı¯riyya, 1960. ˙ Anawati, M. Hodeiri and F. Ahwani, Cairo: –– , al-Sˇifa¯’, al-Madhal, eds G.C. ˘ al-Mutba’a al-amı¯˘riyya, 1952. ˙¯’, al-Sama¯‘ al-tabı¯‘ı¯, ed. S. Zayed, Cairo: al-Hay’a al-misriyya al-‘a¯mma –– , al-Sˇifa ˙ ˙ li-l-kita¯b, 1983. –– , al-Ta lı¯qa¯t, ed. A. Badawı¯, Cairo: al-Hay’a al-misriyya al-‘a¯mma li-l-kita¯b, 1973. ˙ ˇ ifa¯’’, ed. F. Rahman, London/ –– , ‘De Anima’ being the Psychological Part of ‘Kita¯b al-S New York/Toronto: Oxford University Press, 1959, reprinted 1970. –– , Die Metaphysik Avicennas, transl. and ed. M. Horten, Leipzig, 1907; repr. Frankfurt am Main: Minerva, 1960. –– , Fı¯ l-siya¯sa l-manziliyya, in ‘A. Sˇams al-Dı¯n, al-Madhab al-tarbawı¯ ‘inda Ibn Sı¯na¯ ¯ al-‘a¯lamiyya lil-kita¯b, 1988, min g˘ila¯li falsafatihi al-‘amaliyyati, Beirut: al-Sˇarka pp. 232 – 60. –– , Kita¯b al-Mabda’ wa l-ma‘a¯d, ed. ‘A. Nu¯ra¯nı¯, Tehran: The Institute of Islamic Studies, McGill University, Tehran Branch, 1984. –– , La mtaphysique du Sˇifa¯’, I-V, transl. and ed. G.C. Anawati, Paris: Vrin, 1978. –– , Liber de philosophia prima sive Scientia divina, V-XS, ed. S. Van Riet, Louvain/ Leiden: Peeters/Brill, 1980. –– , Libro della guarigione. Le cose divine, transl. and ed. A. Bertolacci, Torino: UTET, 2008. –– , Metafisica. La ‘Scienza delle cose divine’ (al-Ila¯hiyya¯t) del ‘Libro della Guarigione’ (Kita¯b al-Sˇifa¯’), transl. O. Lizzini, pref. P. Porro, 2nd rev. edn, Milano: Bompiani, 2006. –– , Risa¯la l-adhawiyya fı¯ l-ma‘a¯d, ed. Fr. Lucchetta, Padova: Antenore, 1969. ˙ ˙ wa l-itm, in ‘A. Sˇams al-Dı¯n, al-Madhab al-tarbawı¯ ‘inda Ibn Sı¯na¯ min –– , Risa¯la l-birr ¯ ¯¯lamiyya lil-kita¯b, 1988, pp. 353 – g˘ila¯li falsafatihi al-‘amaliyyati, Beirut: al-Sˇarka al-‘a 68. –– , Sˇarh Kita¯b Utu¯lu¯g˘iya¯, in Aristu¯ ‘inda l-‘Arab, ed. A. Badawı¯, Cairo: Maktaba al¯ ˙ – 74. nahd˙a al-misriyya, 1947, pp. 35 ˙ ˙ –– , The Life of Ibn Sina. A Critical edition and Annotated Translation, transl. and ed. W. Gohlman, Albany, New York: State University of New York Press, 1974. –– , The Metaphysics of ‘The Healing’, transl. and ed. M. Marmura, Provo, Utah: Brigham Young University Press, 2005. ˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

25

26 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Jules Janssens

J. Janssens, Bahmanya¯r, and his Revision of Ibn Sı¯na¯’s Metaphysical Project, Medioevo, 32, 2007, pp. 99 – 117. –– , Bahmanya¯r ibn Marzuba¯n: A Faithful Disciple of Ibn Sı¯na¯ ?, in Before and After Avicenna, ed. D. Reisman with the assistance of Ahmed H. al-Rahim, Leiden: Brill, 2003, pp. 177 – 97 (now reprinted in J. Janssens, Ibn Sı¯na¯ and his Influence on the Arabic and Latin World, Aldershot: Ashgate, 2006, pp. 177 – 79.). –– , Experience (tajriba) in Classical Arabic Philosophy (al-Fa¯ra¯bı¯–Avicenna), Quaestio, 4, 2004, pp. 45 – 62. –– , Ibn Sı¯na¯ : An Important Historian of the Sciences, in Uluslararası ˙Ibn Sn Sempozyumu. Bildiriler. International Ibn Sina Symposium. Papers, II, eds M. Mazak and N. zkaya, Istanbul: ˙Istanbul Byks¸ehir Belediyesi Kltr A.S¸. Yayınları, 2008 – 9, pp. 83 – 93. –– , Le Da¯nesh-Na¯meh d’Ibn Sı¯na¯ : un texte  revoir?, Bulletin de philosophie mdivale, 28, 1986, pp. 163 – 77 (now reprinted in J. Janssens, Ibn Sı¯na¯ and his Influence on the Arabic and Latin World, Collected Studies Series 843, Aldershot, Hampshire: Ashgate, 2006, pp. 163 – 77). R. Marcotte, Preliminary Notes on the Life and Work of Abu¯ al-‘Abba¯s al-Lawkarı¯ (d. ca. 517/1123), Anaquel de Estudios rabes, 17, 2006, pp. 133 – 57. Y. Michot, La destine de l’homme selon Avicenne, Louvain: Peeters, 1986. ¸ elebi –– , Un important recueil avicennien du VIIe/13e sicle: La Majmu¯‘a Hseyin C 1194 de Brousse, Bulletin de philosophie mdivale, 33, 1991, pp. 121 – 9.

Essence and Existence in the Eleventh- and Twelfth-Century Islamic East (Masˇriq): A Sketch* Robert Wisnovsky Introduction

˘

˘

In a well known section of his Hikmat al-isˇra¯q (Philosophy of Illumination), ˙ Sˇiha¯baddı¯n as-Suhrawardı¯ (i. e., asˇ-Sˇayh al-Maqtu¯l, d. 587 H./1191) attacked ˘ the doctrine that existence (wug˘u¯d) is something superadded to (ma nan za¯ idun ala¯) the substance or quiddity of things in the concrete, extramental world (fı¯ la ya¯n)–a doctrine he associates with those he refers to as the followers of the Peripatetics (atba¯ al-Masˇˇsa¯ ¯ın).1 Suhrawardı¯ maintains, by contrast, that existence is among those aspects (i tiba¯ra¯t) of a thing that belong purely to the intellect.2 Partly because of Suhrawardı¯’s insistence on the subjective nature

˘ ˘

˘

˘

˘

1 2

Earlier versions of this paper were presented at the American Philosophical Association, Eastern Division Meeting, New York (December 2005); at the conference entitled ‘The Arabic, Hebrew and Latin reception of Avicenna’s metaphysics’, Centro Italo-Tedesco Villa Vigoni, Menaggio, Italy (July 2008); and as part of McGill’s Philosophy Department Colloquium series (November 2009). I am grateful for the feedback I received on those three occasions, as well as to the students and colleagues who participated in two graduate seminars: the first, at Harvard in 1998, was devoted to issues of ontology in Islamic philosophy; the second, at McGill in 2009, was devoted specifically to Suhrawardı¯. More particularly, Heidrun Eichner first helped me work through Ra¯zı¯’s position and graciously shared photocopies of relevant manuscripts that she had acquired; and Reza Pourjavady and Stephen Menn each made crucial suggestions that largely shaped the final form of my argument. Needless to say, all mistakes are my own. Suhrawardı¯, Hikmat al-isˇra¯q, I.3, § 59, p. 46,8-ult. ˙ ikmat al-isˇra¯q I.3, § 56, p. 45,1-ult., and § 60, p. 47, 1 – 13. The term Suhrawardı¯, H ˙ i tiba¯r is difficult to translate into English. In his extended discussion of this issue, T. Izutsu cites the famous passage in the Madhal (Isagoge) section of the Mantiq (Logic) of ˙ (ma¯hiyya) ˘ Avicenna claims that quiddity Avicenna’s Kita¯b asˇ-Sˇifa¯ (The Healing) where has three i tiba¯ra¯t: as a universal existing in the mind; as an essence existing in a concrete individual; and taken in and of itself, i. e., as neutral with respect to either mental or concrete existence. In light of this passage, and in light of Suhrawardı¯’s uses of the term in the Hikmat al-isˇra¯q, Izutsu takes i tiba¯r to mean ‘… a subjective manner of looking at a thing,˙ something produced or posited through the analytic work of the reason. It is an aspect of a thing which primarily appears in the subject and which, then, is projected onto the thing itself as if it were an objective aspect of the thing.’ See T. Izutsu, The Distinction between essentia and existentia, pp. 49 – 70 at 65. The Avicenna passage is ˘

*

˘

˘

˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

28

˘

˘

˘

of existence, his view was later used as an essentialist foil by Mulla¯ Sadra¯ and other members of that school of Islamic metaphysics which saw˙ itself as upholding ‘the fundamentality of existence (asa¯lat al-wug˘u¯d)’.3 ˙ Suhrawardı¯’s use of the phrase ‘the followers of the Peripatetics’ in this context is usually taken by medieval as well as modern commentators to refer to Fa¯ra¯bı¯ and especially Avicenna. In a broad sense this is perfectly plausible. After all, Suhrawardı¯ attacks the idea that existence is an attribute that itself has real existence in the concrete world, in the context of pointing to ‘the fact that some followers of the Peripatetics construct their entire metaphysical project on [the basis of ] existence (anna ba da atba¯ i l-masˇˇsa¯ ¯ına banaw kulla amrihim fı¯ lila¯hiyya¯ti ala¯ l-wug˘u¯di)’. This˙ must at least partially refer to Avicenna, who explicitly claimed, in the final chapter of Section 4 (‘On existence and its causes’) of his al-Isˇa¯ra¯t wa-t-tanbı¯ha¯t (Pointers and Reminders), to have created a new proof of God’s existence (viz., burha¯n as-siddı¯qı¯n) that was superior, by virtue of its basis in existence alone, to proofs ˙of˙ God’s existence from motion, such as Aristotle’s proof of the need for an Unmoved Mover based on the impossibility of an infinite regress of movers and moved things. Since existence provides us with such a shaky foundation, Suhrawardı¯ argues, we need to turn elsewhere, and create an alternative metaphysical basis in the form of ‘light’ (nu¯r). Nevertheless, a question arises, because to my knowledge Avicenna never explicitly committed himself to the thesis that existence is something ‘superadded to’ (za¯ id ala¯) a thing’s quiddity.4 True, there is one passage in the Ta lı¯qa¯t (Marginal Notes) where Avicenna states that ‘The existence of each category is extrinsic to its quiddity and superadded to it (fa-inna kulla maqu¯latin fawug˘u¯duha¯ ha¯rig˘un an ma¯hiyyatiha¯ wa-za¯ idun alayha¯); whereas the quiddity of ˘ of Existence is its thatness; hand its thatness is noti superadded to the Necessary [its] quiddity’.5 But given our current uncertainty about the circumstances in which the Ta lı¯qa¯t were composed, it would be rash to extrapolate a full-fledged theory from this isolated instance. In Book IV, Chapter 3, of the Ila¯hiyya¯t of his Kita¯b asˇ-Sˇifa¯ (The Healing), Avicenna does use the phrase al-wug˘u¯d az-za¯ id. But ˘

˘

˘

˘

˘

˘

˘

˘

˘

˘

˘

˘

Kita¯b asˇ-Sˇifa¯ /Mantiq (1): al-Madhal I.2, p. 15,1 – 7. J. Walbridge translates i tiba¯ra¯t ˙ ˘ the basis of an analogy with ‘legal fictions’: The aqliyya as ‘intellectual fictions’, on Science of Mystic Lights, pp. 45 – 6, n. 43. This label would probably have surprised Suhrawardı¯, given that in the passages just cited he maintains the subjective nature not only of existence but also of the various ways of conceptualizing essence, including quiddity (ma¯hiyya), thingness (sˇay iyya) and innerreality (haqı¯qa). On Suhrawardı¯’s role in Mulla¯ Sadra¯’s historiography of Islamic ˙ see S. Rizvi, An Islamic Subversion of the ˙ Existence-Essence Distinction?, philosophy, pp. 219 – 27. Walbridge raises this question but does not offer an answer: Science of Mystic Lights, pp. 47 – 8. Ibn Sı¯na¯, at-Ta lı¯qa¯t, IV.32, p. 164,18–ult.

˘

˘

3

4 5

˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Robert Wisnovsky

29

Essence and Existence in the Eleventh- and Twelfth-Century

˘

˘

what he is clearly referring to in this passage is the extra existence that God, who is above perfection (fawqa tama¯m), does not need for Himself and which He therefore passes on to other, lower beings.6 In a broader sense, Avicenna’s ontology could doubtless be interpreted as implying the thesis that existence is ‘superadded to’ (za¯ id ala¯) a thing’s quiddity. As I have discussed extensively in other publications, Avicenna’s general position on essence (or quiddity) and existence is that essence and existence are extensionally identical but intensionally distinct. In other words, every essence must either be an individual existing in the concrete, extramental world (fı¯ l-a ya¯n), or a universal existing in the mind (fı¯ d-dihn). Even so, essence and ¯ ¯ existent have different meanings: essence refers to what a thing is, whereas existence refers to the fact that a thing is. More important for my discussion here is the series of hints, given by Avicenna, that despite the fact that essence and existence are co-implied (the term he uses in Kita¯b asˇ-Sˇifa¯ /Ila¯hiyya¯t I.5 is mutala¯zima¯ni), essence nevertheless enjoys some kind of logical priority over existence. The sense that essence is logically prior to existence is conveyed by – among other clues – Avicenna’s frequent uses of the terms la¯zim (‘is logically entailed [by]’), a¯rid (‘attaches accidentally [to]’), la¯hiq (‘is a concomitant [of ]’) ˙ [to]’) to describe how existence ˙ connects to essence.7 An and muda¯f (‘is related ˙ interpreter could reasonably infer that describing existence as za¯ id (‘is superadded [to]’) would be perfectly in line with these other descriptions of how existence connects to essence – despite the fact that, apart from its lonely appearance in the Ta lı¯qa¯t, za¯ id is never used by Avicenna in this way. Given the prominence of Suhrawardı¯’s critique of the thesis that existence is something superadded to quiddity, and given the uncertainty about its Avicennian genealogy, we should still try to find out more precisely who Suhrawardı¯ was referring to when he targeted ‘the followers of the Peripatetics’ in this context. Avicenna may well have been in Suhrawardı¯’s sights, as has been commonly assumed. But the fact remains that the most prominent exponent of the thesis that existence is superadded to quiddity was Fahraddı¯n ar-Ra¯zı¯ (d. ˘ 606/1210), a contemporary of Suhrawardı¯’s and a fellow alumnus of ˘ ¯ılı¯’s (n.d.) circle in Mara¯g˙a. My hypothesis is that the balance Mag˘daddı¯n al-G of evidence compels us to think that Suhrawardı¯ was not so much targeting Avicenna’s own ontology as he was targeting an emerging Avicennian ontology – that is, the systematic reconstruction of Avicenna’s ontology that Ra¯zı¯ was just beginning to undertake. Because Ra¯zı¯ appears to have left Azerbaijan in 580/ ˘

˘

˘

˘

Ibn Sı¯na¯, Kita¯b asˇ-Sˇifa¯ /Ila¯hiyya¯t (1), IV.3, p. 188,11 – 13. I discuss Avicenna’s developing ideas about the relationship between essence and existence in my Notes on Avicenna’s concept of thingness (sˇay iyya), pp. 181 – 221; Avicenna’s Metaphysics in Context, pp. 143 – 80; and Avicenna and the Avicennian tradition, pp. 105 – 13. ˘

˘

6 7

˘

˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

30

˘

1184 for Transoxania, where he wrote most of the works in which he claims that existence is superadded to essence; and because Suhrawardı¯ finished his Hikmat al-isˇra¯q in 582/1186, it is unlikely that Ra¯zı¯ himself was in Suhrawardı¯’s ˙ sights. But regardless of the identity of the particular person whom Suhrawardı¯ ˇ ¯ılı¯ as well as Ra¯zı¯’s father and his Asˇ arite circle in saw himself as opposing (G Rayy present themselves as possibilities, but further research will be needed in order to determine this), Suhrawardı¯’s arguments in favor of the conceptual nature of existence clearly recapitulate an earlier attack, by the mathematicianpoet Umar Hayya¯m (d. 517/1123), against the tendency of certain Asˇ arite ˘ uwaynı¯ (d. 478/1085), to appeal mutakallimu¯n,˘ such as Ima¯m al-Haramayn al-G ˙ to the theory of modes (ahwa¯l) – a theory associated with the Basran Mu tazilite ˘ ubba¯ ¯ı (d. 321/933) and his followers, the mutakallim Abu¯ Ha¯ˇsim˙ al-G Bahsˇamites – as the best way to construe and promote Avicenna’s concept of existence. ˘

˘

˘

˘

Avicenna’s Two Distinctions

˘

To begin this story, I must turn first to Avicenna himself. Avicenna’s two key metaphysical distinctions were between essence (or more properly, ‘quiddity’, ma¯hiyya) and existence (wug˘u¯d); and between the necessary of existence in itself (wa¯g˘ib al-wug˘u¯d bi-da¯tihi) and the necessary of existence through another ¯ (wa¯g˘ib al-wug˘u¯d bi-g˙ayrihi), which Avicenna appears to have taken as convertible with the possible (or ‘contingent’) of existence in itself (mumkin al-wug˘u¯d bi-da¯tihi). Avicenna’s distinction between essence and existence, in its ¯ mature formulation in al-Ila¯hiyya¯t (Metaphysics) I.5 of his Sˇifa¯ and in Isˇa¯ra¯t 4, can be seen from one angle as a compromise position, stated in ArabicAristotelian terminology, between the view of the early Mu tazilite mutakallimu¯n and that of al-Asˇ arı¯.8 Like Asˇ arı¯, Avicenna maintains – in Sˇifa¯ , Ila¯hiyya¯t I.5 – that thing (sˇay ) and existent (mawg˘u¯d), and by implication quiddity and existence, are extensionally identical: every existent will also be a thing, and vice versa. This is in contrast to the position of the early Mu tazilites, who believed that thing was a broader category than existent, in that thing subsumes both the non-existent (ma du¯m) and the existent. To the Mu tazilites, entities that had not yet come to be, and concepts in the mind, are examples of non-existent things: therefore, non-existents as well as existents possess thingness (sˇay iyya). By contrast, Avicenna argues that things such as concepts in the mind do enjoy a kind of existence – they simply possess mental existence (al-wug˘u¯d ad-dihnı¯ or ¯ ¯ al-wug˘u¯d fı¯ d-dihn) as opposed to the concrete existence found in individuals ¯ ¯ (al-wug˘u¯d al- aynı¯ or al-wug˘u¯d fı¯ l-a ya¯n; also referred to as ‘extra[mental] ˘

˘

˘

˘

˘

˘

˘

˘

˘

8

Ibid.

˘

˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Robert Wisnovsky

31

Essence and Existence in the Eleventh- and Twelfth-Century

˘

existence’, al-wug˘u¯d al-ha¯rig˘¯ı). But unlike Asˇ arı¯, who maintained that thing and ˘ existent were also intensionally identical, in the sense that thing means no more or less than existent, and vice versa, Avicenna claimed that quiddity or thingness (abstracted from thing) on the one hand, and existence (abstracted from existent) on the other hand, were intensionally distinct. As I mentioned above, for Avicenna, thingness and quiddity refer to what X is (i. e., as opposed to what Y is); existence, by contrast, refers to the fact that X is (i. e., as opposed to X’s not existing). Unlike his distinction between essence and existence, which appears to have evolved over the course of his career but only in a subtle way, Avicenna’s distinction between intrinsically and extrinsically necessary existence underwent some dramatic developments from its first appearance in al-Hikma al- Aru¯diyya ˙ ˙ (Philosophy for Aru¯d¯ı) to its final appearance in the Isˇa¯ra¯t. More directly relevant ˙ to this chapter is the fact that in addition to articulating each of these two distinctions in slightly different ways in books that he wrote at various points in his life, Avicenna appears to have bound the two distinctions more closely together as his ideas developed over time.9 Thus in the very early al-Hikma al˙ Aru¯diyya, the distinction between essence and existence had hardly crystallized, ˙ and the distinction between intrinsically and extrinsically necessary existence was not thought through; and the two distinctions do not touch upon one another at all.10 In the slightly later al-Mabda wa-al-ma a¯d (Origin and Destination), the distinction between quiddity and existence is still only latent, while the distinction between necessary and possible existence is quite fully articulated; still, neither is linked directly to the other.11 In chapters I.5 and I.6 of the Ila¯hiyya¯t of his Sˇifa¯ , from Avicenna’s middle period, the two distinctions receive their fullest expression.12 And while in those chapters neither distinction is brought directly to bear on the other, they are later, in Book VIII of the Ila¯hiyya¯t. 13 There Avicenna buttresses the distinction between God, the Necessary of Existence in itself, and all other beings in the universe, which are necessary of existence through another, by appealing to the notion that in God quiddity and existence are identical, while in all other beings, quiddity and existence are distinct. In his final major work, the Pointers and Reminders (alIsˇa¯ra¯t wa-t-tanbı¯ha¯t), the two distinctions operate entirely in tandem, and the distinction between quiddity and existence lays the basis for, and leads directly to, the distinction between intrinsically and extrinsically necessary existence.14 ˘

˘

˘

˘

˘

˘

Wisnovsky, Avicenna’s Metaphysics in Context, pp. 245 – 63. Ibn Sı¯na, al-Hikma al- Aru¯diyya (MS Uppsala Or. 364), fol. 2v8 – 10 and fols 3v16 – 4r12. ˙ ˙ a¯d, pp. 2,4 – 3,15. Ibn Sı¯na¯, al-Mabda wa-l-ma ˇ Ibn Sı¯na¯, Kita¯b asˇ-Sifa¯ /Ila¯hiyya¯t (1), I.5, pp. 31,5 – 33,18 and I.6, pp. 37,7 – 38,5. Ibn Sı¯na¯, Kita¯b asˇ-Sˇifa¯ /Ila¯hiyya¯t (2), VIII.4, pp. 343,10 – 347,16. Ibn Sı¯na¯, Kita¯b al-Isˇa¯ra¯t wa-t-tanbı¯ha¯t, pp. 138,2 – 139,13 and pp. 140,12 – 141,2. ˘ ˘

˘

˘

9 10 11 12 13 14

˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

32

˘

˘

Why did Avicenna decide to bind the two distinctions together in his middle and later works? I maintain that it is because he realized, during the process of writing the Ila¯hiyya¯t of the Sˇifa¯ , that he could use the intensional distinctiveness between quiddity and existence to show how beings other than God were composites, that is, composed of quiddity and existence. God, by contrast, could be held to be simple by virtue of the identity of quiddity and existence in Him. Because every composite needs a composer to bring its components together, and because of the impossibility of an infinite regress, the chain of composites and composers must originate in a non-composite composer, namely, God. Partly because Avicenna’s use of the quiddity-existence distinction to support the intrinsically necessary-extrinsically necessary distinction was most obvious in the Isˇa¯ra¯t, and partly because of that work’s abbreviated and allusive style, which invites decompression and commentary, the Isˇa¯ra¯t received more attention from subsequent Muslim philosophers (including the mutakallimu¯n) than any other of Avicenna’s writings – at least until the sixteenth century CE, when the attention of commentators shifted to the Sˇifa¯ . Avicenna’s pressing of the essence-existence distinction into the service of his intrinsically necessary-extrinsically necessary distinction was a crucial event in the history of metaphysics. This is because it provided a method of distinguishing God from both eternal and non-eternal beings that was based on God’s simplicity and all other beings’ compositeness; and because, when understood as the Necessary of Existence in itself, whose essence is not even conceptually distinct from its existence, Avicenna’s God enjoyed a more watertight simplicity than that of the Neoplatonists, whose God as One could be held to be conceptually distinguishable from their God as Good. 15 As will become apparent, Fahraddı¯n ar-Ra¯zı¯’s ontology can be seen as a continuation of ˘ this trend in Avicenna’s own thought, that is, the trend towards using the essence/existence distinction to explain the compositeness of all extrinsically necessary beings.

Theological Ramifications To be sure, it was not Avicenna’s proof of God’s existence from the distinctiveness (and hence compositeness) of essence and existence in all beings other than the Necessary of Existence in itself, which first made Avicenna’s metaphysics attractive to Sunni mutakallimu¯n (specifically, those of the Asˇ arite and Ma¯turı¯dite schools) from the two or three generations immediately ˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Robert Wisnovsky

15 On this see my Final and Efficient Causality in Avicenna’s Cosmology and Theology, pp. 97 – 123, and Avicenna’s Metaphysics in Context, pp. 181 – 95.

33

Essence and Existence in the Eleventh- and Twelfth-Century

following his death. Instead, it was the use that they could make of Avicenna’s distinction between intrinsically and extrinsically necessary existence in order to solve some serious problems with their theology, and particularly with their conception of how God’s attributes (sifa¯t) were related to the divine self (da¯t). ¯ ˙ Sunni mutakallimu¯n from the generations immediately following Avicenna’s death in 1037 CE, appropriated his distinction between intrinsically and extrinsically necessary existence with a view to repairing a two-century-old weakness of Sunni theology. More specifically, Sunni mutakallimu¯n such as the ˇ uwaynı¯ – and to a lesser extent the Ma¯turı¯dite al-Bazdawı¯ (d. 493/ Asˇ arite G 1099) – sought to use new conceptual tools from Avicenna’s metaphysics in order to modify and thereby strengthen the theory of divine attributes that they had inherited from Ibn Kulla¯b (d. ca. 241/855) and which, among other doctrines, served to distinguish their theology from that of their Mu tazilite competitors.16 ˘ uwaynı¯ and other Sunni mutakallimu¯n The theological problem facing G was an unfortunate legacy of their inferring from Ibn Kulla¯b’s formula that God is knowing ( a¯limun) by virtue of a knowledge (bi- ilmin) that is neither identical to nor other than Him (la¯ huwa wa-la¯ g˙ayruhu), and powerful (qa¯dirun) by virtue of a power (bi-qudratin) and speaking (mutakallim) by virtue of a speech (kala¯m) that are neither identical to nor other than Him – so therefore God is also eternal (qadı¯m) by virtue of an eternality (bi-qidamin) that is neither identical to nor other than Him. The first part of the Sunnis’ theological problem arose because God’s attribute of eternality could not be treated in the same way as His other attributes. This is because the Sunnis held that not only is God Himself eternal, but God’s attributes, such as His knowledge and His power and His speech, are eternal too.17 The Sunni mutakallimu¯n viewed eternality as both an attribute (or first-order predicate) and as a meta-attribute (or second-order predicate). In other words, eternality is a divine quality that is predicable not only of God’s self but of other divine qualities too. Early Mu tazilites, by contrast, tended to collapse all the divine attributes into God’s self. According to the formula associated with Abu¯ l-Hudayl (d. ca. 226/841), for example, God is knowing through a knowledge ¯that is identical to Him ( a¯limun bi- ilmin huwa huwa).18 The Mu tazilites were motivated by the fear of violating their cardinal tenet of tawh¯ıd, or divine ˙ ˘

˘

˘

˘

˘

˘

˘

˘

˘

16 A broader discussion of this development is contained in my One Aspect of the Avicennian Turn in Sunni Theology, pp. 65 – 100. 17 This was not a two-way street, however: just because God’s speech was eternal, it did not mean that conversely, God’s eternality was speaking. Nor was every attribute eternal: the ‘attributes of the act’ (sifa¯t al-fi l) such as God’s ‘providing’ (rizq, i. e., food and water for humans and animals),˙ which appeared to require a creaturely object on which to act, were held (by the Asˇ arites, though not by their Ma¯turı¯dite colleagues) to be originated. 18 al-Asˇ arı¯, Maqa¯la¯t al-isla¯miyyı¯n, p. 165,5. ˘

˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

34

˘

uniqueness. For if God’s attributes were meaningfully distinct from God’s self – in this case, distinct enough that each of the attributes could itself be seen as a subject of which the predicate ‘eternal’ holds – then the attributes would share the quality of eternality with God’s self, and a picture would emerge of a pleroma of eternal, and hence quasi-independent, divine beings, rather than of a single God, isolated in His sole possession of eternality. In a vain attempt to preempt this Mu tazilite critique, early Sunnis tended to stick to Ibn Kulla¯b’s formula ‘neither identical to God nor other than Him’ (la¯ hiya lla¯hu wa-la¯ hiya g˙ayruhu).19 The first problem facing Sunni mutakallimu¯n, therefore, was that God’s eternality did double-service, both as a first-order predicate of God’s self and as a second-order predicate of God’s other attributes. As soon as the other attributes themselves became subjects of which ‘eternal’ was predicated, God no longer enjoyed the complete isolation He had before, since each of His eternal attributes had become sufficiently real and distinct that Sunni theology had opened itself up to the criticism of polytheism. Just as serious was the danger of positing an infinite regress of eternalities. For if each attribute is eternal, is it eternal in itself, or through a further eternality? If the Sunni mutakallim answers that each attribute is eternal in itself, then he will have made an explicit claim that the attributes really are independent entities in their own right, and this in turn will provide his Mu tazilite competitor with clear evidence of polytheism. But if an eternal attribute is eternal through yet another eternality, then each of the second-order meta-eternalities will itself be eternal through a third-order eternality, and an infinite regress will ensue.20 ˘ uwaynı¯ and other As if this first challenge was not enough of a headache, G Sunni mutakallimu¯n had to address another weakness. The traditional kala¯m proof of God’s existence relied on the contradictory nature of the opposition between eternal (qadı¯m) and originated (muhdat). Since everything that is ˙ ¯ originated needs an originator (muhdit) to bring it into existence, and since an ˙ ¯ infinite series of originators and originated things is impossible, the chain of originators and originated things will originate in an originator that is not itself originated; and since everything in the world is either eternal or originated, this non-originated originator will be eternal. That eternal originator is God. This proof worked fine for the Mu tazilites, who, as mentioned above, collapsed God’s attributes into the divine self.21 But the Sunnis’ commitment to the ˘

˘

19 al-Asˇ arı¯, Maqa¯la¯t al-isla¯miyyı¯n, pp. 169,2 – 170,3. 20 This worry is expressed by, e. g., al-Qusˇayrı¯ (d. 466/1072), Sˇarh asma¯ Alla¯h al-husna¯, ˙ ˙ p. 55,8 and pp. 392,5 – 7. ˘ 21 See, for example, the Basran Mu tazilite Abd al-Gabba¯r’s (d. 415/1025) description of ˘ ubba¯ ¯ı’s (d. 303/915) position, in al-Mug˙nı¯ fı¯ abwa¯b athis predecessor Abu¯ Alı¯ al-G tawh¯ıd wa-l- adl, V, p. 233,5. ˙ ˘

˘

˘

˘

˘

˘

˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Robert Wisnovsky

35

Essence and Existence in the Eleventh- and Twelfth-Century

˘

eternality of the divine attributes entailed that God, for them, was not the only eternal thing in the universe: the divine attributes were eternal too. This meant that the Sunnis found themselves with a category of beings – the divine attributes – that were in one sense eternal but which were not themselves originators. With the sudden appearance of this shadowy middle ground, the traditional proof no longer works, since the opposition between eternal and originated is now one between contraries, not between contradictories.22 If the proof is to work, there must be an identity between ‘eternal’ and ‘non-originated originator’; but now there seem to be eternal, non-originated things – the divine attributes – that are not originators. Faced with philosophical challenges such as these, it is not surprising that ˘ uwaynı¯ desperately sought tools to Sunni mutakallimu¯n such as Ba¯qilla¯nı¯ and G help them confront or at least side-step the problematic implications of their doctrinal commitments. One of those tools was Avicenna’s distinction between intrinsically and extrinsically necessary existence. Since necessity can be construed as modifying the predication itself, as opposed to serving as a second-order attribute in the way eternality had, Sunni mutakallimu¯n could come up with a new formula that did not predicate second-order attributes of the first-order attributes, and hence was in less danger of inadvertently positing the attributes as subjects or selves. Thus a Sunni mutakallim could assert the modal propositions ‘God is necessarily an existent’ and ‘God is necessarily a knower’, and thereby avoid stating that ‘God is necessary’ and ‘God’s knowledge is necessary’ – a pair of attributions that would create the same infinite-regress problem with necessity that ‘God is eternal’ and God’s knowledge is eternal’ had created earlier with eternality.23 Nevertheless, the new formulas raised further challenges, one of which was explaining how the divine attribute of existence (wug˘u¯d) connected to the divine self or essence (da¯t) in such a way that it ¯ sounded like Sunni attribute-theory and not Mu tazilite attribute-theory.

The Status of Modes

˘

˘ uwaynı¯ turned for help not only to Partly because of these dilemmas, G Avicenna’s metaphysics. He, and (apparently) Ba¯qilla¯nı¯ before him, also seized ˘ ubba¯ ¯ı, the upon the theory of ahwa¯l, or modes, proposed by Abu¯ Ha¯ˇsim al-G ˙

˘

˘

˘

˘

22 As the Asˇ arite mutakallim al-Ba¯qilla¯nı¯ (d. 403/1013) appears to admit; see his Kita¯b atTamhı¯d, pp. 29,17 – 30,2. ˘ uwaynı¯, asˇ-Sˇa¯mil fı¯ usu¯l ad-dı¯n, p. 292,19 – 20, p. 308,9 – 10, 23 See, for example, al-G ˘ ˙uwaynı¯ credits Ba¯qilla¯nı¯ for this move); p. 358,11 – 13 and p. 365,7 – 11 (where G al- Aqı¯da an-niza¯miyya, p. 23,4 – 5; and Luma fı¯ qawa¯ id ahl as-sunna wa-l-g˘ama¯ a, p. 137,9 – 10. ˙ ˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

36 ˘

˘ uwaynı¯’s and Ba¯qilla¯nı¯’s theological school. arch-rival of Asˇ arı¯, who founded G The theory of modes has commonly been viewed as a theological compromise designed to blunt the horns of the dilemma produced by construing the assertion ‘God is powerful’ (Alla¯hu qa¯dirun) as implying either that God is power (Alla¯hu qudratun) or that God has power (Alla¯hu lahu qudratun). As mentioned above, early Mu tazilites such as Abu¯ al-Hudayl took the former tack and collapsed the attributes into God’s self, while early¯ Sunnis – following Ibn Kulla¯b – took the latter tack and viewed the attributes as entities that were real enough to be meaningfully distinct from God’s self and from each other. Abu¯ Ha¯ˇsim appears to have been at least partly motivated by the worry that the early Mu tazilites’ position entailed that the semantic content of all attributive assertions, such as ‘God is power’, is reducible to the assertion ‘God is God’, and that all such assertions are therefore meaningless. So he appealed to an element of classical Arabic grammar – the ha¯l – for a solution. A ha¯l, which ˙ or phrase, refers to ˙a state or often takes the form of an adverbial accusative condition that modifies the subject or object of a verb at the moment (ha¯l) ˙ when the event described by the verb is taking place. Thinking of a divine attribute as a ha¯l, as a mode which describes God in His ‘act’ of being, allowed Abu¯ Ha¯ˇsim to˙ hold (unlike the early Mu tazilites) that God’s being powerful (kawnuhu qa¯diran) is semantically distinct from, say, His being alive (kawnuhu hayyan), just as the ha¯l ‘riding’ in ‘Smith came riding’ (g˘a¯ a Zaydun ra¯kiban) is ˙semantically distinct˙ from the ha¯l ‘walking’ in ‘Smith came walking’ (g˘a¯ a Zaydun ma¯ˇsiyan); but without ˙implying (as the Sunnis had) the separate existence of real entities such as ‘power’ and ‘life’ or ‘riding’ and ‘walking’. In recent decades, scholars’ focus has extended from the theological motivations and implications of mode theory to its ontological motivations and ˘ uwayni’s prime concern in appropriating implications.24 Similarly, although G aspects of Bahsˇamite (i. e., Abu¯ Ha¯ˇsim’s) mode-theory was theological rather than ontological, his appropriation had implications beyond an analysis of the ˘ uwayni’s most extensive relationship between God’s self and His attributes. G discussion of modes comes in his Kita¯b al-Irsˇa¯d, where he states baldly that ‘The mode is an attribute which belongs to an existent and which is [itself ] qualified ˘ uwaynı¯ held that by neither existence or non-existence’.25 To be precise, G ‘knowledge’, ‘power’ and ‘speech’ are real objects (ma a¯nı¯) that are possessed by a being – God – and in which the modes (ahwa¯l) ‘being-a-knower’, ‘being˙ ˘

˘

˘

˘

˘

˘

˘

24 See A. Alami, L’ontologie modale; the classic treatments are by R. Frank, Abu Hashim’s Theory of ‘States’, and more generally, Beings and Their Attributes; and most comprehensively as well as most comprehensibly, H. Wolfson, The Philosophy of the Kalam, pp. 167 – 234. 25 Cf. al-Irsˇa¯d fı¯ usu¯l al-i tiqa¯d, transl. and ed. J. Luciani, Paris, 1938, p. 80,6 – 13 and ˙ p. 47,11 – 13; cited by M. Allard, Le problme des attributes divins dans la doctrines d’al-Ash arı¯ et de ses premiers grands disciples, p. 389. ˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Robert Wisnovsky

37

Essence and Existence in the Eleventh- and Twelfth-Century

˘

˘

powerful’ and ‘being-a-speaker’ are grounded (mu allala). While the attributes understood as ma a¯nı¯ really exist in the being that possesses them, the attributes understood as ahwa¯l are neither existent nor non-existent. ˙ The intermediate ontological status enjoyed by modes, which are neither existent nor non-existent, lies at the bottom of a series of objections that eventually led to Suhrawardı¯’s attack on the conception of existence held by ‘the followers of the Peripatetics’. The ahwa¯l clearly violate the law of the excluded middle, which holds that a thing S˙ is either P or not-P, with no third option available. Forcing those who uphold a ha¯l-type theory of attribution to commit ˙ themselves to the existence or non-existence of the attribute in question was a theme running through eleventh-and twelfth-century discussions of ontology; and skepticism about the real existence of attributes such as existence motivated the mathematician and poet Umar Hayya¯m (d. 517/1123), who wrote a brief but acute Essay on Existence (Risa¯la ˘fı¯ l-wug˘u¯d). 26 In the first chapter of that work, Hayya¯m distinguishes between different types of attribute (sifa): that which is˘ essential (da¯tı¯) to the characterized thing (mawsu¯f ), and that˙ which is ¯ ˙ (la¯zim) of the accidental ( arad¯ı) to it; that which is a necessary concomitant ˙ characterized thing, and that which is not; and last but not least, that which is separable from the characterized thing only in the faculty of estimation (mufa¯riq bi-l-wahm) – that is, conceptually – and that which is separable in the faculty of estimation as well as in (concrete) existence (mufa¯riq bi-l-wahm wa-bi-lwug˘u¯d).27 This last division of attributes is most relevant to my argument, because it lays the ground for Hayya¯m’s distinction between attributes that are conceptual ˘ are existential (wug˘u¯dı¯). Existential attributes are real in (i tiba¯rı¯) and those that the sense that they exist concretely in the characterized thing, whereas conceptual attributes are not real, because they do not exist concretely in the characterized thing, but instead attach to it only in our minds. Hayya¯m’s ˘ with example of an existential-accidental ( arad¯ı wug˘u¯dı¯) attribute is ‘black’ ˙ respect to ‘black body’. Here the attribute ‘blackness’ (sawa¯d) is ‘something superadded to the black thing itself [and] existing in concrete reality’ (ma nan za¯ idun ala¯ da¯ti l-aswadi mawg˘u¯dun fı¯ l-a ya¯ni). To be precise, Hayya¯m holds ¯ ˘ attribution ˇ uwaynı¯’s ha¯l) is an existential that ‘black’ (i. e., corresponding to G ˙ ˇ uwaynı¯’s ma na¯) is (wasf wug˘u¯dı¯), whereas ‘blackness’ (i. e., corresponding to G ˙ an existential attribute (sifa wug˘u¯diyya). Hayya¯m takes the real existence of this ˙ ˘ type of attribute to be evident at the intellectual, estimative and sensory levels, ˘

˘

˘

˘

˘

˘

˘

˘

˘

26 Umar Hayya¯m, Risa¯la fı¯ l-wug˘u¯d, ed. R. Rida¯za¯da-yi Malik, pp. 398,9 – 403,10. This ˙ edition ˘contains quite a few mistranscriptions, and should be compared with that contained in S.G. Tı¯rtha, transl. and ed., The Nectar of Grace. Hayya¯m’s discussion in the Risa¯la is reiterated in his Daru¯riyyat at-tada¯dd fı¯ l- a¯lam wa-l-g˘˘abr wa-l-baqa¯ . ˙ ˙ Risa¯la fı¯ l-wug˘u¯d, pp. 398,13 – 399,16. 27 This and what follows summarize Hayya¯m, ˘ ˘

˘

˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

38

˘

˘

˘ uwaynı¯ on the reality of ma a¯nı¯ (such as ilm – and in this sense he agrees with G knowledge) in which the ahwa¯l (such as kawnuhu a¯liman – being-a-knower) are ˙ grounded. However, when he treats the type of attribute he calls ‘essential-conceptual’ ˘ uwaynı¯. Hayya¯m’s example is ‘being-a(da¯tı¯ i tiba¯rı¯), Hayya¯m departs from G ¯ ˘ ˘ color’, for ‘blackness’. Hayya¯m reasons that colorness (lawniyya) is not an ˘ attribute existing in concrete reality and superadded to blackness (sawa¯diyya), because the fact of being superadded to something else in concrete reality entails that the superadded thing be an accident; yet blackness is itself accidental, and how can one accident serve as a subject for another accident that is predicated of it? According to Hayya¯m’s schema, then, an existential-accidental attribute such ˘ as blackness is indeed something superadded, in concrete reality, to the black thing itself (fa-amma¯ l-qismu l-wug˘u¯diyyu l- aradiyyu fa-huwa ka-wasfi l-g˘ismi bi˙ wug˘u¯diyyatun ay huwa ˙ l-aswadi ida¯ ka¯na aswada fa-inna s-sawa¯da sifatun ma nan ¯ ˙ za¯ idun ala¯ da¯ti l-aswadi mawg˘u¯dun fı¯ l-a ya¯ni). But in the case of a conceptual¯ essential attribute such as the fact that blackness is a color, the attribute colorness is not an attribute superadded in concrete reality to the blackness itself (waamma l-qismu l-i tiba¯riyyu d-da¯tiyyu fa-huwa ka-wasfi s-sawa¯di bi-annahu ¯ ¯ bi-sifatin za¯ idatin ala ˙ ¯ da¯ti s-sawa¯diyyati fı¯ lawnun…wa…l-lawniyyatu laysat ˙ l-a ya¯ni). Hayya¯m proceeds directly to criticizing those who¯ neglect this crucial distinction˘ between conceptual and existential attributions: ˘

˘

˘

˘

˘

˘

˘

˘

˘

˘

˘

Those investigating this topic who do not take these conceptual attributions [awsa¯f ˙ i tiba¯riyya] into account have erred very far indeed, as is the case with some reckless moderns [al-muta assifı¯na l-muta ahhirı¯na], who posit colorness and accidentality and existence [my italics] and similar˘ ˘states as modes (ahwa¯l) that obtain in what can be characterized by neither existence or non-existence.˙ The doubt that makes them fall into this grave mistake pertains to the greatest of First Premises: that there is no middle ground between negation and affirmation, the self-evident nature of which needs no discussion by us, nor is there any way for idiots to contradict it or explain it away.28

˘

˘

˘

˘

Given the Sunni appropriation of ha¯l-theory discussed above; given Hayya¯m’s ˙ ˘ of the insistence in another treatise that Avicenna is to be regarded as ‘the best 29 Modern [philosophers] (afdal al-muta ahhirı¯n)’; given that Hayya¯m is ˙ (admittedly, ˘ ˘ one who is known for˘embellishdescribed by a bio-bibliographer ing his notices with apocrycphal anecdotes) as having gotten himself into trouble by defending Avicenna against Abu¯ l-Baraka¯t al-Bag˙da¯dı¯’s (d. 560/1165) attacks;30 and given that Avicenna himself attacked the Mu tazilite conception of ˘

Da¯nisˇna¯ma-yi hayya¯mı¯, p. 399,18 – 24. ˘ al-kawn wa-t-taklı¯f, p. lxxxvii,8 – 13. Hayya¯m, Risa¯lat ˘ See al-Bayhaqı¯’s notice on Ala¯ ad-Dawla Fara¯marz, King of Yazd, in Tatimmat siwa¯n al˙ hikma, pp. 110,7 – 111,11. ˙ ˘

28 29 30

˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Robert Wisnovsky

39

Essence and Existence in the Eleventh- and Twelfth-Century

ahwa¯l for violating the law of the excluded middle;31 Hayya¯m’s use of ahwa¯l ˙ by implication ma a¯nı) in the passage above makes ˘it likely that his target ˙ (and ¯ in this long attack is not Avicenna himself, but Bahsˇamizing Sunni ˘ uwayni.32 mutakallimu¯n such as Ba¯qilla¯nı¯ and G The appropriation of ahwa¯l in late-tenth and eleventh-century Asˇ arite ˙ kala¯m is well attested by twelfth-and thirteenth-century authors, who level much the same accusation against mode-theory that Hayya¯m does, namely, that ˘ the ahwa¯l violate the law of the excluded middle by falling in between existence ˙ and non-existence. For example, Ibn Hazm (d. 456/1064), a Mag˙ribı¯ near˘ uwaynı¯ (and thus ˙almost certainly referring here to alcontemporary of G Ba¯qilla¯nı¯), stated in his Fisal that ‘One piece of Asˇ arite nonsense is their claim ˙ that people can believe in modes and objects (al-ma a¯nı¯) which are neither existent nor non-existent’.33 One generation after Umar Hayya¯m, asˇ-Sˇahrasta¯nı¯, ˘ in his Milal, describes Abu¯ Ha¯ˇsim as having ‘postulated modes as attributes that are neither existent [nor non-existent] and neither knowable nor unknowable’. In a passage that may have served as a bridge between Hayya¯m’s attack and ˘¯ Ha¯ˇsim’s modes are Suhrawardı¯’s in the Hikmat al-isˇra¯q, Sˇahrasta¯nı¯ says that Abu ˙ ˘

˘

˘

˘

˘

˘

˘

… intellectual aspects (wug˘u¯h) and [subjective] considerations (i tiba¯ra¯t), that is, they are the conceptual (mafhu¯ma) [products of when] when we judge things as being alike by commonality (al-isˇtira¯k) or as being unlike by disjunction (al-iftira¯q). But these aspects are like relationships (al-nasab), correlationships (al-ida¯fa¯t), proximity (al-qurb), remoteness (al-bu d) and the like, which, according ˙to the consensus of opinion, are not to counted among the real attributes.34

˘

Sˇahrasta¯nı¯’s criticism is echoed in his Niha¯yat al-iqda¯m, where he describes Abu¯ Ha¯ˇsim’s ahwa¯l as intellectual aspects (al-wug˘u¯h al- aqliyya) and mental and estimative ˙considerations (i tibara¯t dihniyya wa-taqdı¯riyya).35 In his Musa¯ra a, ¯ ˙ Sˇahrasta¯nı¯, like Hayya¯m, uses colorness as an example of a mental consideration ˘ ˘

˘

˘

31 Ibn Sı¯na¯, al-Muba¯hata¯t, § 181–§ 182, p. 94,1 – 7. ˙ ¯ 32 It is tempting to speculate that another factor at play here may have been the fact that ˘ Guwaynı¯’s pupils were among Hayya¯m’s rivals for Selg˘uq infrastructure-funding, with some money from the Selg˘uq ˘court going to the new Niza¯miyya madrasas (where ˘ uwaynı¯ was employed), and other Selg˘uq money going to Ma˙¯lik-Sˇa¯h’s new observatory G in Isfahan (where Hayya¯m was employed). ˘ wa-l-ahwa¯ wa-n-nihal, no ed., (Cairo: 1317 – 1327 H.), vol. IV, 33 Kita¯b al-Fisal fı¯ l-milal ˙ ˙ Philosophy˙ of the Kalam, p. 170. pp. 208,5 – 6; cited by Wolfson, ˇ 34 Sahrasta¯nı¯, Kita¯b al-milal wa-n-nihal (Book of Religious and Philosophical Sects, W. ˙ Leipzig, 1923), p. 56,3 – 4 and pp. 56,16 – 57,1; Cureton, ed., London, 1846; reprinted cited by Wolfson, Philosophy of the Kalam, pp. 171 – 172 and pp. 198 – 199. 35 Sˇahrasta¯nı¯, Niha¯yat al-iqda¯m fı¯ ilm al-kala¯m (The Summa Philosophiae of al-Sˇahrasta¯nı¯, A. Guillaume, ed., London, 1934), p. 135,2 – 5; cited by Wolfson, Philosophy of the Kalam, p. 199. In his correspondence with ¯Ila¯qı¯, Sˇahrasta¯nı¯ also alludes to the fact that existence is something conceptual (ma nan i tiba¯riyyun): Guftgu¯-yi Sˇahrasta¯nı¯ va-I¯la¯qı¯, p. 101,21. ˘

˘

˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

40

(i tiba¯r dihnı¯).36 In the Niha¯ya Sˇahrasta¯nı¯ also reports that Ba¯qilla¯nı¯ and ˘ uwaynı¯¯ appropriated Abu¯ Ha¯ˇsim’s mode-theory; and in the Milal he G complains that mode-theory is vulnerable to al-Asˇ arı¯’s own criticism that ahwa¯l violate the law of the excluded middle.37 ˙ In the thirteenth century, asˇ-Sˇahrazu¯rı¯ (d. 7th/13th century), in his Divine ˘ ubba¯ ¯ı of the Mu tazilites, Tree (Sˇag˘ara Ila¯hiyya), explicitly names Abu¯ Ha¯ˇsim al-G ˘ ˘ and Guwaynı¯ and ‘Qa¯d¯ı Abd al-Gabba¯r’ from among the Asˇ arites (sic; ‘ Abd al˙ ˘ abba¯r’ is almost certainly G a mistaken scribal insertion after Qa¯d¯ı, which in ˙ contexts such as this normally refers to Ba¯qilla¯nı¯), as those mutakallimu ¯ n who adopted the idea of ahwa¯l, which appear to fall between existence and non˙ existence.38 This is echoed by the Asˇ arite mutakallim Sayfaddı¯n al-A¯midı¯ (d. 631/ 1233), who claims that Qa¯d¯ı Abu¯ Bakr (i. e., Ba¯qilla¯nı¯) and Ima¯m Abu¯ l-Ma a¯lı¯ ˘ uwaynı¯) agreed with˙ Abu¯ Ha¯ˇsim regarding ahwa¯l.39 (i. e., G ˙

˘

˘

˘ ˘

˘

˘

˘

˘

˘

Ra¯zı¯ and Suhrawardı¯ ˘ ¯ılı¯ – we cannot be sure As mentioned above, Ra¯zı¯ (and perhaps his teacher G ˘ because only one of G¯ılı¯’s writings, a treatise on the fourth figure of the syllogism, appears to have survived) followed the late trend in Avicenna’s metaphysics towards pressing the essence/existence distinction into the service of the intrinsically necessary/extrinsically necessary distinction. Motivated by his desire to reinforce the compositeness of contingent beings by hardening the distinction between essence and existence in them, by his Sunni theological commitment to the distinctiveness between self (or quiddity) and attribute, and by his commitment to the univocity of existence (on which more below), Ra¯zı¯ recast ˘

˘

˘

36 wa-anta ta rafu anna l-lawniyyata wa-l-baya¯diyyata tiba¯ra¯ni aqliyya¯ni fı¯ d-dihni la¯ fı¯ ¯ ¯¯nı¯, Kita¯b ˙¯ di g˙ayra baya¯diyyatihi. Sˇahrasta l-ha¯rig˘i wa-alla¯ fı¯ l-wug˘u¯di lawniyyata l-baya ˙ ˙ ˘ al-Musa¯ra a, p. 36,3 – 4 (Arabic) and p. 39 – 40 (English). ˙ ¯nı¯, Niha¯ya, p. 131,5 – 9; Milal, p. 57,5 – 6 and p. 67,2 – 8; cited by Wolfson, 37 Sˇahrasta Philosophy of the Kalam, pp. 175 – 6 and 199 – 200. 38 Rasa¯ il asˇ-sˇag˘ara al-ila¯hiyya, pp. 31,6 – 35,18. ˙ a¯yat al-mara¯m fı¯ ilm al-kala¯m, p. 29,11 – 13; Abka¯r al-afka¯r fı¯ usu¯l ad-dı¯n, 39 al-A¯midı¯, G ˙ vol. II, pp. 458,19 – 459,1; at Abka¯r al-afka¯r, vol. II, p. 604,6 – 9, A¯midı¯ claims that Ba¯qilla¯nı¯ flip-flopped on the issue of the ahwa¯l, sometimes affirming them and other ˙ to Hayya¯m’s discussion, al-A¯midı¯ groups times denying them. In an apparent reference ˘ are superadded to quiddity: G ˙ a¯yat wug˘u¯d with lawniyya on the basis of the fact that both al-mara¯m p. 32,4 – 6, and Abka¯r al-afka¯r, vol. II, p. 607,5 – 8. Ra¯zı¯’s and Suhrawardı¯’s Mag˙ribı¯ contemporary Ibn Rusˇd (d. 595/1198) says in his Taha¯fut at-Taha¯fut that ‘those who reject modes (al-ahwa¯l) reject the belief in existence in general and color in general, ˙ modes say that existence in general and color in general are whereas those who affirm neither existent nor non-existent’: Taha¯fut at-Taha¯fut, vol. III, ed. M. Bouyges, Beirut, 1930, p. 258,10 – 11; cited by Wolfson, Philosophy of the Kalam, pp. 170 – 171. ˘

˘

˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Robert Wisnovsky

41

Essence and Existence in the Eleventh- and Twelfth-Century

˘

˘

Avicenna’s ontology by holding that existence is something superadded to a thing’s quiddity. This in turn raised the question of whether, on Ra¯zı¯’s account, existence is a real attribute or merely a conceptual attribute, to use Hayya¯m’s terms. Holding that the existence that is superadded to quiddities is itself˘ neither existent or non-existent would invite accusations that Ra¯zı¯ had violated the law of the excluded middle. But explicitly affirming that existence is a real attribute superadded to quiddities, i. e., that the existence that is superadded to quiddities is itself existent, would open Ra¯zı¯ up to the criticism that he had inadvertently posited an infinite regress of existences – the criticism that Suhrawardı¯ was to level, and which the Sunni theory of the attributes’ eternality had been vulnerable to. Without wading into the uncertainty and controversy over the chronology of Ra¯zı¯’s writings, I have selected a significant sample of treatises that are generally assumed to be from his middle period: Sˇarh al-Isˇa¯ra¯t wa-t-tanbı¯ha¯t (Commentary on [Avicenna’s] ‘Pointers and Reminders’),˙ al-Maba¯hit al-masˇriqiyyah (Eastern ˙ ¯ al-Mulahhas fı¯ l-hikma Investigations), Luba¯b al-isˇa¯ra¯t (The Gist of the ‘Pointers’), ˙ ¯n ˙(Forty wa-l-mantiq (Epitome of Philosophy and Logic), al-Arba ¯ın fı¯ usu¯˘l ˘ad-dı ˙ ˙ [Chapters] on the Fundamentals of Religion), and ar-Risa¯lat al-kama¯liyya fı¯ l-haqa¯ iq al-ila¯hiyya (Complete Epistle on Metaphysical Realities).40 These treatises are ˙largely consistent in their ontology.41 In these works, Ra¯zı¯ normally begins his discussions of ontology by stating that existence is univocal, that is, that existence applies with a commonality of meaning (isˇtira¯k ma nawı¯) to God and to the rest of the universe.42 Ra¯zı¯ maintains ˘

40 For a discussion of Ra¯zı¯’s writings and his intellectual development, see A. Shihadeh, From al-Ghaza¯lı¯ to al-Ra¯zı¯, pp. 141 – 79; Shihadeh offers a chronology in his The Teleological Ethics of Fakhr al-Dı¯n al-Ra¯zı¯, pp. 4 – 11. As Shihadeh admits, his chronology is tentative because it is based mainly on internal cross-references in Ra¯zı¯’s writings, references that may have been inserted not during the actual composition of a given work, but at a later date, when Ra¯zı¯ was revising that work. Developments in Ra¯zı¯’s ethics are treated in Shihadeh’s book passim. 41 To my knowledge the only extended discussion of the development of Ra¯zı¯’s ontology is M. A. Zarruka¯n, Fahraddı¯n ar-Ra¯zı¯ wa-a¯ra¯ uhu al-kala¯miyyah wa-l-falsafiyyah, Cairo, 1963, pp. 170 – 74; ˘cited by the editors of two of Ra¯zı¯’s works: Sˇarh al-Isˇa¯ra¯t wa-ttanbı¯ha¯t, ed. Nag˘afza¯deh, pp. 47 – 51 and al-Maba¯hit al-masˇriqiyyah,˙ ed. M.M. al˙ ¯are Sˇarh al-Isˇa¯ra¯t wa-t-tanbı¯ha¯t, Bag˙da¯dı¯, p. 114, n. 1. The passages I am summarizing ˙ and pp. 199,4 – 204,7; no ed. [ad Namat 4: Fı¯ al-wug˘u¯d wa- ilalihi], pp. 190,7 – 192,13 ˙ Luba¯b al-isˇa¯ra¯t, p. 79,7-ult.; al-Maba¯hit al-masˇriqiyyah I.1.2 – 5, pp. 106,ult.–130,10; ˙ ¯ Staatsbibliothek Ms. or. oct. 629, I.2 – 3 (see al-Mulahhas fı¯ l-hikma wa-l-mantiq, Berlin ˙ ˙ ˙ ˘ ˘ H. Eichner’s chapter, ‘Essence and existence. 13th-century perspectives in Arabic-Islamic philosophy’, in this volume, for a translation of these two chapters); al-Arba ¯ın fı¯ usu¯l ˙– ad-dı¯n, pp. 53,22 – 58,20; and ar-Risa¯lat al-kama¯liyya fı¯ l-haqa¯ iq al-ila¯hiyya, pp. 33,5 ˙ 34,19. 42 Although see Asa¯s at-taqdı¯s (The Basis for Glorifying [God]), p. 89,8 – 17, for an attack on the Karra¯miyya’s claim of strong univocity between the visible (asˇ-sˇa¯hid) and the invisible ˘

˘

˘

˘

˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

42

˘

the Kulla¯bite line by arguing, against Avicenna and the Mu tazilites, that God’s existence, like the other divine attributes, cannot simply be subsumed in the divine self. But Ra¯zı¯ also argues, in open opposition this time to Asˇ arı¯, that the existence of each being is distinct from its quiddity. Ra¯zı¯ outlines three possible positions on the relation between existence and quiddity, positions that he takes to be exhaustive. The first possible position is that (a) existence is identical to quiddity (al-wug˘u¯d nafs al-ma¯hiyya or al-wug˘u¯d da¯t al-ma¯hiyya); the second is ¯ that (b) existence is extrinsic to quiddity (al-wug˘u¯d ha¯rig˘ an al-ma¯hiyya); and the ˘ third is that (c) existence is intrinsic to quiddity (al-wug ˘u¯d da¯hil fı¯ l-ma¯hiyya). Ra¯zi assigns position (a) to Asˇ arı¯ (as well as – in Ra¯zı¯’s Arba˘¯ın – to the later Mu tazilite Abu¯ l-Husayn al-Basrı¯); position (b) to Avicenna and the fala¯sifa in ˙ ˙ general;43 and position (c) to no one, it being just a logical possibility.44 Ra¯zı¯ then argues as follows in favor of position (b): i. quiddity (ma¯hiyya) is the principle of difference; ii. existence (wug˘u¯d) is the principle of identity, on the basis of the Avicennian doctrine – which Ra¯zı¯ accepts and promotes – that all beings share existence, or have existence in common (musˇa¯rakat al-wug˘ud); iii. given i and ii, existence will not be identical to quiddity, and hence position (a) is incorrect; iv. given i and ii, existence will not be part of quiddity, and hence position (c) is incorrect; v. given that positions (a), (b) and (c) are the only possible positions, and given that positions (a) and (c) are incorrect, it follows that position (b) is correct, and that existence is therefore extrinsic to quiddity. ˘

˘

˘

˘

˘

˘

˘

(al-g˙a¯ ib). For a brief but helpful sketch of how the distinction between isˇtira¯k ma nawı¯ and isˇtira¯k lafz¯ı maps onto the distinction between univocal and equivocal, see T. Mayer, Fakhr ad-Dı¯n˙ ar-Ra¯zı¯’s Critique, n. 48. 43 It is possible that in articulating position (b) as al-wug˘u¯d ha¯rig˘ an al-ma¯hiyya, Ra¯zı¯ may have thought he was reverting to a classical Avicennian ˘formula, since in addition to appearing in the Ta lı¯qa¯t passage mentioned earlier, the same phrase is used by Avicenna’s favorite student, Bahmanya¯r, in his Tahs¯ıl (viz., fa-inna kawna l-wug˘u¯di ma nan ha¯rig˘an ˙ ani l-ma¯hiyyati arafna¯hu bi-baya¯nin ˙wa-burha ¯ nin), at least according to one˘ of the manuscript copies: Kita¯b at-Tahs¯ıl, p. 285, n. 3, ult. ˙ aza¯lı¯’s Taha¯fut al-fala¯sifa, 44 Ra¯zı¯ might have been thinking˙˙here of Discussion V.16 of G ˙ aza¯lı¯ rehearses the argument that existence cannot be a pp. 151,7 – 152,10, where G constituent of (muqawwim bi-) quiddity. The genus and differentia are in a way parts of the quiddity, given that the quiddity is the species of a thing. Ra¯zı¯ could be taking as absurd the idea that the quiddity or species ‘caused [thing]’ contains the parts wug˘u¯d (‘existence’, i. e., as genus) and mumkin (‘possible’, i. e., as differentia). This is implied in his Sˇarh Uyu¯n al-hikma, pp. 80,1 – 82,12; immediately before that passage ˙ ˙ (p. 77,21–22), Ra¯zı¯ quotes Avicenna as denying that existence is ‘intrinsic to’ (da¯hil fı¯) ˘ essence. In his Epitome of the Metaphysics (I, pp. 34 – 43 and III, pp. 34 – 48), Averroes cites Aristotle, Metaph. 3, as denying that being and unity are parts of the being of a thing. ˘

˘

˘

˘

˘

˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Robert Wisnovsky

43

Essence and Existence in the Eleventh- and Twelfth-Century

˘

˘

Ra¯zi further asserts that the correct way to articulate position (b) – the position that existence is extrinsic to quiddity – is that existence is superadded to quiddity (za¯ id ala¯ l-ma¯hiyya). Ra¯zı¯ clearly sees himself as extrapolating from, and in some sense systematizing, Avicenna’s ontology. In his Commentary on the ‘Pointers’ I.11, Ra¯zı¯ interprets the term muda¯f (‘related [to]’), which Avicenna had used in the passage being commented˙ upon to describe how existence is connected to quiddity, in a stronger and less neutral sense, as ‘superadded’:

˘

˘

His [viz., Avicenna’s] statement ‘Like humanity; for in itself it is a quiddity …’, up to the end of the chapter: Know that in defining the essential, he needs to distinguish between the constituents of quiddity and the constituents of existence, so that neither will be confused with the other. This is attained only by explaining that existence is distinct from quiddity and that it is superadded to it [my italics] (anna l-wug˘u¯da mug˙a¯yirun li-l-ma¯hiyyati wa-annahu za¯ idun alayha¯). He argues for this by [citing the fact] that quiddity may be known at a time when its existence is in doubt; given that the known is other than what is in doubt, existence is superadded to quiddity.45

˘

Given Ra¯zı¯’s hard line on the univocity of existence, and given his apparent desire to oppose Mu tazilite theology by placing himself in the Kulla¯bite tradition of holding that the divine attributes are somehow meaningfully distinct from the divine self, we should not be surprised by his argument that in the case of God, as with other beings, existence is also distinct from and superadded to quiddity. But in contrast to a contingent being, which needs a cause to bring its quiddity and existence together, God’s quiddity is sufficient to cause its own existence.46 It is worth noting that Ra¯zı¯’s position is in direct opposition to the doctrine of his school founder, Asˇ arı¯. Asˇ arı¯ held that in created beings, quiddity and existence were intensionally as well as extensionally identical; and by some accounts Asˇ arı¯ also treated the attribute of existence as an exception to Kulla¯bite attribute-theory, by maintaining that in the case of God, too, quiddity and existence were intensionally as well as extensionally identical. Ra¯zı¯’s departure from Asˇ arı¯ is perhaps less surprising than it might first appear, since in the 12th century, when Ra¯zı¯ was active, the strong identification of essence and existence was associated with a newly formed branch of the Mu tazilites as well as with Asˇ arı¯. These Mu tazilites were the followers of Abu¯ l-Husayn al-Basrı¯ (d. 436/ ˙ ˙ Ibn al1044), most prominent of whom was the Hwa¯razmian mutakallim ˘ ˘

˘

˘

˘

˘

˘

˘

˘

45 Ra¯zı¯, Sˇarh al-Isˇa¯ra¯t, ed. Nag˘afza¯deh, I.11, p. 53,7 – 11. 46 Here Ra¯zı˙¯ may be appropriating the distinction Avicenna made in Kita¯b asˇ-Sˇifa¯ /Ila¯hiyya¯t, I.6, pp. 38,17–39,4, between a quiddity that is sufficient for a thing to exist and a quiddity that is not sufficient for a thing to exist (the former will be uncaused, the latter caused). In any case Ra¯zı¯ is not entirely consistent; he says in his Sˇarh Uyu¯n al-hikma ˙ (p. 76,14–15) that ‘we have discussed in our other books decisive proofs that ˙God’s essence (haqı¯qa) is identical to His existence (wug˘u¯d)’. ˙ ˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

44

˘

Mala¯himı¯ (d. 532/1144). These Mu tazilites were openly opposed to Abu¯ ˙ Ha¯ˇsim’s ha¯l-theory, at least as it was construed by its most famous proponent, ˙ Qa¯d¯ı Abdalg˘abba¯r (d. 415/1025), and they subjected it to many of the same ˙ critiques that Hayya¯m and Sˇahrasta¯nı¯ and others had directed at the ˘ such as Ba¯qilla¯nı¯ and G ˘ uwaynı¯.47 Bahsˇamizing Sunnis What might Ra¯zı¯’s motivations have been for embracing Avicenna’s ontology in the case of contingent beings, while rejecting it in the case of God? Ra¯zı¯ may have been unconvinced by Avicenna’s radical understanding of divine simplicity, which Avicenna safeguards by completely identifying essence and existence in God. For Avicenna, God’s simplicity, and thus His noncompositeness, ensures that He is the only uncaused being, in contrast to all other beings, whose composition of essence and existence requires a composer and hence a cause. For Avicenna, in other words, God’s simplicity explained His causal self-sufficiency, His immunity from causedness. But God’s causal productivity, His causation of other beings, was another matter. At most, it could be said that on Avicenna’s account, God’s being wa¯g˘ib al-wug˘u¯d (necessary of existence) conveyed a weak form of transitivity that was perhaps implicit in the active participle wa¯g˘ib (necessary). But even if we were to construe wa¯g˘ib as synonymous with the more clearly transitive fourth-form active participle mu¯g˘ib (necessitating), the kind of transitivity conveyed by w-g˘-b remained one of syllogistic necessitation rather than one of causal production – let alone one of voluntary agency. And God’s necessitating the world just as a true premise necessitates the conclusion that follows from it, was not a robust enough notion of divine causation for Ra¯zı¯. Avicenna’s theology immunized God from causedness very convincingly. But it made God’s causation of the world too automatic, too mechanistic. By contrast, holding – as Ra¯zı¯ does – that God’s quiddity is sufficient to cause His existence, allowed God to retain His attribute of will (ira¯da) and thereby be plausibly seen as a voluntary agent (fa¯ il) rather than as a necessitator. In sum, while Ra¯zı¯ adopts Avicenna’s ontology, he rejects Avicenna’s theology by maintaining that God’s quiddity is not identical to His existence. With this Ra¯zı¯an background in our minds, we can now turn back to Suhrawardı¯, from whence we started, and examine, with fresh eyes, his critique of the notion of existence put forward by the ‘followers of the Peripatetics’. To recapitulate: given that Avicenna was not explicitly committed to the thesis that ˘

˘

˘

47 See Ibn al-Mala¯himı¯, Kita¯b al-Fa¯ iq fı¯ usu¯l ad-dı¯n, pp. 46,3 – 48,ult. (Fı¯ anna wug˘u¯da ˇs-sˇay i hal huwa ˙ da¯tuhu aw ha¯latun za¯˙idatun ala¯ da¯tihi) and pp. 68,10 – 76,15 (Fı¯ ¯ ¯ a¯nin wa-la¯ ahwa¯lin); and Tuhfat annahu ta a¯la¯ qa¯dirun a¯limun˙hayyun li-da¯tihi la¯ li-ma ¯ ˙ ˙ by Samarqandı ˙¯ as al-mutakallimı¯n fı¯ r-radd ala¯ l-fala¯sifa, pp. 61,19 – 62,19. Asˇ arı¯ is cited holding the view that in God essence and existence are identical: as-Saha¯ if al-Ila¯hiyya, ˙˙ ˙ p. 298,5–6. ˘

˘

˘

˘

˘

˘

˘

˘

˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Robert Wisnovsky

45

Essence and Existence in the Eleventh- and Twelfth-Century

˘

˘

existence is something superadded to (za¯ id ala¯) a thing’s quiddity (even though Avicenna could plausibly be construed as implying it), whose doctrine might Suhrawardı¯ have been pointing to here? In both his Talwı¯ha¯t (at least as interpreted by his commentator Ibn ˙ and Hikmat al-isˇra¯q (at least as understood by the Kammu¯na [d. 676/1277]) ˙ baddın al-Sˇırazı [d. 710/1310]), Suhrawardı commentators Sˇahrazu¯rı¯ and Qut ¯ ¯¯¯ ¯ ˙ ¯’s old position, namely, that existence is accepts the incorrectness of Asˇ arı indistinguishable from essence or quiddity either in concrete reality (i. e., extensionally) or in the mind (i. e., intensionally). However, Suhrawardı¯ insists on the incorrectness of the position of ‘the followers of the Peripatetics’, namely, that existence is distinguishable from essence or quiddity both in concrete reality and in the mind. By Suhrawardı¯’s reckoning, essence and existence are distinguishable in the mind but not in concrete reality. This sounds a lot like Avicenna’s position that essence and existence are intensionally distinct but extensionally identical, that essence and existence can be separated conceptually while remaining locked together in actual beings. Where Suhrawardı¯ goes beyond Avicenna is in inferring from this that existence as well as quiddity are merely i tiba¯ra¯t aqliyya, ‘mental constructs’ or ‘expressions of the intellect’. According to Suhrawardı¯, the position of the followers of the Peripatetics suffers from an infinite-regress problem: If existence really existed concretely, it would itself have to be a concrete existent; otherwise the existence would be non-existent, and a contradiction would ensue. But if existence were itself a concrete existent, it would exist through yet another existence; and this would lead to an infinite regress of existences. Suhrawardı¯ and his commentators reason that, given this and other weaknesses, existence cannot be a real attribute of concrete things, but must instead be understood as a mental construct. Existence is therefore not something really superadded to quiddity in order to make a concretely existing being. It seems clear that Suhrawardı¯ and Hayya¯m were part of the same line of ˘ put forward by Sunni mutakallimu¯n critics of the Avicennian theory of existence ˘ such as Guwaynı¯ and Ra¯zı¯. Not only does Suhrawardı¯, like Hayya¯m, deny that ˘ reality; he later existence is an attribute superadded to quiddity in concrete reproduces Hayya¯m’s distinction between attributes that are merely conceptual ˘ that are real as well as conceptual.48 In a brief passage in his and attributes Talwı¯ha¯t, Suhrawardı¯ also attacks the theory of modes, on the basis of its ˙ of the law of the excluded middle; and the example he gives of a violation mental construct (i tiba¯r dihnı¯) that falls into this trap is ‘colorness’ (lawniyya) – the same example that ¯Hayya¯m had given before him. Finally, Suhrawardı¯ ˘ ˘

˘

˘

˘

˘

˘

48 fa-idani s-sifa¯tu kulluha¯ tanqasimu ila¯ qismayni sifatin ayniyyatin wa laha¯ su¯ratun fı¯ l- aqli ¯ ifatin ˙ ˙ wug˘u¯duha¯ fı¯ l- ayni laysa illa¯ nafsu˙wug˘u¯diha¯ fı¯ d-dihni wa-laysa ˙ laha¯ fı¯ g˙ayri … wa-s ¯ ¯ ˙ d-dihni wug˘u¯dun, Hikmat al-isˇra¯q I.3.68, pp. 50,5 – 51,5. ¯ ¯ ˙ ˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

46

˘

˘

˘

˘

famously places his main attack on the conception of existence put forward by ‘the followers of the Peripatetics’, in the section of his Hikmat al-isˇra¯q that is ˙ too may have been devoted to exposing sophistries and logical fallacies. This anticipated by Hayya¯m, given the prominent appearance of hadaya¯n (‘bab¯ 49 Avicenna ˘ art in Arabic sophistics, in Hayya¯m’s discussion. bling’), a term of ˘ also uses h-d-y (specifically, ma¯ hadaw bihi min aqa¯wı¯lihim) in a well known ¯ Ila¯hiyya¯t I.5 of his¯ Sˇifa¯ , when referring to the Mu tazilites’ passage from inconsistency on the issue of whether a non-existent (ma du¯m) is a thing (sˇay ).50 These Avicennian uses of h-d-y, and particularly that contained in the ¯ Ila¯hiyya¯t I.5 passage, are possible antecedents to Hayya¯m’s use of hadaya¯n in his ¯ ˘ to have a common Risa¯lat al-Wug˘u¯d. For Hayya¯m and Avicenna appear target: ˘ those who commit some sophism by, in this case, violating the law of the excluded middle. In Avicenna’s case, it is the early Mu tazilites whose ‘nonexistent thing’ neither is nor is not. In Hayya¯m’s case, it is the Bahsˇamizing Asˇ arites, whose position that existence is˘ superadded to quiddity entails that existence be construed as a mode (ha¯l) (or more precisely, the grounds [ma na¯] ˙ of a mode) that falls between existence and non-existence. The targets are different, but the sophism they commit is ultimately the same. This apparent link between Avicenna and Hayya¯m; the apparent link between Hayya¯m and ˘ Suhrawardı¯ discussed earlier;˘ and the basic similarity between Avicenna’s and Suhrawardı¯’s distinctions between essence and existence; all increase the likelihood that, as I hypothesized at the beginning of this chapter, Suhrawardı¯ was targeting the Avicennian ontology of mutakallimu¯n such as Ra¯zı¯, rather than Avicenna’s own ontology; and that Suhrawardı¯ therefore intended to make an implicit distinction when he criticized the followers of the Peripatetics (atba¯ alMasˇˇsa¯ ¯ın), and not the Peripatetics themselves, for holding that existence is something superadded to quiddity (ma nan za¯ idun ala¯ l-ma¯hiyya). Interestingly, in a late work of his, the Niha¯yat al- uqu¯l, Ra¯zı¯ himself appears to have struck back, targeting Hayya¯m and Suhrawardı¯ by recapitulating Suhrawardı¯’s infinite regress argument˘ against existence’s being something superadded to quiddity, classifying it as a type of sophism used by radical skeptics who deny existence, ˘

˘

˘

˘

˘

˘

˘

˘ ˘

49 Risa¯la fı¯ l-wug˘u¯d 401,11. H-d-y is one of a cluster of roots (the others are h-d-r, h-t-r, h-g˘¯ r, h-m-z and h-m-r) used to render Aristotle’s term !dokeswe?m, for example¯ in Yahya¯ ibn ˙ Adı¯’s translation of Aristotle, Sophistical Refutations (SE 3:165b15): see, ad loc., Mant iq Aristu¯, vol. III, ed. A. Badawı¯. In the corresponding section of the Safsata (Sophistics)˙of ˙ his ˙Sˇifa¯ (I.1, 7,5), Avicenna paraphrases Aristotle but uses the terms al-had aya¯n wa-t¯ takrı¯r; Kita¯b asˇ-Sˇifa¯ /Mantiq (6): as-Safsata, ed. A.F. al-Ahwa¯nı¯; see also SE 13:173a31 – ˙ a II.2: pp. 67,12 ˙ 173b16, and Ibn Sı¯na¯, Safsat – 69,5, where hadaya¯n and hada¯ appear at ¯ I.1, p. 9,1. ¯ ˙ ¯ b asˇ-Sˇifa¯ /Tabı¯ iyya¯t (6): an-Nafs, 68,1.2 tris.3.7.9 bis.10; cf. Kita ˙ ˇ 50 Ibn Sı¯na¯, Kita¯b asˇ-Sifa¯ /Ila¯hiyya¯t (1), p. 33,16 – 18. ˘

˘

˘

˘

˘

˘

˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Robert Wisnovsky

47

Essence and Existence in the Eleventh- and Twelfth-Century

˘

and criticizing those skeptics for recklessness (ta assuf ) – the same flaw that Hayya¯m had seen in the Bahsˇamizing Sunni mutakallimu¯n. 51 ˘

Conclusion

˘

The two opposing positions on this issue largely determined the metaphysical doctrines of the following century. Ra¯zı¯’s Avicennian understanding of existence as something superadded to quiddity is embraced in toto by as-Samarqandı¯ (fl. ca. 690/1291) and al-Bayda¯wı¯ (d. ca. 716/1312), who follow Ra¯zı¯ in ontology as well as in theology; and˙ Ra¯zı¯’s ontology (though not his theology) is accepted by al-Qazwı¯nı¯ al-Ka¯tibı¯ (d. 675/1276) as well as by al-Abharı¯ (663/1264), at least during one phase of his career.52 As was seen above, Suhrawardı¯’s commentators Ibn Kammu¯na, Sˇahrazu¯rı¯ and Qutbaddı¯n Sˇ¯ıra¯zı¯ all endorse ˙ Suhrawardı¯’s view. But it would be a mistake to label this simply as the ‘Isˇra¯qı¯’ ˇ view, given what appear to be Hayya¯m’s and Sahrasta¯nı¯’s roles in propelling it ˘ along, as well as the fact that prominent non-Isˇra¯qı¯s also rejected the position that existence is superadded to quiddity. These include the Mu tazilite followers of Abu¯ l-Husayn al-Basrı¯, such as Ibn al-Mala¯himı¯, as mentioned above. They ˙ ˙ ˙ also include Ra¯zı¯’s and Suhrawardı¯’s Mag˙ribı¯ contemporary, Averroes.53 The

˘

˘

˘

˘

51 Ra¯zı¯, Niha¯yat al- uqu¯l wa-dira¯yat al-usu¯l, Istanbul Ayasofya ms 2376, fols 22r9–v24. ˙ 52 See as-Samarqandı¯, as-Saha¯ if al-Ila¯hiyya, I.1.1.2 (Fı¯ anna l-wug˘u¯d za¯ idun ala¯ l˙ awa ˙ ¯ li al-anwa¯r min mata¯li al-anza¯r I.1.2.2 (Fı¯ kawnihi [viz., ma¯hiyya¯t); al-Bayda¯wı¯,˙ T ˙ I.1.2.3 (Fı¯ kawnihi [viz., ˙ al-wug˙˘u¯d] za¯ idan hila¯fan li-sˇ-Sˇayh ˙ and al-wug˘u¯d] musˇtarak) ˘ ˘ [viz., al-Asˇ arı¯] mutlaqan wa-l-Hukama¯ fı¯ l-wa¯g˘ib); but note I.1.4.1 (Fı¯ annaha ¯ [viz., alwug˘u¯b, al-imka¯n, ˙al-qidam and˙ al-hudu¯t] umu¯r aqliyya la¯ wug˘u¯da laha¯ fı¯ l-ha¯rig˘). Al˙ ¯ (Fı¯ anna l-wug˘u¯d musˇtarak); I.1.1.3 ˘(Fı¯ anna lQazwı¯nı¯ al-Ka¯tibı¯, Hikmat al- ayn I.1.1.2 ˙ wug˘u¯d za¯ id ala¯ ma¯hiyya¯t al-mumkina¯t) and I.1.1.2 (Fı¯ anna l-wug˘u¯d nafs haqı¯qat al˙ ˇtarakun wa¯g˘ib al-wug˘u¯d). Al-Abharı¯, Kasˇf al-haqa¯ iq 1 (Fı¯ kawnihi [viz., al-wug˘u¯d] mus ˙ bayna l-mawg˘u¯da¯t) and 2 (Fı¯ anna wug˘u¯d al-mumkina¯t za¯ id ala¯ ma¯hiyya¯tiha¯); but note his opposite view in Muntaha¯ l-afka¯r fı¯ iba¯nat al-afka¯r I.1.1.1 (al-masˇhu¯r anna l-wug˘u¯d amr musˇtarik min g˘amı¯ al-mawg˘u¯da¯t … wa-l-kull da ¯ıf ) and I.1.1.2 (al-masˇhu¯r anna wug˘u¯d al-mumkina¯t fı¯ l-a ya¯n za¯ id ala¯ ma¯hiyya¯tiha˙¯ … wa-huwa da ¯ıf ). For a brief ˙ discussion of the major 13th-century developments in Avicennian ontology, along with translations of key passages, see Eichner, Essence and existence. 13th-century perspectives in Arabic-Islamic philosophy. 53 fa-inna-ma¯ buniya l-qawlu fı¯ha¯ ala¯ madhabi bni Sı¯na¯ wa-huwa madhabun hata un wa¯ ˙ ˇsay un da¯lika annahu ya taqidu anna l-anniyyata wa-huwa kawnu ˇs-sˇay ¯i mawg˘u˘¯ dan ¯za¯ idun ala¯ l-ma¯hiyyati ha¯rig˘u n-nafsi … amma¯ qawlu l-qa¯ ili inna l-wug˘u¯da amrun za¯ idun ala¯ l-ma¯hiyyati …˘ fa-qawlun mug˙allatun g˘iddan … wa-huwa madhabu bni Sı¯na¯, ¯ see Averroes, ˙ Averroes, Taha¯fut at-Taha¯fut, pp. 302,13 – 304,14; cf. p. 197,15 – 16. Also Tafsı¯r ma¯ ba da t- tabı¯ a, pp. 1279,12 – 1280,11. Averroes’ critique of Avicenna’s ontology ˙ ˙by S. Menn in his Fa¯ra¯bı¯ in the Reception of Avicenna’s Metaphysics: is discussed fully Averroes against Avicenna on Being and Unity, in this volume. ˘

˘

˘

˘

˘

˘

˘

˘

˘

˘

˘

˘

˘

˘

˘

˘

˘

˘

˘

˘

˘

˘ ˘

˘

˘

˘

˘ ˘

˘

˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

48

˘

only difference between Suhrawardı¯ and Averroes is that Averroes assigns this position to Avicenna himself, while Suhrawardı¯ associates it to unspecified ‘followers of the Peripatetics’. The Mag˙ribı¯ philosophical mystic Ibn Arabı¯ (d. 638/1240) similarly condemns construing existence as something superadded; although to be precise, what Ibn Arabı¯ denies is that existence is superadded to the existent, not that existence is superadded to the essence.54 It remains an open question whether any of the critics of what I have labeled Avicennian ontology felt that Avicenna’s own theory of existence could be salvaged if it had been interpreted in a different way. However fertile they doubtless were to subsequent Muslim thinkers, Avicenna’s various articulations of the essence-existence distinction, and his attempts to press that distinction into the service of his distinction between the necessary of existence in itself and the necessary of existence through another, were perhaps so tentative and inconsistent that they resisted systematization. Suhrawardı¯’s creation of a new metaphysics of light might have partly resulted from his frustration with the flaws that emerged from recent and ongoing attempts to systematize Avicennian ontology. ˘

Bibliography ˘

˘ abba¯r, al-Mug˙nı¯ fı¯ abwa¯b at-tawh¯ıd wa-l- adl, eds M.M. Hilmı¯ et al., Cairo: alAbd al-G ˙ ˙ f. Mu assasa al-Mis. riyya al- A¯mma, 1958 A. Alami, L’ontologie modale: tude de la thorie des modes d’Abu¯ Ha¯shim al-Jubba¯ ¯ı, Paris: Vrin, 2001. ˙ a¯yat al-mara¯m fı¯ ilm al-kala¯m, ed. A.F. al-Mazı¯dı¯, Beirut: Da¯r al-Kutub alal-A¯midı¯, G Ilmiyya, 2004. –– , Abka¯r al-afka¯r fı¯ usu¯l ad-dı¯n, vol. II, ed. A.F. al-Mazı¯dı¯, Beirut, 2004. al-Asˇ arı¯, Maqa¯la¯t al-isla¯˙miyyı¯n, I-II, ed. H. Ritter, Istanbul: Devlet, 1929 – 30. al-Bayhaqı¯, Tatimmat siwa¯n al-hikma, ed. M. Sˇafı¯ , Lahore: L. Ishwar Das, Registrar, ˙ ˙ University of the Panjab, 1935. al-Ba¯qilla¯nı¯, Kita¯b at-Tamhı¯d, ed. R.J. McCarthy, Beirut: al-Maktaba asˇ-Sˇarqiyya, 1957. ˙ aza¯lı¯, Taha¯fut al-fala¯sifa, ed. M. Bouyges, Beirut: Imprimerie Catholique, 1927. al-G ˘ uwaynı¯, asˇ-Sˇa¯mil fı¯ usu¯l ad-dı¯n, ed. A.M.M. Umar, Beirut, 1999. al-G ˙ ed. M.Z. al-Kawtarı¯, Cairo: Matba at al-Anwa¯r, 1948. –– , al- Aqı¯da an-niza¯miyya, ˙ ¯, Textes apologe´tiques de ˙ ˘ uwaynı –– , Luma fı¯ qawa¯ id ahl as-Sunna wa-l-g˘ama¯¯ a, in al-G ˇ Guwaynı¯, ed. M. Allard, Beirut: Da¯r al-Masˇriq, 1968. –– , al-Irsˇa¯d fı¯ usu¯l al-i tiqa¯d, transl. and ed. J. Luciani, Paris: Leroux, 1938. ˙ M. Allard, Le problme des attributes divins dans la doctrine d’al-Ash arı¯ et de ses premiers grands disciples, Beirut: Imprimerie Catholique, 1965. ˘

˘

˘

˘

˘

˘

˘

˘

˘

˘

˘

˘

˘

˘

˘

˘

˘

˘

˘

˘

˘

˘

˘

54 fa- lam anna l-wug˘u¯da wa-l- adama laysa¯ bi-sˇay in za¯ idin ala¯ l-mawg˘u¯di wa-l-ma du¯mi lakinna huwa nafsu l-mawg˘u¯di wa-l-ma du¯mi lakinna l-wahma yatahayyalu anna l˘ Arabı¯, Kita¯b wug˘u¯da wa-l- adama sifata¯ni ra¯g˘i ata¯ni ila¯ l-mawg˘u¯di wa-l-ma du¯mi, Ibn ˙ – 17. Insˇa¯ ad-dawa¯ ir, p. 6,14 ˘

˘

˘

˘

˘

˘

˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Robert Wisnovsky

49

Essence and Existence in the Eleventh- and Twelfth-Century

˘

˘

˘

˘

al-Qusˇayrı¯, Sˇarh asma¯ Alla¯h al-husna¯, eds T. A. Sa d and S.H.M. Alı¯, Cairo: Da¯r al˙ ˙ ˙ ¯t, 2001. ˙ Haram li-t-Tura ˙ ¯ al-Ra¯zı¯, al-Arba ¯ın fı¯ usu¯l ad-dı¯n, no ed., Hyderabad: Da¯ irat al-Ma a¯rif al- Utma¯niyya, ˙ 1934 – 35. –– , al-Maba¯hit al-masˇriqiyyah, vol. 1, ed. M.M. al-Bag˙da¯dı¯, Beirut: Da¯r al-Kita¯b al˙¯ Arabı¯, 1990. –– , ar-Risa¯lat al-kama¯liyya fı¯ l-haqa¯ iq al-ila¯hiyya, ed. A. Muhyuddı¯n, Beirut: Da¯r al˙ Kutub al- Ilmiyya, 2002. ˙ –– , Asa¯s at-taqdı¯s, ed. A.H. Saqqa¯, Cairo: Maktabat al-Kulliyya¯t al-Azhariyya, 1986. ˙ Atiyah, Cairo, 1936. –– , Luba¯b al-isˇa¯ra¯t, ed. A.S. ˇ –– , Sarh al-isˇa¯ra¯t wa-t-tanbı¯ha¯t,˙ ed. A.R. Nag˘afza¯deh, Tehran, 1384 H. –– , Sˇarh˙ al-Isˇa¯ra¯t wa-t-tanbı¯ha¯t, no ed., in Kita¯b ˇsarhay al-isˇa¯ra¯t li-l-Hwa¯g˘a Nas¯ıraddı¯n ˙ ˙ ˘ at-T˙u¯sı¯ wa-l-Ima¯m Fahraddı¯n ar-Ra¯zı¯, Qum, 1983/4. ˙ ˙ ˘ ˇ –– , Sarh Uyu¯n al-hikma, ed. A. as-Saqqa¯, Tehran, 1994. ˙ u¯ , ed. A. Badawı¯, vol. III, Cairo: Maktabat an-Nahda alAristotle,˙ Mantiq Arist . ˙ Mis. riyya, 1952. ˙ asˇ-Sˇahrasta¯nı¯, Kita¯b al-Musa¯ra a, ed. W. Madelung, transl. T. Mayer, London: Tauris, ˙ 2001. asˇ-Sˇahrasta¯nı¯ and ¯Ila¯qı¯, Guftgu¯-yi Sˇahrasta¯nı¯ va-I¯la¯qı¯, ed. M.T. Da¯nisˇpazhu¯h, Na¯ma-yi A¯sta¯n-i Quds 3/9, 1348 H., pp. 97 – 104. asˇ-Sˇahrazu¯rı¯, Rasa¯ il asˇ-sˇag˘ara al-ila¯hiyya, vol. III, ed. N. Habı¯bı¯, Tehran: Mu assasa-i ˙ Pazˇu¯hisˇ¯ı-i Hikmat wa Falsafa-i ¯Ira¯n, 1385 H. ˙¯ b at-Tahs¯ıl, ed. M. Mutahharı¯, Tehran: Da-nishkadah- i Ila-hı¯ya-t wa Bahmanya¯r, Kita ˙˙ H. ˙ Ma a-rif-i Isla-mı¯, 1349 R. Frank, Abu Hashim’s Theory of ‘States’: Its Structure and Function, in Actas do quarto congresso de estudios rabes e islmicos, Coimbra-Lisboa, 1 a 8 de setembro de 1968, Leiden: Brill, 1971, pp. 85 – 100. –– , Beings and Their Attributes: The Teaching of the Basrian School of the Mu tazila in the Classical Period, Albany, New York: State University of New York Press, 1978. Umar Hayya¯m, Daru¯riyyat at-tada¯dd fı¯ l- a¯lam wa-l-g˘abr wa-l-baqa¯ , in S.G. Tı¯rtha, ˙ The Nectar of˙Grace: Omar Khayya¯m’s Life and Works, Hyderabad: ˘ and ed., transl. Kitabistan, 1941, pp. xcix–civ. –– , Risa¯lat al-kawn wa-t-taklı¯f, in S.G. Tı¯rtha, transl. and ed., The Nectar of Grace: Omar Khayya¯m’s Life and Works, Hyderabad: Kitabistan, 1941, pp. lxxxiii–lxxxix. –– , Risa¯la fı¯ l-wug˘u¯d, in Da¯nisˇna¯ma-yi hayya¯mı¯, ed. R. Rida¯za¯da-yi Malik, Tehran, ˙ ˘ 1377 H., pp. 398,9 – 403,10. –– , Risa¯la fı¯ l-wug˘u¯d, in S. G. Tı¯rtha, transl. and ed., The Nectar of Grace: Omar Khayya¯m’s Life and Works, Hyderabad: Kitabistan, 1941, pp. cx-cxiv. Ibn al-Mala¯himı¯, Kita¯b al-Fa¯ iq fı¯ usu¯l ad-dı¯n, ed. W. Madelung and M. McDermott, ˙ wa Falsafa-i ¯Ira¯n, 2007. Tehran:˙Mu assasa-i Pazˇu¯hisˇ¯ı-i .Hikmat –– , Tuhfat al-mutakallimı¯n fı¯ r-radd ala¯ l-fala¯sifa, ed. H. Ansari and W. Madelung, ˙ Tehran, 2008. Ibn Arabı¯, Kita¯b Insˇa¯ ad-dawa¯ ir, in Kleinere Schriften des Ibn al- Arabı¯, ed. H. Nyberg, Leiden: Brill, 1919. Ibn Rusˇd, On Aristotle’s ‘Metaphysics’: An Annotated Translation of the So-called ‘Epitome’, ed. R. Arnzen, Berlin: De Gruyter, 2010. –– , Tafsı¯r ma¯ ba da t- tabı¯ a, ed. M. Bouyges, 3 vols, Beirut: Imprimerie Catholique, ˙ ˙ 1938 – 48. –– , Taha¯fut at-Taha¯fut, ed. M. Bouyges, Beirut: Imprimerie Catholique, 1930. ˘

˘

˘

˘

˘

˘

˘

˘

˘

˘

˘

˘

˘

˘

˘

˘

˘

˘

˘

˘

˘

˘

˘

˘

˘

˘

˘

˘

˘

˘

˘

˘

˘

˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

50 ˘

˘

˘

Ibn Sı¯na¯, al-Mabda wa-l-ma a¯d, ed. A. Nu¯ra¯nı¯, Tehran: The Institute of Islamic Studies, McGill University, Tehran Branch, 1984. –– , al-Muba¯hata¯t, ed. M. Bı¯da¯rfar, Qum: Intisˇa¯ra¯t Bı¯da¯r, 1371 H. –– , at-Ta lı¯qa¯˙t,¯ed. H.M. Al- Ubaydı¯, Baghdad, 2002 ˙ –– , Kita¯b al-Isˇa¯ra¯t wa-t-tanbı ¯ha¯t, ed. J. Forget, Leiden: Brill, 1892. ˇ ˇ –– , Kita¯b as-Sifa¯ /Ila¯hiyya¯t (1), eds G. Anawa¯tı¯ and S. Za¯yid, Cairo: al-Hay’a al-‘a¯mma li-sˇu‘u¯n al-mata¯bi‘ al-amı¯riyya, 1960. ˙ ¯ hiyya¯t (2), eds M.Y. Mu¯sa¯, S. Dunya¯ and S. Za¯yid, Cairo: al-Hay’a –– , Kita¯b asˇ-Sˇifa¯ /Ila ˇ al-‘a¯mma li-su‘u¯n al-mata¯bi‘ al-amı¯riyya, 1960. ˙ al-Madhal I.2, eds G. Anawa¯tı¯, M. al-Hudayrı¯ and F. al–– , Kita¯b asˇ-Sˇifa¯ /Mantiq (1): ˙ ˘ ˙ Ahwa¯nı¯, Cairo: al-Matba‘a al-amı¯˘riyya, 1952. ˙ ˇ ˇ –– , Kita¯b as-Sifa¯ /Mantiq (6): as-Safsata, ed. A.F. al-Ahwa¯nı¯, Cairo, 1958. ˙ Avicenna’s De anima; Being the Psychological –– , Kita¯b asˇ-Sˇifa¯ /Tabı¯˙iyya¯t (6): an-Nafs. ˙ ˇ Part of Kita¯b al-Sifa¯ , ed. F. Rahman, London: Oxford University Press, 1959. T. Izutsu, The Distinction between essentia and existentia, in The Fundamental Structure of Sabzawari’s Metaphysics, in (viz., the introduction to) Sabzawa¯rı¯, Sˇarh-i ˙ manzu¯mah, eds M. Muhaqqiq and T. Izutsu, Tehran, 1969, pp. 49 – 70. ˙¯zı¯’s Critique of Ibn Sı¯na¯’s Argument for the Unity of God T. Mayer,˙ Fakhr ad-Dı¯n ar-Ra in the Isˇa¯ra¯t, and Nas¯ır ad-Dı¯n at-Tu¯sı¯’s Defence, in D. Reisman, ed., Before and ˙ ˙ ˙ 2002, pp. 199 – 218. After Avicenna, Leiden/Boston: Brill, S. Rizvi, An Islamic Subversion of the Existence-Essence Distinction? Suhrawardı¯’s Visionary Hierarchy of Lights, Asian Philosophy 9/3, 1999, pp. 219 – 27. A. Shihadeh, The Teleological Ethics of Fakhr al-Dı¯n al-Ra¯zı¯, Leiden/Boston: Brill, 2006. –– , From al-Ghaza¯lı¯ to al-Ra¯zı¯: 6th/12th Century Developments in Muslim Philosophical Theology, Arabic Sciences and Philosophy 15/1, 2005, pp. 141 – 79. Suhrawardı¯, Hikmat al-isˇra¯q, transls and eds J. Walbridge and H. Ziai, Provo, Utah: Brigham ˙Young University Press, 1999. J. Walbridge, The Science of Mystic Lights: Qutb al-Dı¯n Sˇ¯ıra¯zı¯ and the Illuminationist ˙ Massachusetts: Harvard University Tradition in Islamic Philosophy, Cambridge, Press, 1992. R. Wisnovsky, Avicenna and the Avicennian tradition, in P. Adamson and R. Taylor, eds, The Cambridge Companion to Arabic Philosophy, Cambridge: Cambridge University Press, 2005, pp. 92 – 136. –– , Avicenna’s Metaphysics in Context, Ithaca, New York: Cornell University Press, 2003. –– , Final and Efficient Causality in Avicenna’s Cosmology and Theology, Quaestio: The Yearbook of the History of Metaphysics 2, 2002, pp. 97 – 123. –– , Notes on Avicenna’s Concept of Thingness (sˇay iyya), Arabic Sciences and Philosophy 10/2, 2000, pp. 181 – 221. –– , One Aspect of the Avicennian Turn in Sunni Theology, Arabic Sciences and Philosophy 14/1, 2004, pp. 65 – 100. H. Wolfson, The Philosophy of the Kalam, Cambridge, Massachusetts: Harvard University Press, 1976. ˘

˘

˘

˘ ˘

˘

˘

˘

˘

˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Robert Wisnovsky

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Fa¯ra¯bı¯ in the Reception of Avicenna’s Metaphysics: Averroes against Avicenna on Being and Unity Stephen Menn It may seem peculiar to describe Fa¯ra¯bı¯ as part of the reception-history of Avicenna’s metaphysics, given that Fa¯ra¯bı¯ died before Avicenna was born. But I’ll try to show that Fa¯ra¯bı¯ became part of that reception-history, because Averroes made him part of that reception-history.1 Averroes is harshly critical of Avicenna’s attempt to rewrite Aristotle’s Metaphysics in what Avicenna thinks is the more appropriate logical order, systematically developing the central ontological concepts and making the theological conclusions depend on properly ontological rather than physical demonstrations: Averroes thinks that Avicenna’s improvements are disimprovements which weaken the demonstrative ˙ aza¯lı¯’s force of Aristotle’s arguments, and leave the philosophers open to G criticisms. Now on some issues Averroes groups Fa¯ra¯bı¯ and Avicenna together, as deviating from Aristotle and his ancient interpreters; but on some central ontological issues Averroes chooses to follow Fa¯ra¯bı¯ against Avicenna, and indeed takes Fa¯ra¯bı¯ to have diagnosed Avicenna’s errors before Avicenna made them. Avicenna has of course great respect for Fa¯ra¯bı¯, and credits Fa¯ra¯bı¯ with showing him the object [sjop|r, g˙arad] of metaphysics as a science and of ˙ Aristotle’s Metaphysics as a treatise; so it will be particularly effective if Averroes can show that Fa¯ra¯bı¯ had already warned against the errors Avicenna was to make, and that Avicenna had disregarded his warning. And in the Epitome of the Metaphysics, Taha¯fut al-Taha¯fut, and Great Commentary on the Metaphysics, Averroes will try to show that Avicenna was involved in fundamental confusions about the central ontological concepts, being and unity – that he confused different senses of these concepts with each other, and that he misconstrued the logical syntax of these concepts – and that Fa¯ra¯bı¯ had already warned against these confusions in his Kita¯b al-Huru¯f and Kita¯b al-Wa¯hid wa’l-wahda. These ˙ ˙ hard to passages in Averroes have remained very obscure, in part ˙because it was 1

I am grateful to Peter Adamson, Amos Bertolacci, Dag Nikolaus Hasse and Robert Wisnovsky for their comments on a draft of this paper, as well as to everyone at the Villa Vigoni conference. Rafael Njera first alerted me to the texts on unity in Averroes’ Epitome of the Metaphysics. Conversations with Marwan Rashed and Richard Taylor were also helpful. This paper forms part of the promised sequel to my ‘Fa¯ra¯bı¯’s Kita¯b al-Huru¯f ˙ and his Analysis of the Senses of Being’. I will often have to summarize here briefly arguments which are developed more fully in that article.

52

understand what Averroes was saying before Muhsin Mahdi edited these treatises of Fa¯ra¯bı¯; even since then, work on the Kita¯b al-Huru¯f has concentrated ˙ and almost nothing on its religio-political rather than its metaphysical aspects, has been done on the Kita¯b al-Wa¯hid, and so Mahdi’s work has not had the payoff that it might.2 I will try to˙ build here on my reconstruction of the metaphysical program of the Huru¯f in ‘Fa¯ra¯bı¯’s Kita¯b al-Huru¯f and his Analysis ˙ some of what I said in ˙ that article may be of the Senses of Being’. While controversial, I think it will be clear that Averroes is reading the Huru¯f ’s central ˙ also that he theses about being in pretty much the same way that I was, and takes Avicenna to have fallen into the same errors about being that Fa¯ra¯bı¯ is criticizing (which I argued were views of the Kindı¯ circle). Furthermore, Averroes’ use of the Kita¯b al-Wa¯hid will show that he takes these two Fa¯ra¯bian ˙ treatises to be very closely connected (as Mahdi had suggested); and Averroes’ use of the Kita¯b al-Wa¯hid in refuting metaphysical errors may give us a clue to ˙ treatise.3 Fa¯ra¯bı¯’s intentions in that While Aristotle speaks of unity and other supremely universal attributes as per se accidents of being (e. g. Metaphysics C1 1003a21 – 2), Avicenna’s considered opinion seems to be that both ‘one’ and ‘being’ or ‘existent’ [mawgˇu¯d] are both per se accidents of the quasi-genus ‘thing’.4 Aristotle says that

˘

˘

˘

˘

˘

˘

˘

˘

˘

˘

4

˘

3

I will cite Fa¯ra¯bı¯’s Kita¯b al-Huru¯f, ed. Mahdi, either by Mahdi’s ba¯b and paragraph ˙ numbers, and his Kita¯b al-Wa¯hid wa’l-wahda by Mahdi’s numbers or by his page and line ˙ paragraph numbers. Translations of the two treatises, by ˙Charles Butterworth and Ahmed Alwishah respectively, have been announced. I will cite the Metaphysics of the Sˇifa¯ from Marmura, Avicenna, The Metaphysics of ‘The Healing’; the Taha¯fut al-Taha¯fut from Bouyges, Averroes, Tahafot at-Tahafot, cited as ‘TT’; the Great Commentary from Bouyges, Averroes, Tafsı¯r ma¯ ba d al-tabı¯ at, cited as ˙ ‘Tafsı¯r’. I will cite the Epitome of the Metaphysics, where there is no satisfactory edition, from Quirs Rodrguez, Averroes, Compendio de Metafisica, by book and paragraph numbers. See Davidson, Alfarabi, Avicenna and Averroes, pp. 221 – 2, n. 5, for the state of the text of the Epitome; see also Davidson’s comments against Bruno Nardi on its authenticity, and on Averroes’ later supplements and corrections, incorporated in some of the manuscripts but not others. On the problems of the Epitome see also Puig Montada, Cuanto se encuentra ms all de la naturaleza. I have not yet seen Arnzen, Averroes, On Aristotle’s ‘Metaphysics’. So in the Nagˇa¯t: ‘lakinna tabı¯ ata ’l-wa¯hidi min al-a ra¯di ’l-la¯zimati li’l-asˇya¯ i, wa-laysa ’l˙ ¯ hiyyati ˇsay˙in min al-asˇya¯˙i, bal taku¯nu ’l-ma¯hiyyatu ˇsay an wa¯hidu muqawwiman li-ma ˙ imma insa¯nan wa-imma farsan aw aqlan aw nafsan, tumma yaku¯nu da¯lika mawsu¯fan bi¯ For other˙ relevant annahu wa¯hidun wa-mawgˇu¯dun’ (Kita¯b al-Nagˇa¯t, ed.¯ Fakhry, p. 245). ˙ discussion, see Wisnovsky, Avicenna’s Metaphysics, pp. 158 – 60. As passages and Wisnovsky points out, Sˇifa¯ Metaphysics I,5 says only that ‘thing’ [sˇay ] and ‘being’ [mawgˇu¯d] are coextensive and mutually entailing but different in meaning, without specifying one of them as underlying and the other as its attribute, but other texts including the Nagˇa¯t passage make it clear that ‘thing’ underlies and that ‘being’ and ‘one’ are its attributes. (This becomes particularly clear in texts, such as those Wisnovsky cites from Sˇifa¯ Metaphysics V,1 and VII,1, where the same thing can have either wugˇu¯d in the ˘

2

˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Stephen Menn

53

Averroes against Avicenna on Being and Unity

the definition of a thing does not include ‘being’ or ‘one’, and Avicenna takes the underlying reason for this to be that if X is a thing other than God, being and unity are not contained in the essence of X, which is what the definition expresses. (And God does not have a definition.) The essence of (say) horse has one kind of existence in an individual horse, another kind of existence as a universal in someone’s mind, but both individual and universal existence are superadded to the essence: neither individuality nor universality, neither unity nor multiplicity belong to the essence considered purely in itself, and this is why none of these are mentioned in the definition of horse. ‘In itself it is neither one nor many, neither existent in re [fı¯ ’l-a ya¯n] nor existent in the soul, neither potentially nor actually in any of these things in such a way that this would enter into horseness; rather, of itself, it is only horseness’ (Sˇifa¯ Metaphysics V,1,4). The accident of existence outside the soul will be important because this is what God, or some subordinate cause of being, adds to an essence in creating a thing; and the accident of unity is important because numbers, in the accidental category of quantity, are just collections of such accidents of unity (III,3 and III,5 – 6). Averroes thinks this is all a mistake: he rejects Avicenna’s theses that being and unity are accidents of horseness, and that there is a combination in re of these accidents with the essence, and a cause for this combination; or rather, Averroes thinks that being and unity are accidents only in senses of ‘being’ and ‘one’ in which being and unity are accidents existing only in the soul, and there is no composition in re. (And he thinks that these are not the senses of ‘being’ and ‘one’ which are objects of metaphysics, and which Aristotle is discussing in texts like Metaphysics C1 – 2.) And Averroes supports all of these claims out of Fa¯ra¯bı¯. ˘

˘

mind or wugˇu¯d in individuals.) Avicenna’s comparison of thing, being and one contrasts with Aristotle, who never mentions ‘thing’ (‘ti’?) as a universal attribute, and always speaks of being as the substratum of which unity and so on are per se accidents. Wisnovsky sees a tension between Sˇifa¯ Metaphysics I,5 and the other texts he cites, suggesting that the Metaphysics I,5 thesis that ˇsay and mawgˇu¯d are mutually entailing and coextensive but different in meaning would be undermined if ˇsay were more fundamental and mawgˇu¯d were its attribute, but I see no tension here, just as Aristotle sees no tension in Metaphysics C1 – 2 in saying that being and one are mutually entailing and coextensive but different in meaning and that unity is an attribute of being. Sˇay is more fundamental in that it is what the others are predicated of, but ˇsay and mawgˇu¯d and one are equally fundamental in the sense that none of them can be reduced to or defined in terms of the others. ˘

˘

˘

˘

˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

54

I Averroes against Avicenna on Essence and Existence in the Taha¯fut al-Taha¯fut and the Great Commentary on the Metaphysics Averroes’ most direct and extended confrontation with Avicenna’s metaphysics (not necessarily the Metaphysics of the Sˇifa¯ , which he may not have read)5 comes ˙ aza¯lı¯ in the Taha¯fut al-fala¯sifa assembles the in the Taha¯fut al-Taha¯fut. G philosophers’ arguments – for which he mainly follows Avicenna – in support of the three theses on which they are infidels and the seventeen on which they are heretics, and tries to show that their arguments are not demonstrative. Averroes on some points defends the philosophers, but on a remarkable number of points ˙ aza¯lı¯ against Avicenna, and tries to restate what he thinks are he agrees with G the authentic Aristotelian arguments, which he thinks Avicenna either misinterpreted or else deliberately replaced with arguments of his own invention; the authentic Aristotelian arguments, so Averroes claims, will be ˙ aza¯lı¯’s critique. immune to G In three related passages, in the fifth, seventh and eighth discussions in the ˙ aza¯lı¯ discusses the opinion of the philosophers that, in Taha¯fut al-fala¯sifa, G things other than God, there is a distinction between being or existence [wugˇu¯d or inniyya] and quiddity or essence [ma¯hiyya or haqı¯qa], that in such things being is not contained in their essence, whereas ˙God is free of all kinds of plurality including the essence-existence distinction. Averroes answers that this is ˙ aza¯lı¯ has been misled by not the opinion of the philosophers, and that G following Avicenna, who deviates from the consensus of the philosophers in thinking that ‘the existence [inniyya] of a thing, that is, the thing’s being existent [mawgˇu¯d], is a thing superadded to the quiddity outside the soul and as if an accident in it’ (Discussion 5, TT 302,14 – 15). As Averroes complains, ˘

as for the account of someone who says that existence [wugˇu¯d] is something superadded and that the existent [mawgˇu¯d] is not constituted by it in its substance [i.e. wugˇu¯d is not a substantial constituent of the mawgˇu¯d], it is a very mistaken account, since it necessitates that the name “mawgˇu¯d” signifies an accident common to the ten categories outside the soul: this is the opinion of Avicenna. And it will be asked of this accident, when it is said of it that it is mawgˇu¯d, whether this signifies the meaning of “true” [sa¯diq] or an accident which is mawgˇu¯d in this accident; and ˙ there will be [tu¯gˇadu] infinitely many accidents, and this is absurd (304,13 – 305,2).

For evidence that Averroes’ knowledge of the Sˇifa¯ is a reconstruction from at best partial ˙ aza¯lı¯’s Taha¯fut al-fala¯sifa, see Davidson, versions and from secondary sources including G Proofs for Eternity, pp. 318 – 20 and pp. 334 – 5. Still, while Averroes sometimes seriously misinterprets Avicenna, and often shows that he does not know the details of Avicenna’s account, sometimes (as we will see below) he gives surprisingly accurate statements of Avicenna’s arguments as we find them in the Metaphysics of the Sˇifa¯ , whether from firsthand knowledge or otherwise. ˘

5

˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Stephen Menn

Averroes against Avicenna on Being and Unity

The criticism of Avicenna (apart from the question about ‘true’) is clear enough. Avicenna says if X exists, its existence must either be contained in its essence, or an accident superadded to its essence. If the existence of X is contained in its essence, so that ‘X exists’ will have the status of ‘man is an animal’, then X is through itself necessarily existent, and Avicenna claims to be able to prove that there is only one such thing, God. So anything that exists other than God must exist through some accident of existence [wugˇu¯d]. But this wugˇu¯d too exists: if its existence is contained in its essence, then this wugˇu¯d is God; if its existence is an accident superadded to its essence, there will be an infinite regress of wugˇu¯da¯t. (If we try breaking the regress by saying that accidents, especially odd accidents like wugˇu¯d, don’t have their own wugˇu¯da¯t but rather exist through the wugˇu¯d of the substance they inhere in, then the wugˇu¯d of the wugˇu¯d of X is the wugˇu¯d of X, and so the wugˇu¯d of X is through itself necessarily existent, and so the wugˇu¯d of X is God.) One possible solution, adopted both by some Arabic and by some Latin writers after Avicenna, is to say that the wugˇu¯d of things is God, so that when I say ‘Suqratu mawgˇu¯dun’, ‘Socrates exists’, the word ˙ way that when I say ‘Suqratu abyadu’, ‘mawgˇu¯d’ signifies God in the same ˙ Avicenna ˙ ‘Socrates [is] white’, the word ‘abyad’ signifies whiteness:6 but both ˙ and Averroes reject this solution. What is not so clear in Discussion 5 is Averroes’ own account (the account that he attributes to the real philosophers from whom Avicenna deviated) of what the term ‘mawgˇu¯d’ signifies – an account which he thinks is necessary if we are to understand and avoid the error that Avicenna fell into, and not merely to refute his conclusions. The philosophers’ account turns on distinguishing two senses of ‘mawgˇu¯d’, and correspondingly of ‘wugˇu¯d’ or ‘inniyya’:

˘

Existence [inniyya] in reality [haqı¯qa] in the things-that-are [mawgˇu¯da¯t] is a mental ˙ intention, namely a thing’s being outside the soul as it is in the soul, and what signifies it is equivalent [mura¯dif ] with the true [sa¯diq], which is what is signified by ˙ judgement. For the expression the hyparctic [wugˇu¯diyya] copula7 in a predicative ‘wugˇu¯d’ is said in two senses: one is what is signified by the true, like our saying ‘does the thing exist [hal al-sˇay u mawgˇu¯dun] or does it not exist?’ and ‘is such-andsuch such-and-such [hal kada¯ yu¯gˇadu kada¯] or is it not such-and-such?’, and the ¯ the things-that-are ¯ other is what occupies among the place of a genus, as when being [al-mawgˇu¯d] is divided into the ten categories or into substance and accident. And if there is understood by ‘mawgˇu¯d’ what is understood by ‘true’, then there is no 6

Nobody thinks that God inheres in Socrates: the view is that Socrates is called ‘mawgˇu¯d’ by extrinsic denomination, as a diet is called healthy, not on account of a health in it, but because of its relation to an extrinsic health, existing in the person or animal who eats it. There are clear statements of this view in post-Avicennian thinkers notably by Eckhart of ˙ anı¯ al-Na¯bulusı¯ Cologne in his Prologues to the Opus Tripartitum and by Abd al-G (1641 – 1731) in his ¯Ida¯h al-maqsu¯d min wahdat al-wugˇu¯d. ˙ ˙ ˙ That is, a non-modal copula, ‘S is˙ P’, as opposed to a modalized copula ‘S is necessarily P’ or ‘S is possibly P’. ˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

55

7

56 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Stephen Menn

multiplicity [of existence and quiddity] outside the soul, and if there is understood by it what is understood by ‘essence [da¯t]’ and ‘thing’, then the name ‘mawgˇu¯d’ is ¯ of what is other than it by priority and said of the necessary of existence and posteriority, like the name ‘heat’ which is said of fire and of hot things: this is the opinion of the philosophers. (TT 302,1 – 12)

Several things here should seem at first face puzzling (and have as far as I know never been discussed in the modern literature): the distinction between two senses of being, the identification of one of them with truth, the assertion that this is the sense we use when we ask whether X exists [hal X mawgˇu¯dun], and the claim that wugˇu¯d in this sense is a merely mental entity, whereas wugˇu¯d in the other sense is a quasi-genus of the ten categories, so that neither of them is a real accident really inhering in the things-that-are. What should be especially puzzling is that all this is attributed to ‘the philosophers’, which (especially when ‘the philosophers’ are contrasted with Avicenna) seems to mean Aristotle. Aristotle distinguishes the senses of being in Metaphysics D7, but he distinguishes four senses, not two: being per accidens, being per se, being as truth, being as actuality and potentiality. Of these, being per se is divided into sub-senses corresponding to the different categories, and so perhaps it is the quasi-genus of the categories, although Aristotle’s sentences illustrating being per se in D7 are not 1-place assertions ‘S is’ or ‘S is a being’ [S mawgˇu¯dun] but 2-place assertions ‘S is P’. Being as truth (the sense Averroes distinguishes from being as the quasigenus of the categories) is another sense mentioned in D7, but nothing in D7 would suggest that when we ask ‘does S exist?’ we are asking about being as truth: all of Aristotle’s examples for being as truth (like his examples for being per se) are 2-place assertions ‘S is P’, and it is hard to see what it could mean to interpret a 1-place assertion ‘S is’, ‘S exists’, as ‘S is true’. Furthermore, D7 seems to be interested just in the things which are said in different senses to be, and which are thus emta or mawgˇu¯da¯t, and not in the status of the wugˇu¯d/eWmai on account of which things are called mawgˇu¯da¯t/emta, as things are called white on account of whiteness: D7 says nothing about whether such a wugˇu¯d/eWmai is an accident superadded to the things which are, much less about whether such a wugˇu¯d/eWmai might exist only in the soul and not in the things themselves. So where is Averroes getting the account of mawgˇu¯d and wugˇu¯d which he attributes to ‘the philosophers’ and puts forward as an alternative to Avicenna’s false account? The key to the answer can be found in another difficult passage, in Discussion 7: Avicenna erred in this [sc. in holding that mawgˇu¯d is not the genus of the categories, even a genus predicated of them per prius et posterius rather than univocally] only because he thought that the name ‘mawgˇu¯d’ signified the true in the language of the Arabs [i.e. of the Bedouin ; and thus that this is the lexically original meaning of the word in Arabic] and that what signifies the true signifies

57

Averroes against Avicenna on Being and Unity

an accident – in reality8 it signifies a second, i. e. a logical, intelligible – he thought that wherever the translators used [‘mawgˇu¯d’] it signified only this meaning. But it is not so. Rather, the translators intended to signify by it only what is signified by the names ‘essence’ [da¯t] and ‘thing’. And Abu¯ Nasr [al-Fa¯ra¯bı¯] has shown this in his Kita¯b al-Huru¯f and¯made it known that one of ˙the causes of the error which arises about ˙this is that the name ‘mawgˇu¯d’ is in the shape of a paronymous [musˇtaqq] [term], and paronymy signifies an accident. Indeed, it is in the original [Arabic] language paronymous, but when the translators did not find in the language of the Arabs an expression which signifies this meaning which the ancients had divided into substance and accident and into potentiality and actuality, I mean an expression which is a first imposition, some of them signified it by the name ‘mawgˇu¯d’.9 But they did not understand by it the paronymous sense, so that it would signify an accident; rather [they used it to signify] what is signified by the name ‘essence’ [da¯t] – it is a technical term and not a term of [ordinary] language. ¯ But other translators saw fit, on account of the ambiguity which arises from this [word], to express the sense which was meant in the language of the Greeks by deriving, from the pronominal expression which signifies the connection of the predicate with the subject [i.e., in Arabic, huwa/hiya as ‘pronoun of separation’], [a word] which signifies this meaning, since they thought that this was closer to signifying this meaning. So they used, in exchange for the name ‘mawgˇu¯d’, the name ‘huwiyya’, even though they thus had to use a [grammatical] form which does not exist in the language of the Arabs; and for this reason the other school inclined to the name ‘mawgˇu¯d’. And what is understood by ‘mawgˇu¯d’ in the sense of ‘true’ is other than what is understood by ‘quiddity’, and for this reason someone can know the quiddity who does not know the wugˇu¯d [i.e. someone can know what X is without knowing that X is]. This meaning is necessarily other than the quiddity in composite things, but in simple things it and the quiddity are one. [This is] not the meaning which the translators signified by the name ‘mawgˇu¯d’, since this is identical with the quiddity. And when we say that some mawgˇu¯d is substance and some is accident, by the name ‘mawgˇu¯d’ must be understood the meaning which the ˘

˘

˘

Translating the text as Bouyges prints it, yadullu ala¯ aradin, wa-la¯ budda bal fı¯ ’l-haqı¯qati ala¯ ma qu¯lin min al-ma qu¯la¯ti ’l-tawa¯nı¯, a nı¯ ’l-mant˙iqiyyati. But la¯ budda bal ¯ ˙ Bouyges notes that the is strange, and something may have gone wrong in the pointing. Cairene editio princeps of 1885 prints, instead of la¯ budda bal, la¯ yadullu, and while this does not yield a reasonable sense, it will make sense if an illa¯ has dropped out, i. e. if Averroes wrote wa-la¯ yadullu fı¯ ’l-haqı¯qati illa¯ ala¯ ma qu¯lin min al-ma qu¯la¯ti ’l-tawa¯nı¯, ¯ a nı¯ ’l-mantiqiyyati, ‘but in reality it signifies only a second, i. e. logical, intelligible’. ˙ Averroes means to contrast ‘expression of first imposition’ here with a paronymous or derived term [musˇtaqq], i. e. a term derived by a morphological change from a term of first imposition, rather than with a term of second imposition, in the sense of a metalinguistic term such as ‘verb’ or ‘accusative’. Crudely speaking, concrete adjectives and active and passive participles as well as finite verbs are paronymous, while the masdar ˙ (infinitive or nomen actionis) of a verb, the abstract noun associated with an adjective, and ordinary concrete nouns not derived from verbs, are non-paronymous. On the somewhat tortuous history of first and second impositions in Greek and Syriac and Arabic see Zimmermann (ed.), al-Fa¯ra¯bı¯, Commentary, pp. xxx–xxxiv. On the importance of the distinction between paronymous and non-paronymous terms see Menn, Analysis of the Senses of Being, pp. 65 – 6 and more broadly. ˘

˘

˘

˘

9

˘

˘

8

˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

58

translators signified, the signification which is said by priority and posteriority of the essences [dawa¯t] of different things; but when we say that substance [is] mawgˇu¯d, there¯must be understood by it what is understood by ‘true’. (TT 371,9 – 373,3)

˘

˘

Here again Averroes speaks of two senses of ‘mawgˇu¯d’ (and two corresponding senses of the masdar, the infinitive or abstract noun, ‘wugˇu¯d’), of which one is ˙ ‘true’, and the other is divided into the ten categories and is synonymous with synonymous with ‘essence’ and ‘thing’. He thinks that Avicenna was led into error by confusing these two senses, more precisely that he took occurrences of the word ‘mawgˇu¯d’ in (Arabic translations of ) Aristotle and other Greek philosophers which in fact meant ‘essence’ or ‘thing’, as if they meant the true. He adds an important element which was not in Discussion 5, that Avicenna was here captive to systematically misleading translations: Avicenna went wrong because he took the word ‘mawgˇu¯d’ in a sense closer to its normal Arabic meaning, not realizing that the translators, finding no natural Arabic equivalent for the Greek word for whatever it is that is divided into the ten categories, had resorted to various compromises, one of which was to artificially impose a new sense on the word ‘mawgˇu¯d’. ‘mawgˇu¯d’ most literally means ‘found’, and when it is extended beyond this basic sense perhaps its next-most-basic sense is ‘present’, but this doesn’t seem to be what Averroes is worried about. What he seems to find most misleading about the translators’ choice of ‘mawgˇu¯d’ is not so much the content of the word as its grammatical form, the fact that it is grammatically paronymous (on the theory that Averroes is presupposing, every active or passive participle, and also every finite verb-form, is paronymous from the masdar of the verb). When a word is paronymous, its grammatical form suggests˙that it signifies an accident: just as ‘white’ signifies that something has an accident (a quality) of whiteness in it, and ‘runner’ or ‘runs’ signifies that something has an accident (an action) of running in it, so ‘mawgˇu¯d’, according to its grammatical form, would signify that something has an accident of wugˇu¯d in it. Indeed, Averroes more-or-less agrees that when the word ‘mawgˇu¯d’ signifies the true – as surely it sometimes does even in Arabic translations of Greek philosophical texts – it signifies an accident. Only ‘more-or-less agrees’, because the wugˇu¯d through which a thing is mawgˇu¯d-in-the-sense-of-the-true is not an accident in one of the nine accidental categories, really existing in its subject (if it were, there would be an infinite regress of wugˇu¯da¯t), but rather ‘a second, i. e. a logical, intelligible’, as Averroes says here, or ‘a mental intention’, thus something existing only in the mind, as he said in Discussion 5. But when someone says that X is mawgˇu¯d in the sense in which ‘mawgˇu¯d’ is divided into the categories, this doesn’t signify anything beyond the essence of X, and the corresponding wugˇu¯d just is the essence of X: indeed, Averroes says that mawgˇu¯d in this sense signifies the same as ‘thing’, ‘sˇay ’, and so wugˇu¯d should signify the same as ˇsay iyya, and Avicenna agrees that the ˇsay iyya of a thing is ˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Stephen Menn

Averroes against Avicenna on Being and Unity 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

59

precisely its essence.10 In sum, the wugˇu¯d of a thing, depending on the sense in which it is taken, is either something mental or else the essence of the thing: it is never something really existing in the thing and really distinct from its essence. Fa¯ra¯bı¯ and Averroes on mawgˇu¯d and wugˇu¯d Anyway, this is what Averroes thinks. But why does he believe it, how can he attribute all this to ‘the philosophers’ (i. e. to Aristotle), and where is he getting the linguistic claim about a non-paronymous word in Greek for which the translators substituted a paronymous Arabic word? The key is the reference to Fa¯ra¯bı¯ and the Kita¯b al-Huru¯f. Averroes’ reference does not make it clear how ˙ much of this story he is taking from Fa¯ra¯bı¯, and before the Huru¯f was printed there was no way to know. But it is now clear that Averroes is˙ taking the whole thing from Fa¯ra¯bı¯. As I argued in ‘Fa¯ra¯bı¯’s Kita¯b al-Huru¯f and his Analysis of the Senses of Being’, one of Fa¯ra¯bı¯’s main concerns ˙in the Huru¯f is with what ˙ Aristotle calls sophisms of sw/la t/r k´neyr, logical errors arising when the grammatical form of some word or sentence does not correspond to its logical form and so misleads us about its logical form: he is particularly concerned with errors that arise when a term is grammatically but not logically paronymous, i. e. when ‘F’ does not signify ‘that in which an F-ness is present’, where ‘F-ness’ is the masdar corresponding to the paronymous term ‘F’. Fa¯ra¯bı¯ thinks that this ˙ especially when a term that is not paronymous in Greek is translated by happens a paronymous term in Arabic, and he thinks that this is what has happened to the Greek word ‘astı¯n’, which is rendered into Arabic by some translators as mawgˇu¯d and by others by terms derived from huwa in its use as a pronoun of separation; Averroes’ story about the translators is entirely taken over from Fa¯ra¯bı¯.11 Fa¯ra¯bı¯ also thinks, like Averroes, that being is said in two main senses, being-as-truth and being as divided into the categories: this agreement is particularly important since, as noted, Aristotle lists four senses of being in Metaphysics D7, and no one before Fa¯ra¯bı¯ had reduced them to two in this way. Most importantly, Fa¯ra¯bı¯ gives an entirely new interpretation – in which 10 See Wisnovsky, Avicenna’s Concept of Thingness; developed in Avicenna’s Metaphysics, chapters 7 – 9. 11 Fa¯ra¯bı¯’s account of the translators is in Kita¯b al-Huru¯f I,83 – 6; see the discussion in Menn, Analysis of the Senses of Being, pp. 74 – 6˙and more broadly pp. 71 – 6. Usta¯t, ¯ whose translation of the Metaphysics Averroes normally takes as his base for ˙his commentary, favors huwiyya, but Averroes generally rephrases in terms of mawgˇu¯d and wugˇu¯d, which had been the standard terms in Arabic philosophy at least since the time of Fa¯ra¯bı¯. For everything we know about the translators of the Metaphysics see Bertolacci, On the Arabic Translations.

60

˘

Averroes follows him without reservation – to ‘being as truth’. From Metaphysics D7 we might think that this was only a sense of 2-place being, i. e. a sense of ‘is’ in contexts like ‘S is P’ (in the philosophers’ artificial Arabic ‘S mawgˇu¯dun Paccusative’), but Fa¯ra¯bı¯ thinks that being-as-truth is also a sense of 1-place being, i. e. a sense of ‘is’ in contexts like ‘F is’, ‘F exists’ (‘F mawgˇu¯dun’): for Fa¯ra¯bı¯, to ask whether F is, in the sense of the true, is to ask whether the concept of F is instantiated, and he thinks that this is what we are very frequently asking when we ask whether there are F’s.12 Fa¯ra¯bı¯ also, like Averroes, is very much concerned with the senses of ‘wugˇu¯d’ corresponding to these two senses of ‘mawgˇu¯d’: this is the connection with his story about how the Arabic translators of Greek philosophical texts chose a paronymous word to render ‘astı¯n’, since the paronymy of ‘mawgˇu¯d’ tempts Arabic-speakers into thinking that, when F exists, it does so through a wugˇu¯d present in it. Fa¯ra¯bı¯, like Averroes, thinks that this is not in fact correct for either sense of mawgˇu¯d. Averroes is following Fa¯ra¯bı¯ in saying that when ‘mawgˇu¯d’ is used in the sense of the true the wugˇu¯d that it signifies is a ‘second intelligible’ [ma qu¯l ta¯nı¯]: Fa¯ra¯bı¯ is apparently the inventor ¯ of this phrase, and he thinks that such second intelligibles are the domain of logic, which is why Averroes speaks of a ‘second, i. e. logical, intelligible’.13 What Fa¯ra¯bı¯ means by saying that ‘mawgˇu¯d’ in the sense of the true signifies a second intelligible is that when I say ‘F mawgˇu¯dun’ = ‘F exists’ in this sense, I am not asserting something of a thing F, but rather asserting something of a concept of F in the mind, namely that that concept is instantiated outside the mind or, as Fa¯ra¯bı¯ says, that it ‘is outside the soul as it is inside the soul’ (Huru¯f, p. 116,5), ˙

˘

12 For all this see Menn, Analysis of the Senses of Being. Fa¯ra¯bı¯ distinguishes the whetherit-is question (eQ 5sti, hal mawgˇu¯dun) which Aristotle discusses in Posterior Analytics II – are there phoenixes, are there eclipses – into two questions, the first asking whether the concept of F is instantiated, and then, if the answer to the first question is yes, a second question asking whether F has a real essence outside the mind. See below for discussion. 13 Fa¯ra¯bı¯ apparently coins the phrase ‘second intelligible’ [ma qu¯l ta¯nı¯] (for which Avicenna substitutes the equivalent ‘second intention’ [ma na¯ ta¯nı¯]) on¯ the model of the older notion of a term of second imposition [wa d ta¯nı¯], in ¯the sense of a metalinguistic term; see, as before, Zimmermann (ed.), al-Fa¯ra¯˙bı¯¯, Commentary, pp. xxx–xxxiv, and Menn, Analysis of the Senses of Being, p. 81. A second intelligible is a thought about other thoughts, or a predicate of other predicates. Wa d awwal/ta¯nı¯ translate Greek pq¾tg/ ˙ deut´qg h´sir, but ma qu¯l awwal/ta¯nı¯ and ma na¯ awwal/t a¯nı¯¯do not seem to translate any ¯ ¯ Greek terminology. The phrase ‘de¼teqai 1p¸moiai’ which Zimmermann p. xxxiii cites from Dexippus, In Categorias 12,14 – 15, may be an antecedent, but in context Dexippus is talking about concepts signified by compound nouns and the like. Dexippus also, without the phrase ‘de¼teqai 1p¸moiai’, talks about other terms that do not signify things in the categories, including what Latin writers will call syncategorematics and also including some metalinguistic terms, but he does not seem to have the idea of metaconceptual concepts analogous to metalinguistic terms, as Fa¯ra¯bı¯ very clearly does, see esp. Kita¯b al-Huru¯f I,7 – 10. Still, this kind of thing in the Greek commentators, one ˙ way or another, gives Fa¯ra¯bı¯ his start. ˘

˘

˘

˘

˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Stephen Menn

61

Averroes against Avicenna on Being and Unity

the phrase which Averroes also uses14 in the passage cited above from Discussion 5 to paraphrase being-as-the-true in both 1-place and 2-place contexts. So although the fact that F exists in this sense is non-tautological, it does not involve any composition of subject and attribute in re: the wugˇu¯d in question can exist only in the mind, being an attribute of a concept or ‘first intelligible’ which itself exists in the mind. By contrast, the wugˇu¯d which is signified by ‘mawgˇu¯d’ in the other sense, the sense which is divided into the categories, is something real outside the mind, but it is not an accident inhering in the thing which is mawgˇu¯d, being rather the quiddity of the thing (or perhaps a part of the quiddity). Fa¯ra¯bı¯ glosses this sense of ‘mawgˇu¯d’ as ‘having an essence outside the soul’ or more literally ‘delimited by [munha¯z bi-] some essence outside the soul’ (Huru¯f, p. 116,7).15 ˙ ˙ At first hearing this sounds close to the gloss on being-as-the-true as ‘being outside the soul as it is inside the soul’, but it is importantly different. Fa¯ra¯bı¯ insists that being-as-the-true, i. e. ‘being outside the soul as it is inside the soul’, is predicated univocally of everything – what it is for a concept to be instantiated is the same whether the concept is of something in a category like ‘man’ or ‘white’, a negation like ‘not-white’ or a privation like ‘blind’, or presumably an ens per accidens like ‘white man’. But being as ‘having an essence outside the soul’ will not apply univocally to all these things, since while both man and white have essences, accidents have essences only derivatively from substances (whatit-is-to-be-white is just what-it-is-for-some-substance-to-be-white), and white man and not-white and blind do not have essences at all; so being as ‘having an essence outside the soul’ will apply non-univocally to things in the different categories, and will not apply at all to things outside the categories. Averroes is following all this in Discussion 5, a little below the passage that I cited above, when he describes the sense of being other than the true as ‘what privation is opposed to [i.e. what applies only to positive things and not to negations and privations] … what is divided into the ten genera and is like their genus’ (TT 303,12 – 13). It is like a genus in that mawgˇu¯d in this sense is an essential predicate of things in all ten categories, only like a genus because ‘it is said of the ten categories by priority and posteriority’ (303,15), whereas being in the sense of the true is said equally of all the categories (303,17 – 304,1) and presumably also of privations and other things outside the categories. ˘

14 Although not quite verbatim: Averroes says that being as the true is kawnu ’l-sˇay i ha¯rigˇa ˘ qu¯l ’l-nafsi ala¯ ma¯ huwa alayhi fı¯ ’l-nafsi (TT 302,2 – 3), Fa¯ra¯bi that it is said of every ma ˇ which ka¯na ha¯riga ’l-nafsi wa-huwa bi- aynihi ka-ma¯ huwa fı¯ ’l-nafsi. But both Fa¯ra¯bı¯ and ˘ themselves some variation in expression. Averroes allow 15 In ‘Analysis of the Senses of Being’ I said ‘circumscribed’ for munha¯z. ‘Delimited’ is less awkward, but runs the risk of falsely suggesting a close connection ˙with ‘limit’, niha¯ya (= p´qar). To avoid this risk I will translate the latter, somewhat less naturally, as ‘boundary’. ˘

˘

˘

˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

62 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Stephen Menn

I argued in ‘Fa¯ra¯bı¯’s Kita¯b al-Huru¯f and his Analysis of the Senses of Being’ that Fa¯ra¯bı¯ in the Kita¯b al-Huru¯˙f is restating and reworking, for an Arabicspeaking Muslim audience, ˙ what he sees as the central achievements of Aristotle’s Metaphysics; that Part I of the Kita¯b al-Huru¯f in particular rework Metaphysics D, and that the chapters on mawgˇu¯d and˙ wugˇu¯d (I,80 – 103) rework D7. Since Averroes accepts Fa¯ra¯bı¯’s analysis of the different senses of mawgˇu¯d, we would expect him to draw on these chapters of the Huru¯f in commenting on ˙ D7 in his Great Commentary on the Metaphysics. And indeed he does, although without mentioning Fa¯ra¯bı¯’s name as he does in the Taha¯fut al-Taha¯fut;16 but much of what he says is closely dependent on Fa¯ra¯bı¯, and much of it is closely parallel to discussions we have seen from the Taha¯fut al-Taha¯fut; and the aim is, as there, to avert what he sees as Avicenna’s confusions. While Averroes, constrained by Aristotle’s text, has a discussion of being per accidens at the beginning (pp. 553 – 4, commenting on the text of Aristotle p. 552), and a page on being actually and potentially at the end (562,5 – 563,7), his overwhelming interest is in being per se and being as the true: specifically in the relations between the terms ‘mawgˇu¯d’ (and cognates) and ‘huwiyya’ (and cognates), which the Arabic translators used to render the same Greek word; in the danger of being misled by the paronymous grammatical form of the word ‘mawgˇu¯d’ into thinking that it signifies an accident; and in the difference between the ‘mawgˇu¯d’ or ‘huwiyya’ which signifies the essence of the thing and the ‘mawgˇu¯d’ or ‘huwiyya’ which signifies the true (he says indifferently ‘haqq’ or ‘sa¯diq’).17 All ˙ he tells ˙ about the of this is taken directly from the Kita¯b al-Huru¯f, as is the story ˙ dilemmas of the translators (should they say ‘mawgˇu¯d’, despite its misleading paronymous form, or coin a new word and say ‘huwiyya’?), very closely parallel to Taha¯fut al-Taha¯fut 371,9 – 373,3, cited above; here as in the Taha¯fut alTaha¯fut passage he explicitly names Avicenna as having fallen into the trap of taking ‘mawgˇu¯d’ to signify an accident.18 (He does not say here, as he does in the 16 According to Bouyges’ index, Averroes in the Great Commentary cites Fa¯ra¯bı¯ by name (or rather by his kunya, Abu¯ Nasr) only three times, at 886,1 and 1498,6 and 1499,1: none ˙ to the Huru¯f. of these mentions are relevant ˙ long polemic against Avicenna, on whether a 17 Averroes also gets occupied in another concrete accidental term like ‘white’ primarily signifies the accident or the underlying substance (558,7 – 559,14). Averroes’ argument here turns on a misconstrual of Aristotle’s text for which the translator is at least partly to blame. I hope to return to these issues elsewhere. 18 ‘You must know that the name “huwiyya” is not the form of an Arabic name in its origin. Rather, some of the translators were compelled to this, and derived this name from the connective particle [i.e. the copula, i. e. “huwa” or “hiya” as a pronoun of separation], I mean what signifies among the Arabs the connection of the predicate with the subject in its substance: this is the particle “huwa” in their saying “Zayd, huwa an animal, or a man.” For the utterance of someone who says “man, huwa an animal” signifies what is signified by our saying “man, his substance or essence is that he is an animal.” And when

Averroes against Avicenna on Being and Unity 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

63

Taha¯fut al-Taha¯fut parallel, that Avicenna thought that ‘mawgˇu¯d’ in the translations meant the true, but he does say this in the commentary on Iota 2, in a passage which also parallels the Taha¯fut al-Taha¯fut infinite regress refutation of making existence a superadded accident.)19 In the commentary on D7 (559,15 – they found that this particle was of this character, they derived this name from it, according to the custom of the Arabs in deriving one name from another, for [the Arabs] do not derive a name from a particle; and this name signifies what is signified by “the essence of the thing.” Some of the translators were compelled to this, as we have said, since they thought that it would signify in the translation what was signified by the expression which was used in the language of the Greeks corresponding to “mawgˇu¯d” in the language of the Arabs, indeed that it would be more apt to signify this than the name “mawgˇu¯d”, since when the name “mawgˇu¯d” is used in the sciences to signify the essence of a thing, then since it is a paronymous/derived [musˇtaqq] name in the speech of the Arabs and paronymous/derived names only signify accidents, it is imagined that it signifies an accident in [the thing which is said to be mawgˇu¯d]: this is what happened to Avicenna. And so some of the translators turned away from this expression to the expression “huwiyya,” since this [misunderstanding, or grammatical feature giving rise to a misunderstanding] does not happen with this [expression]. And if the word “mawgˇu¯d” in the speech of the Arabs signified what “thing” signifies, then it would be truer [i.e. more adequate] in signifying the ten categories than the name “huwiyya”, since this name [sc. “mawgˇu¯d”] does occur in the speech of the Arabs. But since this [misunderstanding or grammatically misleading feature] happens with the name “mawgˇu¯d”, some of [the translators] preferred to it the name “huwiyya”. For this reason, when it is used here, there must not be understood by it anything of the meaning of paronymy, even though its form is the form of a paronymous name, I mean the name “mawgˇu¯d”’ (557,5 – 558,6). 19 ‘Avicenna thinks that “being [mawgˇu¯d]” and “one” signify of a thing something superadded to its essence, since he does not think that a thing exists [that it {is} mawgˇu¯d] by its essence, but rather by an attribute superadded to it, like our saying of a thing that it is white. “One” and “being” according to him signify an accident in the thing; we have already in another place mentioned the absurdities which follow necessarily from this thesis. The first thing that follows necessarily is that it is said to him, “this attribute or this accident by which the one becomes one or the existent [mawgˇu¯d] becomes existent, does it also become one or existent by something superadded, or by its essence?”. If he says “by something superadded”, there necessarily follows an infinite regress; and if he says “by its essence”, he has agreed that something exists [and is] one by its essence. [This argument is also given in the commentary on C2, 314,19 – 315,9.] The man was deceived by two things: first, that he thought that the one which is the principle of quantity was the one which is equivalent [mura¯dif ] to the name “being”, so he thought, instead of [li-maka¯ni] [thinking] that this one is enumerated among accidents, that the “one” which signifies all the categories was an accident; and second, that he confused the name “being” which signifies the genus [i.e. the quasi-genus said per prius et posterius of the ten categories] and [the name “being”] which signifies the true: for [the name “being”] which signifies the true is an accident, but [the name “being”] which signifies the genus signifies each of the ten categories by a signification corresponding to the way that “huwiyya” is said’ (1279,12 – 1280,11). I’ll come back below to Averroes’ criticism of Avicenna on unity, as developed in the Epitome of the Metaphysics. Note that Averroes can block the infinite regress argument also in the case of being-as-truth, since the wugˇu¯d in this sense is just an accident in the mind: even if there is an infinite sequence of such

64

560,14) Averroes insists, following the Kita¯b al-Huru¯f, that ‘mawgˇu¯d’ or ˙ and 2-place [murakkab] ‘huwiyya’ can signify the true in both 1-place [mutlaq] ˙ being as truth in D7 are 2contexts, even though all of Aristotle’s examples of place. Averroes also stresses that being as truth is something in the mind, which D7 does not say but the Kita¯b al-Huru¯f does.20 He also stresses, following the Kita¯b al-Huru¯f without any text in˙ Aristotle, that being as truth, unlike being ˙ per se, signifies an accident, ‘and for this reason the commentators on the second book of the Topics differ about the simple question, i. e. about our asking [not whether S is P but] whether the thing exists [hal al-sˇay u mawgˇu¯dun], whether this falls under questions of accident or under questions of genus. Whoever understands by the word “mawgˇu¯d” here what he understands by “true” says that it falls under the question of accident’ (561,8 – 13). Averroes is here copying the claim about the Topics commentators directly from Fa¯ra¯bı¯ Kita¯b al-Huru¯f ˙ of III,246, which refers to Alexander’s Topics commentary: Fa¯ra¯bı¯ is thinking Alexander In topica 53,2 – 10, which does mention a disagreement about whether ‘does X exist?’ is a question of genus or of accident, but the interpretation of the dispute as turning on two different senses of being is entirely Fa¯ra¯bı¯’s invention. ˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Stephen Menn

Fa¯ra¯bı¯ and Averroes on Posterior Analytics II: a Scientific Treatment of Essence and Existence? Both for Fa¯ra¯bı¯ and for Averroes, this distinction between two senses of being is bound up with an interpretation not only of Metaphysics D7 but also of Posterior Analytics II.21 Aristotle says that we can give a scientific definition of a thing (say t¸ 1sti, grasp its quiddity) only after we know that it exists. More specifically, after we know that F exists, we can ask why F exists, and grasping this cause will thoughts about thoughts, the higher ones are derivative of the lower ones, not presupposed by them, cf. Fa¯ra¯bı¯’s discussion in Kita¯b al-Huru¯f I,7 – 10 of why there are ˙ when e. g. ‘accusative’ is put no ‘third intentions’, just as there are no ‘third impositions’ in the accusative. It may not be as easy to force Avicenna to a regress of unities as to a regress of wugˇu¯da¯t: if the wugˇu¯d of X is mawgˇu¯d by its essence it’s an intrinsically necessary existent and there’s supposed to be only one of those, but why can’t the oneness of X be one by its essence? Perhaps because there would be nothing to individuate things whose essence consists just in their being one, which is a point Aristotle makes against the Academics; but Avicenna can reply that they’re accidents individuated by their subjects, unlike the Academics’ units, and also unlike anything which is mawgˇu¯d by its essence. 20 So Averroes says that Aristotle in his examples of being as the true ‘means only that there is a difference between the expression “huwiyya” which signifies the copula in the mind [dihn] and [the expression “huwiyya”] which signifies the essence [da¯t] outside the mind’, ¯ ¯ 561,17 – 562,1. 21 For all of this in Fa¯ra¯bı¯ see Menn, Analysis of the Senses of Being, pp. 84 – 90.

Averroes against Avicenna on Being and Unity 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

65

be grasping the essence of F: thus once we know that there are lunar eclipses, we can ask why there are lunar eclipses, and the answer (‘because the earth blocks the sun’s light from reaching the moon’) also gives the definition of lunar eclipse (‘darkening of the moon due to interposition of the earth between moon and sun’). And although lunar eclipse is an accident, Aristotle thinks a similar procedure will work for matter-form composite substances. So, as Averroes says, ‘the knowledge of the quiddity of the thing cannot be sought until it is known that [the thing] exists’ (TT, Discussion 5, 304,3 – 4). But how can we know whether there are lunar eclipses, if we don’t yet know what a lunar eclipse is? As Averroes says, ‘as for the quiddity which precedes the knowledge of existence in our minds, it is not really a quiddity; it is only the explanation of the meaning of some name’ (TT 304,4 – 6): thus if we can explain the name ‘lunar eclipse’ as ‘darkening of the moon at opposition [i.e. when it is roughly opposite to the sun and so would normally be full]’, this is enough for us to be able to determine whether there are lunar eclipses, i. e. whether the moon is sometimes dark at opposition, although we do not yet know the quiddity of lunar eclipse. So we start with a nominal definition spelling out the concept of F, then ask whether this concept is instantiated, then (if the answer is yes) we investigate why F exists, and in so doing we discover the quiddity of F. So far this is all Aristotelian, and does not turn on a distinction in senses of being. But Fa¯ra¯bı¯ and Averroes add that when we come to know the wugˇu¯d of F (i. e. come to know that F exists) without knowing the quiddity of F, what we know is the wugˇu¯d of F only in the sense that F is true, i. e. that the concept of F is instantiated, and this is a predicate of the concept of F rather than of the thing F: ‘the mawgˇu¯d which is in the sense of the true is a meaning [or “entity”] in the minds, namely the thing’s being outside the soul as it is in the soul, and this knowledge [sc. that the thing is mawgˇu¯d in this sense] precedes the knowledge of the quiddity of the thing’ (TT 304,1 – 3). Before I have investigated the quiddity of lunar eclipse I can’t know that lunar eclipses are mawgˇu¯d in the sense of having an essence, the sense of mawgˇu¯d said by priority and posteriority of the different categories, since I won’t yet know which category lunar eclipses are in, or whether they are in no category at all (as if an eclipse is a negation or a privation, or a being per accidens like white man); but I can know that lunar eclipses are mawgˇu¯d in the sense of truth, since mawgˇu¯d in this sense is univocal, to things in any category and to things in no category. Once we know that F is mawgˇu¯d in the sense of truth, we can go on to ask whether it is mawgˇu¯d in a second sense. The meaning of our saying ‘is the thing mawgˇu¯d?’, in the case of what has a cause which determines its existence [wugˇu¯d], has the same force as our saying ‘does the thing have a cause or does it not have a cause?’, as Aristotle says at the beginning of Book II of On Demonstration [i.e. the Posterior Analytics]; and if it does not have a cause, the meaning is ‘does the thing have some necessary attribute [la¯zim] which

66

determines its existence?’. When there is understood by ‘mawgˇu¯d’ what is understood by ‘thing’ and ‘essence’, it follows the pattern of a genus which is said per prius et posterius; whatever [sense of ‘mawgˇu¯d’] in which what has a cause and what does not have a cause do not differ, it does not signify anything superadded to the mawgˇu¯d, and this is what is meant by ‘true’, and if it signifies something superadded to the essence, it is only something mental [ma nan dihniyun] which has no existence ¯ case for the universal. (TT outside the soul except potentially, as is also the Discussion 8, 392,10 – 14). ˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Stephen Menn

This second kind of mawgˇu¯d, unlike mawgˇu¯d as truth, has different senses for uncaused things (i. e. for God or immaterial substances) and for caused things, and indeed for different kinds of caused things: the cause will in each case give the quiddity, and to say that it has some cause is just to say that it has some quiddity (unlike negations and privations and beings per accidens), and the investigation into wugˇu¯d in this sense will be completed by determining more precisely that cause and that quiddity. In all of this Averroes is following Fa¯ra¯bı¯: in particular, both are supplementing Metaphysics D7 with Posterior Analytics II to give a systematic account of the meanings of mawgˇu¯d and wugˇu¯d and of the relations between the existence [wugˇu¯d] of F and essence or quiddity of F. But Averroes would not follow uncritically any authority except Aristotle’s – he has very high regard for Alexander of Aphrodisias, but he is perfectly capable of criticizing him, and Fa¯ra¯bı¯ as well. And while Fa¯ra¯bı¯’s exegesis of Posterior Analytics II is to a large extent (certainly not entirely) faithful to the text, his account especially of beingas-truth (and his reading it into the Posterior Analytics) is, at the least, a construction sufficiently remote from the surface of the text that we would usually expect Averroes to be suspicious. The reason why Averroes is willing to accept all this from Fa¯ra¯bı¯ is that it gives him a way of dealing with the problem of Avicenna. At the most obvious level, in writing the Taha¯fut al-Taha¯fut ˙ aza¯lı¯’s summary of the arguments of ‘the Averroes is confronted with G philosophers’, taken mainly from Avicenna and full of Avicennian metaphysical concepts, essence and existence and contingency and necessity and so on. His ˙ aza¯lı¯ is right that Avicenna’s metaphysical strategy, often, is to say that while G arguments do not work, these arguments are Avicenna’s innovations, and the ˙ aza¯lı¯’s objections. But Averroes still has to say authentic Aristotle is immune to G what he thinks about essence and existence and contingency and necessity and so on, so he needs an ‘authentically Aristotelian’ theory of these things to counterpose to Avicenna, and Fa¯ra¯bı¯ shows him how to find unsuspected metaphysical depths in Posterior Analytics II, not merely a methodology for proving existence and defining essences, but a theory of essence and existence that supports that methodology. More than this, Fa¯ra¯bı¯ offers Averroes an explanation of how Avicenna could go so badly wrong. This is a real problem for Averroes, just as it is a

Averroes against Avicenna on Being and Unity 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

67

˙ aza¯lı¯ could go so badly wrong. Averroes’ epistemology problem for him how G does not admit the possibility of radically different world-views, there are just different degrees of knowledge and ignorance, and anyone sufficiently familiar with scientific concepts should spontaneously recognize the truth of the fundamental scientific principles, from which the other propositions of the ˙ aza¯lı¯ are unquestionably intelligent sciences can be deduced. But Avicenna and G and familiar with the sciences; so Averroes needs some explanation in their ˙ aza¯lı¯ he sometimes says – perhaps with some justification cases. (In the case of G ˙ – that Gaza¯lı¯ really agrees with the philosophers and is for some reason disguising his true opinion.)22 Fa¯ra¯bı¯, writing before Avicenna’s birth, gives Averroes some explanation for Avicenna’s errors. In the first place, as we have seen, Averroes accepts Fa¯ra¯bı¯’s story about the difficulties of the Arabic translators in rendering the Greek non-paronymous word for being; Avicenna would not be aware of the discrepancy between the translation and the original, and would assume that what is designated by the grammatically paronymous ‘mawgˇu¯d’ is also logically paronymous, so that things are mawgˇu¯d through the presence of a wugˇu¯d really distinct from them.23 As we have seen, Averroes says that Avicenna thought on the basis of Arabic grammar and idiom that the word ‘mawgˇu¯d’, as used by the translators of Greek philosophical texts, signified the true, which is indeed logically paronymous, in the sense that ‘F is mawgˇu¯d’ meaning ‘there is an F’ cannot hold on account of the essence of F unless F is per se necessarily existent, and so Avicenna thought that ‘mawgˇu¯d’ was neither a genus over the ten categories or a non-univocal quasi-genus over them. But Avicenna assumes that ‘mawgˇu¯d’ is a first intelligible, signifying something in re, as indeed it must be if al-mawgˇu¯du bi-ma¯ huwa mawgˇu¯dun, being qu being, is ˙ aza¯lı¯’s proceeding against the 22 At TT 347,10 – 13 he speaks of the ‘insolence’ [qubh] of G philosophers when he agrees with them on most of˙ their opinions; at TT 117,6 – 8 he says that ‘it is clear from the books attributed to him that he relies in the divine sciences on the opinion of the philosophers’, and that this is clearest in the Misˇka¯t al-anwa¯r [Niche for Lights]. Similar ideas in the extended personal attack at TT 352,14 – 354,5, ˙ aza¯li’s fame and the brilliance of his books where he says in particular that most of G derived from his reading of the books of the philosophers. In none of these texts can ˙ aza¯lı¯, who is just ‘this man’; but at TT 159,10 – 160,2, Averroes bring himself to name G ˙ aza¯lı¯’s kunya] after complaining about sophistical reasoning, he says ‘but Abu¯ Ha¯mid [G ˙ was of greater stature than this, but perhaps the people of his time compelled him to [write] this book so that he could divert from himself the suspicion that he followed the opinion of the sages [i.e. the philosophers].’ 23 In the case of God, whose wugˇu¯d is identical with himself, Avicenna must admit a discrepancy between the grammatical and the logical form of ‘mawgˇu¯d’. But if F is contingent, Avicenna cannot say that F’s wugˇu¯d is F itself, or he would make ‘F is mawgˇu¯d’ a per se necessary truth. Averroes thinks that he can avoid this problem, where Avicenna cannot, by saying that ‘mawgˇu¯d’ in this context means the true, and that in this sense the wugˇu¯d is a mere second intention, so neither really identical with F nor really distinct from it.

68 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Stephen Menn

an object of metaphysics rather than of logic (all parties seem to accept Fa¯ra¯bı¯’s characterization of logic as the science of second intelligibles); so he is left with wugˇu¯d as a real accident really distinct from contingent beings, and thus he falls into the problem of an infinite regress of wugˇu¯da¯t and other problems that Averroes is delighted to raise for him. The fact that Fa¯ra¯bı¯ has exposed all of these errors of Avicenna before Avicenna made them makes his exposure all the more persuasive: this shows (so Averroes can say) that what Averroes is taking from Fa¯ra¯bı¯ is not an ad hoc response to Avicenna, rather Fa¯ra¯bı¯ is pointing to objectively misleading features of the situation, and notably of the Arabic translations of Greek philosophical texts, which make it understandable that even someone as brilliant as Avicenna might be misled. And Fa¯ra¯bı¯ speaks with such authority about the structure of the Greek language, the way it expresses the concept of being, and the difficulties of the translators, that Averroes probably assumes in all innocence that Fa¯ra¯bı¯ knew Greek, or at least had reliable sources for what he said about the language. It is probably also important for Averroes that, while Avicenna says that we can know the essence of F without knowing whether F exists, on the alternative account which Averroes (following Fa¯ra¯bı¯) extracts from the Posterior Analytics, what we can know prior to knowing that F exists is only the explication of the name ‘F’, and not the real essence of F. So Averroes can conclude that when Avicenna thinks that he has grasped a real essence, independently of its existence in re, he has in fact grasped only a mental concept which he wrongly regards as a real nature. Fa¯ra¯bı¯ in the Kita¯b al-Huru¯f draws a contrast between dialectic, ˙ which proceeds from the what-F-is question (asking for the explication of the name) to the whether-F-is question (asking for being-as-truth, i. e. for whether the concept of F is instantiated), and science, which picks up once the work of dialectic is done, and proceeds from the distinctly scientific whether-F-is question (asking for being-as-having-an-essence, i. e. asking whether F has a real essence, once we know that F exists in the sense of truth) to the what-F-is question (investigating that essence). If Averroes takes this over from Fa¯ra¯bı¯, he would also be taking over and applying against Avicenna the charge that Fa¯ra¯bı¯ brings against his opponents in the Kita¯b al-Huru¯f, that they are taking their metaphysics of essence and existence from the˙ practice of dialectic, when their should be taking it from the practice of science, as analyzed in the Posterior Analytics. In ‘Fa¯ra¯bı¯’s Kita¯b al-Huru¯f and his Analysis of the Senses of Being’ I argued ˙ that Farabı¯ drew these various distinctions, and (or sketched an argument) ¯¯ reconstructed the syntax of being in Greek as he did, in order to attack what he saw as the metaphysical error of the Kindı¯ circle and Arabic neo-Platonica like the Liber de causis which take God as the being-itself in which other things participate in order to exist, the wugˇu¯d through which other things are

Averroes against Avicenna on Being and Unity

mawgˇu¯da¯t: Fa¯ra¯bı¯ argues that if we mean mawgˇu¯d-as-having-an-essence, the wugˇu¯d of F is not extrinsic to the essence of F, and that if we mean mawgˇu¯d-asthe-true, the wugˇu¯d of F is a mere second intention, and so in either case it cannot be God. Avicenna, of course, does not think that the wugˇu¯d of F is God (although, as we have seen, he may be logically forced to that view to avoid an infinite regress of wugˇu¯da¯t), rather he thinks it is an accident of F, but Averroes thinks that this is still committing the same fundamental error, and he turns against Avicenna the machinery (including the interpretation and harmonization of Metaphysics D7 and Posterior Analytics II, and at least some of the speculative reconstruction of the Greek language and of the difficulties facing the translators) that Fa¯ra¯bı¯ had apparently first designed against the Kindı¯ circle and the neo-Platonica. And, as I have argued elsewhere, Averroes’ Fa¯ra¯bian criticisms of Avicenna on being, and the need to save Avicenna’s essenceexistence distinction in face of them, are the starting point for Thomas Aquinas’ distinction between the senses of esse; and doubtless for much other Latin scholastic ontology as well.24 Avicenna’s Response to Fa¯ra¯bı¯?

˘

Thus far in this section when I have spoken of Avicenna I have been speaking entirely of Averroes’ construction of Avicenna, which does not seem to be based on intimate knowledge of the Metaphysics of the Sˇifa¯ , and which in any case serves Averroes’ polemical and apologetic ends.25 But it is reasonable to ask how fair Averroes’ picture of Avicenna is, and, in particular, how far he is justified in making Fa¯ra¯bı¯ his ally against Avicenna. Avicenna too has enormous respect for Fa¯ra¯bı¯, which makes Averroes’ use of Fa¯ra¯bı¯ all the more rhetorically effective. Avicenna famously claims to find in Fa¯ra¯bı¯’s On the Aims of the Metaphysics the key to understanding the object and structure of Aristotle’s Metaphysics and thus the object and structure of metaphysics as a science. Averroes, however, claims to find affinities rather between Avicenna and the people Fa¯ra¯bı¯ is criticizing in the Kita¯b al-Huru¯f, that is (whether Averroes knows it or not) mainly the Kindı¯ ˙ 24 See my Review of Kenny, Aquinas on Being. I develop these points further in work in progress. 25 The most notorious discrepancy between Averroes’ Avicenna and the Metaphysics of the Sˇifa¯ is that Averroes takes Avicenna’s things ‘contingent in themselves, necessary through another’ to be the heavenly bodies or their movers, not recognizing that Avicenna gives this status to everything except God (see Davidson, Proofs for Eternity, pp. 318 – 20). Also Averroes says he thinks that for Avicenna ‘mawgˇu¯d’ is said univocally of things across categories (TT 370,4 – 7; and, without the hesitation, Epitome of the Metaphysics III,38 – 9), whereas Avicenna in Sˇifa¯ Metaphysics I,5,21 says that it is said of them by priority and posteriority. ˘

˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

69

70

˘

circle. And it is not out of the question that, despite Avicenna’s debts to Fa¯ra¯bı¯, he does have important agreements with the Kindı¯ circle or the Arabic neoPlatonica against Fa¯ra¯bı¯, not on the issue of the object of metaphysics, but on the analysis of the concept of being. (And, of course, he may perfectly well be right to disagree with Fa¯ra¯bı¯ on this or other issues.) This is a very large issue on which I will make only a few basic comments here. While Avicenna had of course read Fa¯ra¯bı¯, I have not found clear evidence that he had read the Kita¯b al-Huru¯f. If he did, evidently he did not agree with ˙ everything he read. In particular, Avicenna seems never to interpret ‘being as the true’ from Metaphysics D7 as Fa¯ra¯bı¯ does, as the sense of being found in assertions like ‘there is a phoenix’, ‘there are lunar eclipses’, and as signifying a second intelligible rather than something in re. 26 Indeed Avicenna, like Kindı¯, seems to take Metaphysics a as a model for metaphysical argument, and in his brief version of Metaphysics a1 at the beginning of Sˇifa¯ Metaphysics I,8 he says that ‘true’ [haqq] is said of what exists in re [mawgˇu¯d fı¯ ’l-a ya¯n], and is said in a ˙ stronger degree of what exists eternally and of the causes of other things’ existing, or of propositions which are true eternally and which cause other propositions to be true (Sˇifa¯ Metaphysics I,8, following Aristotle Metaphysics a1 993b23-31–contrast Fa¯ra¯bı¯’s insistence that being as truth is univocal). Avicenna says here that every contingent existent is ‘true through another [sc. and thus ultimately through God] and false in itself ’ (I,8,1), which sounds close to Kindı¯.27 On the other hand, Avicenna comes strikingly close to Fa¯ra¯bı¯ when he distinguishes two senses of mawgˇu¯d and of wugˇu¯d, wugˇu¯d in the sense of affirmation or positing [wugˇu¯d itba¯tı¯] and the wugˇu¯d proper to a thing [wugˇu¯d ¯ ha¯ss : perhaps, roughly, a thing’s being what it is], corresponding respectively to ˙ ˘Fa¯˙ra ¯bı¯’s wugˇu¯d-as-truth and wugˇu¯d-as-having-an-essence (I,5,9); and he agrees with Fa¯ra¯bı¯ that mawgˇu¯d in the second of these senses is equivalent with ˇsay

˘

˘

˘

˘

26 This despite the fact that Avicenna takes over Fa¯ra¯bı¯’s theory of second intelligibles as the objects of logic. Usually Avicenna says ‘second intention’ [ma na¯ ta¯nı¯] but sometimes he says ‘second intelligible’ [ma qu¯l ta¯nı¯] like Fa¯ra¯bı¯, and there is no¯difference in meaning; ¯ the subject-matter of logic is al-ma a¯nı¯ al-ma qu¯la alat Sˇifa¯ Metaphysics I,2,4 he says that ta¯niya which depend on al-ma a¯nı¯ al-ma qu¯la al-u¯la¯. For Avicenna on second intentions ¯as the object of logic, see Sabra, Avicenna on Logic; but Sabra does not have a full picture of the Fa¯ra¯bian background, and does not discuss the Kita¯b al-Huru¯f. ˙ is the Truth 27 ‘The cause of the existence [wugˇu¯d] and affirmation [taba¯t] of everything ¯ [inniyya] has a truth/reality [al-haqq–i.e. God], since everything that has a being ˙¯qa]; so the Truth exists [mawgˇu¯d] for beings [inniyya¯t] which exist. And the most [haqı ˙ noble part of philosophy and its highest in degree is first philosophy, I mean the knowledge of the first Truth which is the cause of every Truth’ (Kindı¯, On First Philosophy 9,12 – 14 in Kindı¯, Oeuvres Philosophiques et Scientifiques, eds Rashed and Jolivet). Kindı¯ is here reflecting on the same passage of Aristotle, Metaphysics a1 993b23 – 31, as Avicenna in Sˇifa¯ Metaphysics I,8. ˘

˘

˘

˘

˘

˘

˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Stephen Menn

71

Averroes against Avicenna on Being and Unity

˘

[thing or something], and that the wugˇu¯d in this sense is identical with the quiddity. Thus Avicenna’s differences with Fa¯ra¯bı¯ seems to be concentrated on the status of wugˇu¯d in the sense of affirmation: Fa¯ra¯bı¯ thinks it exists only in the mind and produces no real composition with the essence of the thing, whereas Avicenna thinks that it is really composed with the essence in re. And the reason for this difference seems to be a different view of what subject wugˇu¯d is being predicated of when I say ‘F is mawgˇu¯d’ in sentences like ‘there is a phoenix’, ‘thunder exists’. Fa¯ra¯bı¯ and Avicenna agree that ‘F is mawgˇu¯d’ must be capable of being false without being meaningless, and therefore that ‘F’ in such sentences must signify something which it can signify whether it exists in re or not: whatever ‘F’ signifies here will be the subject of which wugˇu¯d is being predicated. Fa¯ra¯bı¯ says that in this context ‘F’ signifies the concept of F, and to say that F is mawgˇu¯d means that this concept is instantiated. Avicenna rejects this: perhaps his reason is that, when I say that thunder exists, I am not saying that some concept exists. At the same time, Avicenna agrees with Fa¯ra¯bı¯ that when I say ‘F is mawgˇu¯d’ in this sense, I am not referring to some individual F (some instance of the concept of F) which might be either mawgˇu¯d or ma du¯m [non-existent], as some Mu tazilites thought (basing themselves on the Qur a¯nic ‘when God decrees a thing, he just says to it “be!” and it is’ [3:47, 40:68, slightly differently 36:82] where what God is addressing must be a non-existent thing [sˇay ma du¯m]; for Avicenna’s attack on the Mu tazilite thesis see I,5,12 – 18 and cf. I,5,25 – 7). Avicenna’s alternative is to say that what ‘F’ signifies here is an quiddity which is neutral to existence in re or in the soul: it can never be absolutely non-existent, it must always exist either in re or in the soul, and it is this neutral quiddity rather than a concept (and rather than a Mu tazilite notyet-actualized individual) of which we affirm that it is mawgˇu¯d not only in the soul but also in re; and so it is this quiddity to which wugˇu¯d is superadded to constitute a really existing individual. Assuming Avicenna’s objection to Fa¯ra¯bı¯’s account is as I have suggested, Fa¯ra¯bı¯ might reply that when I say ‘F is mawgˇu¯d’, I am not saying that the concept of F exists, but rather than the concept of F is instantiated: although the grammatical structure of the sentence suggests that I am asserting of something that it exists, the underlying logical structure does not assert either of some object or of some concept that it exists, but of some concept that it is instantiated. This is what Fa¯ra¯bı¯ ought to say, but it is not clear that he actually does: Fa¯ra¯bı¯ says that the mawgˇu¯d as the true is what ‘is outside the soul as it is inside the soul’ (Huru¯f, p. 116,5, cited above), but what is the ‘it’ that is both ˙ outside and inside the soul? If Fa¯ra¯bı¯’s wording is to be taken at face value, the ‘it’ of which being-as-truth is predicated is not the concept but an Avicennian neutral quiddity which can be realized either in the soul or outside the soul; if ˘

˘

˘

˘

˘

˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

72 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Stephen Menn

Fa¯ra¯bı¯’s wording is not to be taken at face value, it is incumbent on him to explain what the correct formulation would be.

II Averroes against Avicenna on Being and Unity in the Epitome of the Metaphysics The Epitome of the Metaphysics gives the clearest picture of how Averroes sees the main results of the Metaphysics. The Taha¯fut al-Taha¯fut is not about Aristotle, ˙ aza¯lı¯; the and discusses metaphysical issues only piecemeal as they are raised by G Great Commentary is of course full of information, but in most places sticks so closely to the text that it is hard to see the larger picture. But in the Epitome of the Metaphysics, as in his other Epitomes, Averroes does not feel bound to follow the order of Aristotle’s text, but rather restates what he takes to be the chief scientific contributions of Aristotle’s text in whatever he finds the most appropriate scientific order, eliminating (for instance) refutations of now deceased errors, but adding discussions of more pressing errors that have arisen since Aristotle’s time – and, in particular, those of Avicenna. I have argued elsewhere, and recalled above, that Fa¯ra¯bı¯ also sees himself in the Kita¯b al-Huru¯f ˙ as restating central contributions of the Metaphysics in this way for a new linguistic-religious community, and in particular sees himself as reworking Metaphysics D (far more central to the Metaphysics for Fa¯ra¯bı¯ than it has been for most readers) in Part I of the Huru¯f. What emerges from Averroes’ Epitome of ˙ the Metaphysics is that he too read Part I of the Huru¯f in this way, and took it as ˙ his immediate model for the part of the Epitome corresponding to D ; and that he took the confusions Fa¯ra¯bı¯ was warning against to be close enough to Avicenna’s confusions that he thought that a close restatement of Fa¯ra¯bı¯ would serve to refute Avicenna as well. Because he is trying to bring out what he sees as Aristotle’s main underlying points in the Metaphysics, without being bound to Aristotle’s text, he makes more systematic and structural use of Huru¯f here than ˙ in the Taha¯fut al-Taha¯fut or the Great Commentary on the Metaphysics, although without ever naming Fa¯ra¯bı¯; but the dependence on the Huru¯f is often very ˙ close. I will not document this in full (it is generally immediately obvious on comparing the texts, and was noted by Mahdi in his introduction to the Huru¯f, ˙ p. 39). Instead, I will first briefly review what Averroes says in the Epitome about mawgˇu¯d and wugˇu¯d – which adds no real content to what we have already seen, but which makes his use of the Huru¯f and its relation to D more obvious – and ˙ then go on to something that does add real content, the Epitome’s very interesting account of the senses of unity and multiplicity, and criticism of Avicenna on unity and multiplicity, drawn not from the Kita¯b al-Huru¯f but ˙ from the Kita¯b al-Wa¯hid wa’l-wahda. ˙ ˙

Averroes against Avicenna on Being and Unity

Senses of Being and Senses of Unity in Epitome I In the Epitome of the Metaphysics, before settling down to the systematic study of substances and their attributes and principles, Averroes gives in Book I first a relatively brief discussion of the science of metaphysics and its subdivisions and their objects, corresponding structurally to Metaphysics E1 (Epitome I,1 – 17), and then a much longer discussion of the different meanings of each of the terms that signify the things treated in the science (I,18 – 66). This latter part of Book I corresponds to Metaphysics D, but in fact for much of this text Averroes is immediately dependent not on D but on Part I of the Kita¯b al-Huru¯f, much of ˙ which he extracts or closely paraphrases, without ever naming his source. In particular, in his account of the meanings of ‘mawgˇu¯d’ and ‘huwiyya’ (I,19 – 23) is heavily dependent on Fa¯ra¯bı¯’s. Like Fa¯ra¯bı¯, he distinguishes between two main senses of ‘mawgˇu¯d’, the sense that is said of the ten categories per prius et posterius and the sense that is said of the true, glossed as in Fa¯ra¯bı¯ as ‘what is in the mind as it is outside the mind’; as in Fa¯ra¯bı¯, being-as-the-true is illustrated with examples of the eQ 5sti question from the Posterior Analytics, whether nature exists, whether void does not exist (I,19).28 Because of the paronymous form of the word ‘mawgˇu¯d’, and because it is transferred from the sense of ‘found’, some people thought that it signified an accident in the thing rather than its essence [da¯t], but Averroes explains that in fact we are using it the word ¯ (in one of its senses) to signify what is signified by ‘thing’ and ‘essence’ and other non-paronymous terms (I,21). Averroes also adds that some people thought that the two main philosophical senses of ‘mawgˇu¯d’ were the same, where presumably this is the error that he attributes to Avicenna in Taha¯fut alTaha¯fut Discussion 7, of thinking that the translators always used ‘mawgˇu¯d’ to mean the true. Also, some argued that ‘mawgˇu¯d’ must signify an accident, since if it signified the essence, then when we say ‘substance is mawgˇu¯d’ (i. e. ‘a substance exists’), it would be a tautology (still I,21); this is in fact an argument that Avicenna makes in Sˇifa¯ Metaphysics I,5,10 – 11 to show that the wugˇu¯d of a thing in the sense of affirmation [itba¯t] is different from its essence [haqı¯qa] or ¯ thingness. Averroes has presumably been thinking mainly of ˙ Avicenna throughout I,21, but I,22 is explicitly a refutation of Avicenna’s thesis that ‘mawgˇu¯d’ signifies an accident in the thing: if it is a first intelligible, then either it must belong to one of the nine categories of accidents, in which case being won’t be predicable of substance or of the other categories of accidents except inasmuch as this accident inheres in them, or else it is an accident somehow ˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

73

28 Averroes subdivides the first main sense of ‘mawgˇu¯d’ into the things that have a quiddity outside the soul and the quiddities of these things, but he notes that the ‘mawgˇu¯d’ applies to the ten categories in both of these senses (I,19). Beyond the two main senses, he also mentions being per accidens (I,20).

74

˘

˘

predicated in common of the ten categories, which is the thesis that in the Taha¯fut al-Taha¯fut (Discussion 7, 370,4 – 11) he says Avicenna probably held, and here as there he says that it is absurd. So if ‘mawgˇu¯d’ signifies an accident, the only possibility is that it is a second intelligible, i. e. an intelligible whose existence is only in the mind; Averroes says that this is possible and indeed correct for ‘mawgˇu¯d’ in the sense of the true, but that this is very different from ‘mawgˇu¯d’ meaning what has a quiddity outside the mind; ‘all this is clear on the slightest reflection, but that is the nature of this man [sc. Avicenna] in much of what he brought forth out of himself ’ (all I,22). Thus far, as we have seen, Averroes is closely following the Kita¯b al-Huru¯f. ˙ to But when in Epitome of the Metaphysics Book I, in the section corresponding Metaphysics D6, Averroes comes to deal with the concepts of unity and multiplicity, he takes as his model, not the Kita¯b al-Huru¯f, which indeed has no ˙ account of unity, but Fa¯ra¯bı¯’s Kita¯b al-Wa¯hid wa’l-wah da.29 Metaphysics D of ˙ ˙ course has an account of unity (D6) alongside the accounts of being (D7) and of the other concepts that Averroes discusses in Epitome of the Metaphysics Book I; since Averroes’ discussions follow the Kita¯b al-Wa¯hid wa’l-wahda for unity and ˙ much all the multiplicity, and the first Part of the Kita¯b al-H˙uru¯f for pretty other concepts, it seems that he thought of ˙ these two texts together as constituting Fa¯ra¯bı¯’s updated version of D, and the model for his own updated version. With unity, as with being, Averroes’ overriding aim in the Epitome is to expose what he sees as Avicenna’s errors, and again he thinks that the errors Fa¯ra¯bı¯ is exposing are the same or close enough that Fa¯ra¯bı¯’s distinctions will still be the relevant ones to draw. The main relevant texts here are Epitome I,34 – 43 and III,34 – 48. As usual, Averroes mentions Avicenna by name only a few times, but in such a way as to make clear that Avicenna is on his mind throughout. The main aim is to refute, or restrict the scope of, Avicenna’s claim that being and unity, as predicated of things other than God, signify something superadded and accidental to the essence of the thing. Avicenna goes beyond Aristotle’s theses that being and unity are not parts of the essence of a thing (and therefore should not be mentioned in its definition), and that unity is a per se attribute of being, to the conclusion that both being and unity are per se attributes of the most universal quasi-genus, al-sˇay , ‘thing’, consequences of it rather than contained in its essence. Nor are they contained in the essence of any particular species of al-sˇay : ‘in itself it is nothing at all but horseness: in itself it is neither one nor many, neither existent in re [fı¯ ’l-a ya¯n] nor existent in the soul, neither potentially nor actually in any of these things in such a way that this would enter into horseness’ (Sˇifa¯ Metaphysics V,1,4). Such an essence is capable of receiving existence in re and thus individuality, or existence in the mind and thus universality, and it is ˘

˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Stephen Menn

29 As Mahdi notes in his Arabic introduction to the Kita¯b al-Wa¯hid wa’l-wahda, p. 19. ˙ ˙

Averroes against Avicenna on Being and Unity 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

75

likewise capable of receiving unity or multiplicity. Averroes’ response in the case of unity is parallel to his response in the case of being. He thinks that Avicenna has misinterpreted the reasons why being and unity are not mentioned either as genera or as differentiae in the definitions of things; and Averroes thinks that the idea of something that is in itself not existent or one, and receives existence and unity as accidents, involves the same absurdities that Aristotle attributes to the Platonists who generate the many things from being and what is not in itself being but comes to participate in being, or from the one and what is not in itself one but comes to participate in unity. And Averroes’ way of avoiding these absurdities, and of diagnosing and avoiding Avicenna’s conclusion, is again closely parallel in the case of unity and in the case of being. With unity, as with being, he distinguishes two main senses of these attributes. In the sense of mawgˇu¯d as having a quiddity outside the mind, man is mawgˇu¯d by his essence, and the reason that mawgˇu¯d is not mentioned in the definition alongside wingless and biped and animal is that a wingless biped animal is already mawgˇu¯d, that ‘mawgˇu¯d’ signifies the whole quiddity rather than signifying some part that could be superadded to another part that is not already mawgˇu¯d by its essence; and Averroes will say that there is likewise a sense of ‘one’ in which man is one by his essence, and in which ‘one’ signifies the whole quiddity rather than a part of the quiddity or an accident superadded to the quiddity. In the sense of ‘mawgˇu¯d’ as the true, Averroes concedes to Avicenna that ‘mawgˇu¯d’ signifies an accident not included in the quiddity, but he argues that this is something existing only in the mind, and he likewise concedes that there is a sense of ‘one’ in which it signifies something not included in the quiddity, but again he argues that this is something existing only in the mind. Averroes seems to think that these conclusions, in the case of unity as in the case of being, follow from Fa¯ra¯bı¯’s distinctions. Fa¯ra¯bı¯ does indeed distinguish senses of ‘one’, and he does indeed think that one of these senses of ‘one’ is equivalent to the sense of ‘mawgˇu¯d’ as signifying the categories and to ‘thing’, and that a thing’s unity in this sense is just its quiddity; it is less clear that he thinks, as Averroes does, that ‘one’ has a second main sense in which the thing’s unity exists only in the mind. Senses of Unity and Ways of Being Delimited Epitome of the Metaphysics I,34 – 43 are devoted to unity, corresponding structurally to Metaphysics D6; much of the content is also drawn, directly or indirectly, from Metaphysics D6. But as Averroes’ discussion of mawgˇu¯d differs from Aristotle’s by his interest in determining the wugˇu¯d by which things are mawgˇu¯d, and in particular whether this wugˇu¯d is a real accident of the things or is identical with the things or exists only in the mind, so his discussion of ‘one’ differs from Aristotle’s by his interest in the unity by which each thing is one,

76 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Stephen Menn

whether this is a real accident or the essence of the things or exists only in the mind. Both here in Epitome I,34 – 43 and later in III,34 – 48 (which corresponds structurally to Metaphysics Iota 1 – 2)30 this interest leads Averroes to focus on two particular issues neither of which is really Aristotelian: first, the thesis, taken from Fa¯ra¯bı¯’s Kita¯b al-Wa¯hid, that various kinds of things are called ‘one’ ˙ a¯z] from other things, either by a place or by because they are ‘delimited’ [munh a boundary or by a quiddity; and ˙second, the question how ‘the one which is the principle of number’ (and which is, like number, an accident in the category of quantity) arises and how it is distinguished from other senses of ‘one’. On both of these issues, Averroes is in direct confrontation with Avicenna. Epitome I,34 – 43 is a rearranged and rationalized version of Metaphysics D6. Averroes skips the long treatment of one per accidens which Aristotle insists on putting at the beginning of D6 (1015b16 – 36), and he also plays down the treatment of unity in species or genus or by analogy (D6 1016a17-b1, 1016b31 – 1017a3), inserting short versions of both of these toward the end, in I,41: he concentrates heavily on the different ways in which something is called one in number (corresponding roughly to D6 1015b36 – 1016a17 and 1016b11 – 31). In ordinary usage, something is said to be one in number in the strongest sense if it is continuous (either per se, like a one-, two- or threedimensional continuous quantity, or through an accident present in it, like a homoeomerous natural body which is continuous through possessing a threedimensional continuous quantity), especially if it is something complete, like a circle (Aristotle’s example, D6 1016b16 – 17); something is called one in number in a weaker sense if it is composed of several juxtaposed homoeomerous bodies bound together in such a way that they all move together when any of them is moved (it is one in a stronger sense if it is composed by nature, like a hand, in a weaker sense if it is composed by art, c. D6 1016a4 – 6); and also ‘an individual one in form’, like Socrates, is called one, presumably not just in the way that his right hand is called one, but also because, while he is made up of different juxtaposed homoeomerous bodies, they are all contained by a single complete natural form. Thus far (Epitome I,34 and the beginning of I,35) Averroes is broadly following his Aristotelian model, but he then adds a reflection on the kinds of unity or oneness on account of which ordinary people [al-gˇumhu¯r] call things one in number: these people call these things one only ‘inasmuch as they are delimited [munha¯z] from other things and isolated ˙ [munfarid] in themselves’ (I,35).31 Different senses of ‘one’ correspond to 30 The parallel with Metaphysics Iota continues through Epitome III,63. Epitome III,34 – 48 correspond to Iota 1 – 2 on unity, and then III,49 – 63 discuss multiplicity, sameness, difference, contrariety, opposites and intermediates, corresponding to Iota 3 – 10. 31 The word ‘munfarid’, which I have translated literally enough as ‘isolated’, is sometimes (but not consistently) used in the translation-literature for ‘jah’ 6jastom’, ‘particular’ or

Averroes against Avicenna on Being and Unity 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

77

different ways in which a thing can be delimited from other things. Most obviously, something can be delimited by a place which contains it. Other things are delimited not by a place but just by their boundaries [niha¯ya¯t], ‘and they are contiguous’: the sense is apparently that the boundary of body A and the boundary of body B are ‘together’, i. e. that these boundaries are in the same place, so that bodies A and B, without being separated, are divided from each other by real boundaries, which become two really distinct surfaces, although in the same place, when the division is effected (see Aristotle, Physics V,3). Still others have ‘their delimitation by imagination alone, and in this way number attaches to what is continuous’: what is continuous has no actual internal dividing boundaries (but only the potentiality for being so divided) and thus a three-foot length does not contain three feet actually delimited from each other by boundaries, but nonetheless we can measure it by delimiting these parts in the imagination. In all of these cases – in all the cases recognized by the gˇumhu¯r–‘one in number’ signifies ‘things which are external to their essences, and in general, accidents attaching to them which are inasmuch as they are undivided’ (all I,35). Averroes argues that Avicenna was misled by these ordinary-language cases and overgeneralized to the conclusion that every numerical or individual unity, that is, every delimitation [inhiya¯z], is an accident of the thing which is one. ‘In ˙ this art’ – metaphysics, as opposed to ordinary-language contexts – ‘the one is used equivalently [mura¯difan] with being’ (I,37), and Averroes argues that unity in this sense is not an accident. ‘By the one in number is sometimes signified an individual which cannot be divided inasmuch as it is an individual, like our saying one man and one horse’ – a human being can be divided, but not into two human beings (the example comes from Aristotle, D6 1016b3 – 6). But to show more clearly that the relevant kind of unity is not an accident, and is distinct from the ordinary kinds of unity that he has listed, Averroes turns to a different kind of example, a blend such as oxymel (a blend of honey and vinegar, used as a medicine). The concern is evidently not with the sense in which the oxymel itself is one, but with the sense in which the honey and the vinegar are each one and delimited from each other. Because oxymel is a real blend and not just a juxtaposition, the honey and the vinegar are not delimited from each other by spatial boundaries, like contiguous things; but they are also not like the parts of a continuous magnitude, which have no determinate demarcation in themselves, since honey and vinegar are distinct by their own natures (all I,37). So, while ‘the delimitation of continuous magnitudes is something external to their substance, this is not so for the delimitation of a ‘individual’ (so for instance at Metaphysics Iota 1 1052a32 and a35). But the Arabic, unlike the Greek, strongly connotes isolation or separation from other things, and, as a passive form, raises the question, ‘isolated by what?’.

78 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Stephen Menn

blended thing from what is blended out of it’ (ibid.): what makes the honey in oxymel one thing, and the vinegar in oxymel one other thing, is that each possesses, and is ‘delimited’ from the other by, its own intrinsic nature, not by a distinct place or boundary but also not merely in the imagination. It is clear that by ‘one’ here, when there is meant by it what is individually one, there is signified only the delimitation of the designated [musˇa¯r ilayhi, rendering ‘t|de ti’] individual by its essence and its quiddity, not the delimitation by anything external to its essence, as when we say about this designated water that it is one in number: for the delimitation, in something like this, is only an accident in the water, and therefore the water remains the same when it is delimited and when it is not delimited, in the way that accidents succeed each other in the substratum without it changing in its substance. (I,38)

That is: I can say that this water is one, in the sense that it is continuous: this is to assert that it is distinguished from everything else by being delimited by a determinate bounding surface, and, negatively, that it is not divided by any further boundaries internal to that surface. The unity of the water, in this sense, is an accident superadded to the water, since the water can persist in its substance whether it is divided into many drops or reunited into a continuous mass. But any individual can also be called ‘one’ as being ‘delimited’ by its own quiddity, an attribute which all of it possesses and which distinguishes it from everything else; and its unity in this sense is essential to it, and the thing could not continue to exist if this unity were removed.32 Averroes thinks that Avicenna, overgeneralizing from cases like the one-as-continuous, thought that ‘one in number’ always signifies a delimitation which is an accident superadded to the essence of a thing. He would be doing this, not exactly because he did not recognize a distinctively metaphysical sense of ‘one’ in which it is equivalent to ‘being’, distinct from the ordinary-language senses, but because he did not recognize that this metaphysical sense had a distinct logical syntax, and assumed that the syntax of the ordinary-language senses applied to it as well.

32 But note that this holds only if the object has a quiddity, that is, if it is mawgˇu¯d in the sense divided into the categories. Averroes in the Great Commentary on the Metaphysics (following a misleading translation) takes Iota 2 1053b20 – 21 to say that ‘mawgˇu¯d’ is more widely extended than ‘one’, and he explains this by saying that mawgˇu¯d in the sense of the true need not be one (Tafsı¯r 1271,12 – 17). Thus e. g. the concept ‘pile of stones’ can be instantiated, and yet the pile of stones, not being delimited by a quiddity outside the mind, will be not one but many.

79

Averroes against Avicenna on Being and Unity

Against Avicenna: Why Unity in the Metaphysical Sense is not an Accident However, Averroes thinks that Avicenna also had a supplemental argument to support this conclusion, and therefore that to avoid Avicenna’s conclusion (and the ensuing difficulties), we need not only to point out the difference between the popular and the metaphysical senses of ‘one’, but also to respond to this argument: For this reason Avicenna thought that ‘one in number’ signifies only an accident in substance and in anything else which is delimited, and that it cannot signify the substance of a thing, I mean [that it cannot signify] a delimitation which is not superadded to the essence of the substance. This is because he thought that if he granted that ‘one in number’ signifies a delimitation which is an accident in the accident and a substance in the substance [i.e. that what ‘one in number’ signifies, when it is said of X, is really identical with X and so falls under the same category as X], then number would be composed of accidents and substances, and so would not fall under a single category, much less under the category of quantity, and this would be absurd. He also said that if we stipulate that it signifies substance alone, another absurdity would follow, namely that substances would inhere in accidents; and otherwise how will we say about a designated accident that it is one in number? And he went wrong in this only in that he paid attention to the popular [gˇumhu¯rı¯] signification and thought that the delimitations and unities of things were accidents in all delimited things. We will explain this further in our account of the one and the many [i.e. in Epitome III,34 – 63, corresponding to Metaphysics Iota]. (I,39)

Avicenna’s point would be: a number is an accident in the category of quantity (it is a peculiar kind of accident, because it seems to be predicated not of any one subject but of many at once, but Aristotle indisputably says that the category includes both continuous quantities and discrete quantities). But a number is composed of ones, and these ones must be homogeneous with what is composed out of them, which is to say that they too must be accidents of discrete quantity, i. e. unities by which some thing is one. If things were one by their essence, so that the unity of X would be in whatever category X is in, then these unities could not combine to compose numbers. This is in fact a fairly accurate summary of some of Avicenna’s arguments in Sˇifa¯ Metaphysics III (although Avicenna seems not to use the concept of ‘delimitation’, which Averroes takes from Fa¯ra¯bı¯). In particular, Avicenna argues that the unity of an accident cannot be a substance inhering in the accident (III,3,10); and while he does not seem to raise the possibility that a single number might be composed out of some substantial unities and some accidental unities, he does argue that if ‘one’ were said non-univocally of substances and accidents, so that it would signify a substance in substances and an accident in accidents, then there would be some numbers composed out of substantial unities and others composed out of accidental unities (III,3,13), which he takes to be absurd, presumably for the reason Averroes suggests, that these numbers could not all fall under the same ˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

80

category. (Avicenna also gives other arguments that unity in substances cannot be a substance, e. g. by arguing that it is not a constituent in their definition either as a genus or as a differentia, and therefore must be an accident, III,3,10.) Averroes, in response, agrees that numbers must be accidents composed out of accidents of unity, and that if things were simply one by their essence they could not be counted by numbers belonging to the category of quantity. But Averroes says, as usual, that Avicenna has confused two different senses of ‘one’, in this case the distinctively metaphysical sense and ‘the one that is the principle of number’. (The metaphysical sense, ‘individual unity’, can also be called ‘one in number’, but not in the sense that there are numbers composed out of it, at least not numbers in the category of quantity, the objects of arithmetic.) And indeed Avicenna is very insistent that there is a univocal sense of ‘one’ that applies (as an accident) to beings in all categories (III,3,13 – 15), and that the unity from this universal metaphysical predicate is also the unity out of which numbers are composed (III,3,17). Averroes agrees with Avicenna that the kind of one out of which numbers are composed is an accident of unity, but he tries to show that this kind of unity, and also the numbers arising from it, exist only in the soul, contrasting with other senses of unity, including the metaphysical sense equivalent to being, which exist in re independently of souls. (By contrast, Avicenna not only thinks that the unity which is the principle of number exists in re, he also insists that numbers themselves exist in re independently of souls, directly contradicting Aristotle, Physics IV,14 223a22 – 5: at III,5,2 he rejects the formula ‘number has no wugˇu¯d except in the soul’ and corrects it to say ‘number has no wugˇu¯d abstracted from the numbered things which are in re [fı¯ ’l-a ya¯n], except in the soul’ – just as horseness has wugˇu¯d in an individual, and wugˇu¯d in the soul, but no wugˇu¯d abstracted from both.) Averroes says a bit about this in Epitome I,36, as part of his survey of the various ordinary-language senses of ‘one in number’ in I,34 – 6 before he gets to the distinctively metaphysical sense in I,37 – 40; he adds more detail in Epitome III. ˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Stephen Menn

Against Avicenna: Quantitative Unity is Mind-Dependent Averroes says in I,36 that ‘in this way [i.e. from accidents attaching to things inasmuch as they are undivided] there arises in the mind [fı¯ ’l-dihn] the one ¯ from these which is the principle of number, because when the intellect abstracts individuals this meaning, “indivisible into two or more individuals”, this is the one which is the principle of number, and when the mind repeats it, number arises.’ Like the unity of what is delimited by a place or a boundary, this is one of the senses of individual unity that are recognized by ordinary people [algˇumhu¯r]; in all of these cases, something is one through an accident, and so they all contrast with the metaphysical sense of unity, or as Averroes also calls it ‘one

Averroes against Avicenna on Being and Unity 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

81

absolutely [al-wa¯hid al-mutlaq]’, in which a thing is one by its essence. Unlike ˙ the unity of what˙ is delimited by a place or a boundary, the unity that is the principle of number is an act of the soul: ‘the one in number and numerical unity is only something which the soul produces in existing things; were it not for the soul, there would not exist here numerical unity or number at all’ (III,41). As this text says, the soul produces this unity in existing things: the thing that is one is something existing outside the soul, but the unity through which it becomes one is an act of the soul – just as, more famously, intellectus est qui facit universalitatem in rebus, so that the things which are universal exist outside the intellect but it is an act of the intellect which makes them universal.33 Why do numerical unity and number depend on the soul in this way, when (Averroes says by contrast, III,41) continuous quantities of whatever dimension do not? All of these things, as the mathematician studies them, depend on some sort of act of abstraction – the surface which the mathematician treats may be the boundary of a natural body, but he abstracts from this and does not treat it qu boundary of a natural body (III,37) – but number is ‘more abstracted from matter’ than continuous quantities (III,41). The reason emerges from III,35: ‘the numerical one is the thing designated [al-musˇa¯r ilayhi] in thought, not divided in it into quantity’ (a parallel in III,36 adds ‘or into quality or position’). And, although Aristotle never says that unity depends on the soul, this rests on what he says in Metaphysics Iota, that the one in each genus is something either indivisible or taken as indivisible that measures the things in that genus: in many cases, as with the foot as a measure of length or the semitone as a measure of musical intervals, no genuinely indivisible measure is available, and we must treat something divisible as if it were indivisible; and Averroes reasonably concludes that the indivisibility or undividedness which makes it one is conferred by an act of the soul, not merely abstracting from irrelevant features of the object, but treating the object as if it possessed a feature which it does not possess apart from the soul’s act of considering it as an indivisible unit. Averroes stresses the difference between ‘the one absolutely’ and ‘the one as the principle of number’ that the one absolutely is always predicated of a subject in some particular category, and, because it is predicated of its subject essentially, always itself belongs to the same category as its subject, whereas ‘the one which is the principle of number, if it is grasped inasmuch as it is one, does not manifest its dependence on the subject, and only in this way can it fall under the category of quantity’ (III,38). ‘The numerical one is the meaning of the 33 This is the form in which the phrase is standardly cited in Latin; it’s sometimes cited as coming from Averroes’ Great Commentary on the De anima, Book I, text 8, but there the Latin apparently has ‘intellectus est qui agit in eis universalitatem’ (so in Averroes, Commentarium magnum in Aristotelis de anima libros, ed. Crawford, p. 12, lines 25 – 6).

82

individual, abstracted from quantity and quality, I mean that by which the individual is an individual, since it is an individual only by the meaning “undivided”, and the mind abstracts it from matters and grasps it as a separate meaning’ (III,41).34 Once we have abstracted such unities from various individuals, we can collect them into numbers by the act of counting, even if the individuals themselves belong to different categories, and so cannot be called ‘one’ (or anything else) univocally: entities that really exist in such diverse subjects could not be aggregated, but entities produced by the soul in considering them can be.35 Because Avicenna (says Averroes, III,41) failed to distinguish between the one absolutely and the one that is the principle of number, he thought that the one in number was an attribute really existing in its subject (Avicenna ‘wanted to make the case of number like the case of line and surface, I mean that it would have a nature even if there were no soul’, III,41): it would have to be an accident predicated univocally of its diverse subjects, so that the different unities could be aggregated into numbers, and he was thus confronted with the problem what the single subject is of which ‘one absolutely’ is univocally predicated. Evidently, it is being, but being is not a single univocal genus of the different categories; and so, Averroes says, Avicenna was forced to posit that the being which is the subject of the one is an accident predicated univocally of things in the ten categories (III,38 – 9)–that is, although Averroes does not say so here, that this being is the univocal accident being-as-truth, wrongly construed as really present in the things rather than as dependent on the soul.36

˘

34 Compare I,36 cited above. The phrase I have translated ‘only by the meaning “undivided”’ might instead be translated ‘only by an undivided meaning/entity’ – we could read either ‘bi-ma na¯ “g˙ayri munqasimin”’ or ‘bi-ma nan g˙ayri munqasimin’. But I think Averroes means the former: this seems to be what makes the individual an individual, and it seems to be an appropriate building-block of discrete quantities. There is a similar ambiguity in I,36, where the phrase I translated ‘this meaning, “indivisible into two or more individuals”’ might instead be translated ‘this meaning/entity which is indivisible into two or more individuals’. 35 Averroes is here probably thinking of Aristotle’s point in Metaphysics M6 – 7 that mathematical ones are all ‘comparable’ or ‘associable’ [sulbkgt\] with each other and so can be combined into numbers, while other kinds of ones need not be, so that it may not be possible (say) to form a two out of a one of this type and a one of that type. 36 So too, in the Great Commentary on Metaphysics Iota, Averroes takes the last colon of Iota 1, 1053b7 – 8, to distinguish one as indivisible in substance from one as indivisible in quantity, and to say that the ignoring of this distinction is responsible for the error of those who say that the one in discrete quantity is a substance, and therefore that numbers are substances (1267,8 – 14); and then Averroes adds on his own behalf that the same reason led to Avicenna’s believing ‘that the one which is the principle of number is a genus to the ten mawgˇu¯da¯t [i.e. the categories], in such a way that it signifies an accident common to them’, since he did not distinguish the one which is the principle of number from the one which is equivalent [mura¯dif ] with being (1267,15 – 1268,3). ˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Stephen Menn

Averroes against Avicenna on Being and Unity

Thus for Avicenna unity as a per se attribute of being (more precisely, unity as consequent on being and, like being, a per se attribute of ‘thing’) would be an accident in the category of quantity; and manyness or multiplicity (katra), as the ¯ per se attribute predicated of several beings jointly, would be number, as an accident in the category of quantity composed out of quantitative unities aggregated together. As Averroes correctly reports (III,43), Avicenna objected (in Sˇifa¯ Metaphysics III,3,5) to the standard definition of number as ‘multiplicity composed of unities or ones’, on the ground that multiplicity is not the genus of number but simply is number, and that it is essential to all multiplicity to be composed of unities. But, Averroes says, just as unity absolutely is more general than quantitative unity, so multiplicity absolutely is more general than number or quantitative multiplicity: even things in different categories can be called ‘many’ in the first sense (thus Aristotle can argue against Parmenides that if any substance has an accident, beings are more than one), whereas an accident in the category of quantity could not belong to things in several different categories, except that as a mental act of counting it can depend on the unities formed by the mind in abstracting from these categories. Indeed, if number were a real univocal accident as Avicenna describes it, then a number three, being three units, would itself be three in just the same way that three horses or three virtues are three, and there would be a regress to a further accident of threeness; if the number is the soul’s act of counting the three things, it is not itself (in the same sense) three and there is no regress.37 ˘

37 The infinite regress argument is not explicit, but this looks like what Averroes means in Epitome III,43 when he speaks of a difficulty arising from the fact that, on Avicenna’s view, number would be numbered like other numbered things. (Averroes says that this is in fact possible, but only so far as a number is an act of the soul in the things that are numbered.) Avicenna’s answer would be that the number three is not three units, although it is composed out of three units (see Sˇifa¯ Metaphysics III,5,4 – 8, drawing on Aristotelian texts such as Metaphysics Z17 and H3 which argue that the whole is not the same as all its parts but is something over and above them, and specifically that a number is not the same as all its parts but is something over and above them). So the three units will be three through a further accident, namely the number three, and the number three will not itself be three through any further accident. However, perhaps Averroes’ point is not that on Avicenna’s view the number three would have to be three through a further accident of threeness, but rather that numbers can, like other things, be counted (e. g. we can say that there are four prime numbers less than 10), that this will require a further accident of number, and that such an infinite sequence is not absurd in successive acts of the soul, but is absurd in real accidents. ˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

83

84 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Stephen Menn

Averroes on Aristotle’s Aporia about Whether Being and Unity Exist Separately Averroes in Epitome III,44 – 8 steps back to put the issues about unity in a larger perspective by giving his version of an aporia in Aristotle’s Metaphysics B (B4 1001a4-b25), which Aristotle raises there both about being and unity: are being and unity subsisting things whose natures are just being and unity, and which could be the first principles from which other things are derived, as Plato and the Pythagoreans (and Eleatics) say,38 or is it as the pre-Socratic physicists say, that ‘being’ and ‘one’ are always attributes of some other underlying nature, such as fire or whatever the thinker in question takes the first principles to be? Averroes gives his version of the history of this disagreement among the ‘ancients’, which is considerably more reminiscent of debates in Arabic philosophy than Aristotle’s original account was. He is willing to attribute a surprising degree of consensus to the ancients. Both the pre-Socratic physicists and their Platonist opponents agreed with Aristotle that ‘being’ [mawgˇu¯d] and ‘one’ are equivalent, and, more surprisingly, they agreed that there is a first one which is a principle of all other mawgˇuda¯t and a cause to them of their existing [their wugˇu¯d] and also of their being counted and measured and therefore being knowable. The physicists noticed that ‘there is in each genus a first one, which is the cause of the wugˇu¯d of each species of that genus, and the cause of these other species’ being measured and known’ (III,45), e. g. the semitone among musical intervals or the daily motion of the heavens among motions (the idea and the examples, but not the attribution to the physicists, are from Aristotle, Metaphysics Iota 1 – 2 and parallels). Each of these genera is said per prius et posterius, and is said first of some first thing, as ‘the hot is said of fire, and of the other things related to fire, per prius et posterius, and fire is the cause of the wugˇu¯d of other hot things and of their being measured and counted’ (ibid.), fire being the homogeneous unit by which hot things are measured. And since beings, too, are called beings per prius et posterius, substances primarily and accidents derivatively, they speculated that ‘there is a first mawgˇu¯d which is the cause of all the other mawgˇu¯da¯t’s being mawgˇu¯d and numbered and known, as the one is the cause of all the other species of number being mawgˇu¯d and numbered’ (ibid.). These are surprising things to attribute to pre-Socratic physicists: they sound more like Arabic interpretations of Aristotle. The example of fire and the analogy between heat and being, fire and the principles of being, comes from 38 Presumably the Pythagoreans say this about the one, not about being. Aristotle doesn’t mention the Eleatics in the formulation of the question, but in the course of the aporia he introduces Parmenides’ argument that (as Aristotle interprets it) if there is something which is just being itself, nothing else can exist (1001a29-b1, cf. Physics I,2 – 3).

85

Averroes against Avicenna on Being and Unity

˘

Aristotle Metaphysics a1, a text (not especially the fire comparison) favored notably by Kindı¯, and while Aristotle there says nothing about the one or measuring (indeed Aristotle never apparently never says that God is one at all, preferring ‘simple’, see K7 1072a31 – 4), for Kindı¯ it is the ‘true One’ which is the first cause of all beings as fire is of all hot things; and Averroes’ ‘first mawgˇu¯d which is the cause of all the other mawgˇu¯da¯t’s being mawgˇu¯d’ is virtually a quotation of the first sentence of Fa¯ra¯bı¯’s Perfect City, ‘the first mawgˇu¯d is the first cause of the wugˇu¯d of all the other mawgˇu¯da¯t.’ But Averroes represents these as matters of broad agreement: what is distinctive of the physicists is that, believing in the priority of sensible individuals, and being unaware of any but the material cause, they made the first mawgˇu¯d and first cause of wugˇu¯d whatever they thought the first underlying matter was, e. g. fire (III,45; Averroes is perhaps combining the merely analogical mention of fire in Metaphysics a1 with the mention of fire as a candidate for the underlying nature of the one or being at B4 1001a15). By contrast, Plato (not explicitly named, III,46), who was aware of the formal cause but wrongly conceived it as an intelligible or intention or thought [ma qu¯l = m|gla] existing apart from the mind and apart from the individuals that fall under it, ‘said that the universal one, common to everything which is called one, is the cause of wugˇu¯d of all the other mawgˇu¯da¯t which are said to be one, and the cause of their being measured’ (III,46). This, then, is Averroes’ version of the opposition between the Platonic and materialist sides of the aporia of Metaphysics B4 1001a4–b25, filled out with the help of the idea of a first cause of wugˇu¯d from Metaphysics a1 as interpreted by Arabic philosophers;39 he then gives in Epitome III,47 – 8 what he sees as the distinctively Aristotelian solution to the aporia. Against Plato, Aristotle concluded that ma qu¯la¯t or universals have their existence qu ma qu¯la¯t (as opposed to their existence in sensible instances) only dependently on the mind: so the universal ‘one’ cannot be realer than other things and the cause of wugˇu¯d to them. Sensible individuals fall into the ten categories, and there is no real one beyond these categories; but within each category of accidents there is some first one, not as a universal applying to everything in the category, but as a unit (with the usual examples), which is the cause of wugˇu¯d and of being measured to other things in that category; and so Aristotle concluded that within substance too there must be a first one which is the cause of wugˇu¯d to all other substances, and thus to all other mawgˇu¯da¯t as well, since their wugˇu¯d is only through substance. And since he concluded that it is a one separate from matter which is most ˘

˘

˘

39 Including Averroes himself, Tafsı¯r 14,13 – 15,5, culminating ‘al-wugˇu¯du wa’l-haqqu ˙ a¯tihi innama¯ ’stafa¯dathu gˇamı¯ u ’l-mawgˇu¯da¯ti min ha¯dihi ’l- illati fa-huwa ’l-mawgˇu¯du bi-d ¯ ¯ faqat wa’l-haqqu bi-da¯tihi wa-gˇamı¯ u ’l-mawgˇu¯da¯ti innama¯ hiya mawgˇu¯da¯tun wa-haqqun ¯ ˙ ˙ ˙ ˇ bi-wugu¯dihi wa-haqqihi’ (15,2 – 5). Aristotle himself in Metaphysics a1 is more concerned about causes of ˙truth than about causes of being. ˘

˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

86 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Stephen Menn

deserving of the names ‘one’ and ‘mawgˇu¯d’, this investigation of the first one comes to the same as the main investigation of the Metaphysics, ‘whether there is a separate substance which is the principle of sensible substance, or whether sensible substance is sufficient in itself for wugˇu¯d’ (III,48).40 Indeed, Averroes seems to be mimicking Fa¯ra¯bı¯’s discussion, in Kita¯b al-Huru¯f I,92, of how to ˙ investigate whether there is a first mawgˇu¯d,41 but rewriting the discussion for ‘one’ instead. Either way, Averroes’ point is that all that an honest Aristotelian can do by way of finding a first mawgˇu¯d or a first one is to investigate causally, step by step and category by category, rather than leaping from the many instances to a single universal: there is no universal mawgˇu¯d or universal one beyond the categories, either existing separate from the categories, as he thinks Plato believed, or a universal mawgˇu¯d common to the categories with a universal one as its per se attribute, as he thinks Avicenna believed.42 Avicenna is thus represented as a halfway step to Plato, ignoring the fundamental Aristotelian lesson that being and unity are not anything ‘beyond the categories’ [paq± t± c]mg]; presumably Averroes also thinks that Avicenna has answered the part of Aristotle’s aporia about being (rather than about unity) in a Platonic way in saying that God is just a separate wugˇu¯d that is not the wugˇu¯d of any essence. Averroes’ ‘Aristotelian’ answer to the aporia is much closer to what he has attributed to the physicists: it differs in recognizing non-material causes and immaterial substances, but it posits outside the mind only individuals, and it posits no being or unity that is not the being or unity of some determinate essence in one of the ten categories. (However, to listen to Avicenna rather than Averroes: from his own point of view, he too is giving an anti-Platonic answer to the aporia about unity. Indeed, 40 Much, although not all, of this has parallels in the Great Commentary on Metaphysics Iota 1 – 2. Averroes sees Aristotle in these chapters as searching for a ‘one in substance’ (for the Aristotelian basis of this phrase see Iota 2 1054a4 – 9), analogous to the ones in other categories or in sub-categorial genera like colors and sounds, and he thinks that this one in substance will be the first mover (Tafsı¯r 1277,15 – 1278,5). Perhaps due partly to a misleading translation, Averroes takes Iota 2 1053b11 – 15 as asking, not whether the one is itself a substance or an attribute of some other underlying nature, but whether ‘one’ signifies a non-sensible substance or a sensible substance (1269,12 – 1270,4); Aristotle will correct both the physicists’ answers that it is some sensible substance, and Plato’s answer that it is a non-sensible substance which is the universal one, by giving the correct non-sensible substance. 41 See Menn, Analysis of the Senses of Being, pp. 93 – 6. 42 ‘We have stipulated that the subject of the one absolutely is nothing over and above the ten categories … for [otherwise] the subject of the numerical one would have to be either something common to all ten categories, as Avicenna says, or something separate, as many of the ancients thought about the nature of the one that they [sc. the one and being?] are separate things’ (III,38). Averroes says that Aristotle will refute the Platonic view, and he himself turns to refute the Avicennian view in III,39.

Averroes against Avicenna on Being and Unity

he develops his version of this Aristotelian aporia at some length, Sˇifa¯ Metaphysics III,3,11 – 17, in support of his claim that unity is an accident, i. e. that unity could not subsist if separated from the substance to which it belongs. There cannot be an indivisible which is just indivisible, without any underlying subject which is indivisible: at a minimum it must be an indivisible wugˇu¯d. If unity is just indivisibility, it can never be separated from its subject, since it would then have to have some other subject, and its former subject would have to have some other unity, and absurdities would follow [III,3,12]. If unity is not just indivisibility but indivisible wugˇu¯d, then perhaps it might subsist by itself, but such a self-subsisting indivisible wugˇu¯d could not be predicated both of substances and of accidents – an understatement, since Avicenna has argued that there can be only one subsisting wugˇu¯d, God – and so if there is such a unity, it is less universal than the unity that is predicated of beings in all categories, and this maximal universal is what we really mean by ‘unity’ [III,3,13 – 15]. Avicenna thus gives an anti-Platonic answer to the aporia on unity but a Platonic answer to the aporia on being.)43 While Averroes’ critique of Avicenna on unity is logically complex, his fundamental point is that unity, in the sense in which it is an attribute of every mawgˇu¯d, is not an accident of that mawgˇu¯d (except in a sense in which it would be mind-dependent), but rather is constituted by the essence of that mawgˇu¯d, just as the wugˇu¯d of a mawgˇu¯d (except its mind-dependent wugˇu¯d-as-truth) is not an accident but is constituted by the essence. There is no essence neutral to existing-in-the-mind and existing-outside-the-mind, or neutral to unity and multiplicity, and there is no cause bestowing wugˇu¯d or unity on such an essence, just the usual Aristotelian causes to bestow form on a preexisting substratum.44 ‘Wa¯hid’, like ‘mawgˇu¯d’, is grammatically paronymous, so there will be a ˙ grammatical appearance that things are wa¯hid through a wahda other than their ˙ ˙ essence, as that they are mawgˇu¯d through a wugˇu¯d other than their essence. But, more importantly, Averroes thinks Avicenna is led astray by the fact that ordinary-language senses of ‘wa¯hid’, like ordinary-language senses of ‘mawgˇu¯d’, ˙

˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

87

43 You might argue, mimicking Avicenna’s argument about unity, that even if there is a subsisting wugˇu¯d, it cannot be the wugˇu¯d that is predicated of beings in all categories, and that this maximal universal is what we really mean by ‘wugˇu¯d’. But, despite what Averroes says, Avicenna seems not to believe that there is a universal wugˇu¯d predicated univocally of substances and accidents, whereas he does believe this about unity. Averroes insists that, since unity is a per se attribute of being, X and Y cannot be univocally one if they are not univocally beings, but Avicenna seems not to accept that inference. 44 Compare Aristotle, Metaphysics H6, arguing that there are no causes of unity to an essence except, if the thing has matter and form, the efficient cause which actualizes the potentiality of the matter; if the thing has no matter, ‘the essence is immediately a one, just as it is immediately a being, and hence there is no other cause to any of these things of being one or of being a being’ (1045b3 – 5).

88 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Stephen Menn

are indeed logically paronymous; Averroes spells out these ordinary-language senses of (individual) unity, except the mathematical sense of one as the principle of number, as the delimitation [inhiya¯z] of a thing by some accident (a ˙ being divided, and he argues that place or a boundary), and negatively as its not a thing is one, in the distinctively metaphysical sense of ‘one’ equivalent to being, through being delimited by its own essence, rather than by an accident which might perish while the thing continues to exist. Fa¯ra¯bı¯ and Kindı¯ on Unity Now Averroes’ analysis of one sense of ‘one’ as ‘delimited by its quiddity’ [munha¯z bi-ma¯hiyyatihi] is taken directly from Fa¯ra¯bı¯’s Kita¯b al-Wa¯hid wa’l˙ § 17, and is as far as I know entirely original to Fa¯ra¯bı¯. Fa¯ra¯bı¯ has ˙ argued wahda, ˙ in the preceding paragraph that something can be called one either because it is delimited [munha¯z] either by its boundary [niha¯ya] or by its place [maka¯n]. The ˙ in these senses seem to be the continuous and the contiguous things that are one respectively, since Fa¯ra¯bı¯ has said that the continuous is called one because its boundary is one (§ 12), and says later that ‘the continuous is not divided in its boundaries, and the contiguous is not divided in the places of its boundaries, for both [boundaries of two contiguous things] are in one primary place’ (in the summary § 95, drawing on Aristotle’s definitions of continuous and contiguous in Physics V,3). ‘The continuous becomes one only because something else in it is one’, namely its boundary (§ 12), and the contiguous also is one only through whatever binds its parts together, ‘and the meaning of its unity is the bond [riba¯t] by which some of them are attached [lazima] to others, whether by ˙ or by art’ (§ 13). So things are one in these senses, not through their nature essence, but through something else present in them (and this something else will have to be one in a stronger sense than they are, for the contiguous to avoid a regress of bonds, for the continuous because its boundary will be of one fewer dimension). By contrast, ‘one’ is also said of what is delimited by its quiddity – whatever quiddity it is, divided or undivided, conceived or outside the soul – and it is what is delimited by … its share [qist] of wugˇu¯d. For the one in this meaning naturally accompanies [yusa¯wiqa] mawg˙ˇu¯d, like ‘thing’, for there is no difference between saying ‘every thing’ and ‘every one [kullu wa¯hidin = “each”]’. And it is said like this of all the ˙ thing, and of other things – if there are any – categories, and of this designated outside the categories. (§ 17)

And if a thing is one inasmuch as it is mawgˇu¯d, and its oneness is its ‘share of wugˇu¯d’, and as we know from the Kita¯b al-Huru¯f the wugˇu¯d of a thing, inasmuch as it is something having a quiddity˙ outside the soul, is just that quiddity, then also the oneness of a thing in this sense will be its quiddity and

Averroes against Avicenna on Being and Unity

not an accident superadded to it. Indeed, the Huru¯f explains the non-mind˙ dependent sense of ‘mawgˇu¯d’ as ‘delimited by some essence outside the soul’ (Huru¯f, p. 116,7) – the same explanation that the Kita¯b al-Wa¯hid gives for the ˙ ˙ relevant sense of ‘one’. This is not the most obvious ordinary-language sense either of ‘mawgˇu¯d’ or of ‘one’, but Fa¯ra¯bı¯ thinks that if we imitate Aristotle’s method in Metaphysics D6 (on unity) and D7 (on being), starting from the ordinary-language meanings and pushing deeper, we will see that the deepest meanings of ‘one’ and ‘mawgˇu¯d’, as applied to mind-independent things, converge. In this sense ‘one’ will not be an accident of ‘mawgˇu¯d’, nor will they both be accidents of ‘thing’, rather the oneness, the wugˇu¯d, and the thingness or essence will all be really identical outside the mind. Fa¯ra¯bı¯ draws precisely this conclusion in the Principles of the Opinions of the People of the Perfect City: ‘[the First’s] wugˇu¯d by which it is delimited [munha¯z] from the mawgˇu¯da¯t other than it cannot be other than that by which it is ˙in itself mawgˇu¯d; and therefore its delimitation [inhiya¯z] from what is other than it is through a unity which is its essence. For one˙ of the meanings of unity is the specific [ha¯ss] wugˇu¯d by which ˘ ˙˙ it, and this is that each mawgˇu¯d is delimited [munha¯z] from what is other than ˙ by which each mawgˇu¯d is called one inasmuch as it is mawgˇu¯d with respect to the wugˇu¯d which specifies it; and this is the meaning of “one” which accompanies [yusa¯wiqu] mawgˇu¯d. So the First is also one is in this way, and more deserving of the name “one” and its meaning than any one other than it’ (68,7 – 13).45 The Perfect City is here taking up the technical terminology, and the definitions of different senses of ‘being’ and ‘one’, from the Kita¯b al-Huru¯f ˙ and Kita¯b al-Wa¯hid, and applying them to the case of the divine first principle, ˙ in very much the way that, on Fa¯ra¯bı¯’s reconstruction, Metaphysics K applies the conceptual distinctions and definitions of Metaphysics D ; and it gives Fa¯ra¯bı¯’s constructive alternative to uncritical theories of the senses in which God is the first mawgˇu¯d and the first one and the cause to all other things of wugˇu¯d and unity.46 ˘

45 I cite the Perfect City (Maba¯di a¯ra¯ ahl al-madı¯na al-fa¯dila) from Fa¯ra¯bı¯, On the Perfect State, ed. Walzer, by page and line numbers. Fa¯ra¯bı¯ says˙ that the First is ‘also one in this way’ because he has just said that the First is also one by being indivisible, and specifically ‘indivisible in its substance’ and therefore ‘one in substance’ (68,6), ‘since one of the meanings in which “one” is said is what is indivisible’ (68,2 – 3); and because he had said previously that it is ‘isolated [munfarid] in its wugˇu¯d’ (i. e. that there is nothing else of its species) and also ‘isolated in its rank’ (meaning esp. that it has no contrary), and that it is therefore one in these ways (66,5 – 7, cf. 62,6 – 7). Both ‘indivisible’ (in various ways) and ‘isolated by some wugˇu¯d’, as well as ‘delimited by some quiddity’, are senses of ‘one’ discussed in the Kita¯b al-Wa¯hid; for ‘isolated by some wugˇu¯d’, and how it differs from ‘delimited by some quiddity’, ˙see below. 46 The technical sense of what Fa¯ra¯bı¯ is saying here in the Perfect City seems never to have been noticed before, and it could not have been noticed until both the Kita¯b al-Huru¯f ˙ the and Kita¯b al-Wa¯hid, with their respective accounts of ‘mawgˇu¯d’ and ‘one’ as having ˙ ˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

89

90 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Stephen Menn

It seems, then, that Averroes in the Epitome of the Metaphysics is taking the Kita¯b al-Wa¯hid as his model for his accounts of unity and multiplicity, ˙ to Aristotle’s Metaphysics D6 and connected passages in Iota, as he corresponding takes the Kita¯b al-Huru¯f as his model for the rest of Epitome I,18 – 66, corresponding to the˙ rest of Metaphysics D, and especially for his account of mawgˇu¯d and wugˇu¯d, corresponding to D7; and it seems that he takes the Kita¯b al-Huru¯f and Kita¯b al-Wa¯hid to be parts of a single Fa¯ra¯bian project of ˙ ˙ distinguishing senses of fundamental metaphysical terms and averting the errors that arise from confusing their different senses or from confusion about their logical syntax. Furthermore, in both cases it seems that Averroes sees Fa¯ra¯bı¯’s main contribution as in averting a metaphysical error of Avicenna, before Avicenna even made it. And this suggests that Averroes sees the Kita¯b al-Wa¯hid, ˙ like the Kita¯b al-Huru¯f, as devoted to exposing and averting the metaphysical ˙ errors of some earlier philosopher or philosophers whose errors were close enough to Avicenna’s that the same arguments would work against both. And, while the Kita¯b al-Wa¯hid keeps a mysterious silence about its motivations and ˙ it might support, it seems very likely that Averroes is the grander conclusions right in this, that the Kita¯b al-Wa¯hid, like the Kita¯b al-Huru¯f, is directed against ˙ ˙ and the Arabic adaptations Kindı¯ or philosophers of his circle of neo-Platonic texts associated with them. And this helps to remove a difficulty about the interpretation of the Kita¯b al-Huru¯f as directed against Kindı¯: why is Fa¯ra¯bı¯ so ˙ other things are mawgˇud through participation concerned to avert the idea that ¯ in God as wugˇu¯d-itself, when this seems to be a relatively minor commitment of Kindı¯’s, while Fa¯ra¯bı¯’s version of Metaphysics D in Part I of the Huru¯f entirely leaves out D6 on unity and multiplicity, although Kindı¯ is˙ much more concerned with God as the ‘true One’ than as wugˇu¯d-itself ? The answer is that Fa¯ra¯bı¯ is not leaving that out but saving it for the Kita¯b al-Wa¯hid, and that the ˙ the motives of programmatic parts of the Kita¯b al-Huru¯f are needed to explain ˙ the Kita¯b al-Wa¯hid. The two texts must be read as part of the same project: ˙ had suggested (in his Arabic introductions to the Kita¯b alperhaps, as Mahdi Huru¯f, pp. 42 – 3, and to the Kita¯b al-Wa¯hid, p. 25) the Kita¯b al-Wa¯hid was ˙ ˙ transmitted in a fragmented ˙ and originally part of the Kita¯b al-Huru¯f (now ˙ disordered state); or perhaps Fa¯ra¯bı¯ originally intended it as part of the Huru¯f, but it grew out of proportion and he spun it off as a separate treatise. ˙ Fa¯ra¯bı¯, and Averroes to whatever extent he may have been aware of the Kindian position, might reasonably regard Kindı¯ as giving the Platonist answer sense of ‘delimited [munha¯z] by some quiddity’, had become available. The picture that ˙ as applying Fa¯ra¯bı¯’s technical ontology to theology contrasts emerges of the Perfect City sharply with Mahdi’s picture (in Foundation of Islamic Political Philosophy) of the Perfect City as an exoteric or ‘political’ work with a non-serious metaphysics, logically disconnected from the esoteric ontology of the Kita¯b al-Huru¯f. I hope to develop this ˙ picture of the Perfect City in other work.

Averroes against Avicenna on Being and Unity 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

91

to Aristotle’s aporia at Metaphysics B4 1001a4-b25. Kindı¯ describes his aim, in the major extant part of the On First Philosophy, as ‘to make clear the distinction among ones, so that the true One may be manifest […] and [so that it may be manifest] what are the ones by license [bi’l-magˇa¯z, i. e. improperly or metaphorically one], I mean [one] by participation [ifa¯da]47 in the true One’ (97,21 – 99,1). So Kindı¯ wants to show that anything other than God is only improperly one, and then that the cause of whatever unity it has is the true One. If a thing can be multiplied or divided in any way, Kindı¯ will conclude that its unity can be separated from it (since it can continue to exist while ceasing to be one), that it has unity only as an accident and not by its essence, that it is one only by convention (both 47,17 – 23), that it is one only by license (95,25 – 6). He will then argue that any such thing has its unity by participating in something that is one by its essence (53,6 – 15 and 95,20 – 25): this must be entirely simple, outside all the categories and all the predicables, and ‘therefore it is just pure unity, I mean nothing other than unity, and every one [kullu wa¯hidin ˙ = each thing] other than it is multiplied’ (95,13 – 14). Furthermore, since ‘there is no being [huwiyya] except inasmuch as there is unity in it’ – because any multiplicity must consist of unities – ‘and its being unified [tawahhud] is its ˙ ˙ from the receiving being [tahawwı¯]’ (97,16), therefore ‘the emanation of unity first true One is the receiving-being [tahawwı¯] of every sensible and of what attaches to sensibles, and so every one of them exists [yu¯gˇadu] when it receives being through [the true One’s] giving it being’ (97,8 – 10). By contrast, Fa¯ra¯bı¯, and Averroes following him, will determine senses in which each thing, even if it is divisible, can be called ‘one’. These senses are not by license, and they are not all through an accident, since ‘delimited by its essence’ applies to every mawgˇu¯d. Something may still be one ‘through another’, i. e. through some cause other than itself, just as it may exist through some cause other than itself, and we may be able to trace it back causally, step by step, to a first One, just as we may trace it back causally to a first mawgˇu¯d. But the thing will not be one by participating in this first One, as if it were a formal cause of unity, rather the thing will be one formally through its own essence, and dependent on something else as its subject or its efficient cause; and the first One will not be a universal one beyond the categories and participated in by things in all the categories, rather it will be reached by inferring causally from accidents to substances and then from substances to their first efficient cause.48 47 The last phrase might be more literally translated ‘by the benefaction [ifa¯da] of the true One’, i. e. by the true One’s bestowing something (namely, unity) on it. But it seems to be fairly fixed terminology (for instance, in the Liber de causis, rendering Proclus’ language of participation) that the mustafı¯d is the participant, the mustafa¯d is what the participant receives, and the mufı¯d is what it participates in. 48 However, Fa¯ra¯bı¯ thinks we can infer causally back from substance to a first cause which is not in any of the categories and thus in particular is not a substance (see Kita¯b al-Huru¯f ˙

92

Avicenna of course does not think that things other than God are ‘one’ only metaphorically, nor does he think that they are one by participating in God as a formal cause of unity, any more than he things that they are mawgˇu¯d by participating in God as a formal cause of wugˇu¯d; when he abstracts away from a thing’s unity he discovers, not a plurality or an original non-being, but an essence neutral to individual existence in matter and universal existence in a mind, which receives numerical or specific unity by coming to be in matter or being thought by a mind. But he too thinks that unity is an accident superadded to the essence of things, and Averroes might reasonably think that the distinctions in unity that Fa¯ra¯bı¯ develops in order to avert Kindı¯’s errors will serve to avert Avicenna’s errors as well. Whence the Thesis that Quantitative Unity is Mind-Dependent? The only main point of Averroes’ account of unity that does not seem to be in Fa¯ra¯bı¯’s Kita¯b al-Wa¯hid is the thesis that the one as a principle of number exists ˙ act of abstraction, so that unity in this sense, like being only through the mind’s as truth, is a second intention. Presumably Averroes was led to this by reflecting on Avicenna’s thesis that number or multiplicity is an accident in the category of quantity composed out of many unities, each of which is an accident in the category of quantity belonging to an underlying subject in some category. Avicenna, responding to an opponent who says ‘multiplicity is sometimes composed out of things other than unities, such as men and beasts’, replies that ‘as these things are not unities, but things which are subjects of unities, likewise also they are not [collectively] a multiplicity, but things which are subjects of multiplicity; and as these things are ones [but] not unities, likewise they are [collectively] many [but] not a multiplicity’ (Sˇifa¯ Metaphysics III,3,6 – 7). Averroes will distinguish multiplicity as a per se attribute of being, which is not distinct from the essences of its subjects, from the accident numerical multiplicity, but he will agree that if Socrates and Plato are numerically two, their twoness is composed not out of Socrates and Plato but out of the unity of Socrates and the unity of Plato, which must be in the same category, quantity, as the twoness they compose; and this will lead Averroes to reflect on the status of this quantitative unity, which cannot be the same as the unity which is a per se attribute of being. But it may be that here too he found a clue in Fa¯ra¯bı¯. ˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Stephen Menn

I,92; translated and discussed in Menn, Analysis of the Senses of Being, pp. 93 – 6), while Averroes thinks that God is a substance, and has a determinate essence, namely intellect. (Averroes is quite unusual within the Muslim context in saying that God is a substance: credal statements typically say that God is neither a body [gˇism] nor a substance [gˇawhar], by which they mean in the first instance that he is neither a compound of atoms nor himself an atom.)

Averroes against Avicenna on Being and Unity

Recall that Averroes says ‘the numerical one is the meaning of the individual, abstracted from quantity and quality, I mean that by which the individual is an individual, since it is an individual only by the meaning “undivided”’ (Epitome III,41, cited above), and that ‘when the intellect abstracts from these individuals this meaning, “indivisible into two or more individuals”, this is the one which is the principle of number’ (I,36, cited above). This notion of individuality or undividedness seems to be what Fa¯ra¯bı¯ means by the one as ‘that whose essence is not shared [in such a way that] some two things would resemble each other in respect of it’ (Kita¯b al-Wa¯hid § 18) or ‘what has no ˙ partner [qası¯m] in the meaning [= predicate] which is attributed to it – whatever meaning it is – in [the sense] that this is a quiddity which belongs to it, as if it were isolated [munfarid] by wugˇu¯d, and nothing other than it shares with it the quiddity which belongs to it’ (§ 25); this sense of unity ‘accompanies “delimited by some quiddity”, for it is delimited only by that in which it has no partner’ (§ 26). This sense applies to immaterial substances, which are necessarily unique in their species, but also to every designated thing [musˇa¯r ilayhi = t|de ti] which is not in a subject [i.e. individual substances], and every designated thing which is in a subject [i.e. individual accidents as described in Categories 1a23 – 9], since none of these is predicated of more than one thing. And ‘one’ is also said of what is not divided into a subject lower than it [sc. as the individual is ‘lower’ than the species and the species than the genus] but rather it is the lowest subject which underlies a predicate, and the division of each more general predicate terminates with it and does not go beyond it; and many people are wont to call it ‘one in number’, and it is called49 individuals [a ya¯n] and individuals [asˇha¯s]. (§ 18) ˘ ˙ ˘

It seems that this covers both what is predicated of only one thing (or is predicated of nothing, except tautologously of itself ), and what has some predicate that is predicated only of it; and it seems that the more basic notion is ‘predicated of only one thing’, while ‘has some predicate that is predicated of only one thing’ is derived from this. But ‘predicated of only one thing’ is a second intention, since ‘predicated of ’ is a second intention, applicable only to things in the mind (Kita¯b al-Huru¯f, p. 64,11 – 14). To say that something is one ˙ is only one of it, and it is properly only for a in this sense is to say that there predicate-term F that I can say that there is only one F: if I say of something c which is not a predicate that it is unique, that there is only one of it, this is just a loose way of saying that for some appropriate predicate F (perhaps c’s species, if c is an immaterial substance), c is F and there is only one F.50

49 Perhaps we should add a ha¯ at the end of the verb, reading yusammı¯hi instead of yusamma¯ ; then not ‘it is called’ but ‘[Aristotle] calls it’. 50 So something can be ‘one’ in the sense of ‘unique’ in respect of one predicate and not in respect of another: ‘the moon is isolated by the meaning of moonness [since] it has no ˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

93

94

It seems likely that Averroes is reflecting on these passages of Fa¯ra¯bı¯, and finding in them a response to what Avicenna says about the unities of which accidents of number are composed; and thus the concept of individual unity which Averroes takes to be the principle of number, i. e. the concept we use in counting, would be this logical indivisibility rather than quantitative indivisibility into spatially distinguished parts. I can legitimately count to three and say ‘there are three people is the room’ (logically similar to Fa¯ra¯bı¯’s example, ‘there is one dirham in my hand’, Kita¯b al-Wa¯hid § 25), despite the fact that each of the ˙ and lower parts, because the predicate people in the room is divisible into upper ‘is a person in this room’ is truly predicated of three subjects which are not truly predicated of any further subjects.51 Just as ‘existing outside the mind’, which Avicenna takes to be an accident attaching in re to a neutral essence, is only a second intention, saying of some predicate F ‘there is at least one F’, so numerical unity, which Avicenna likewise takes to be an accident attaching in re to a neutral essence, is only a second intention, saying of some predicate F ‘there is exactly one F’; and presumably threeness is also a second intention, saying of some predicate F ‘there are exactly three F’s’. And just as Averroes’ criticisms of Avicenna on mawgˇu¯d and wugˇu¯d are the starting point for Latin scholastic discussions of essence and existence, so his criticisms of Avicenna on unity and number will be the starting point for the scholastic discussions of numerical and transcendental unity, and of categorematic and syncategorematic quantity. But even without pursuing the Latin discussions, the Arabic history makes it clear that the logical syntax of unity, like the logical syntax of being, has a long and complicated history; in particular, the thesis that both unity and being have important senses in which they are second-order predicates has a history going back far beyond the nineteenth-century logicians who are usually said to have invented it.52 partner of its species, although it has a partner in the meaning that it is a star, which is its genus’ (Fa¯ra¯bı¯, Kita¯b al-Wa¯hid wa’l-wahda § 25). ˙ restricted ˙ to things in this room and to the present time, 51 Or, perhaps, ‘is a person’, when is truly predicated of three such subjects; or ‘is in the room’, when restricted to persons and to the present time. Fa¯ra¯bı¯ gives ‘there is one dirham in my hand’ as an example of being isolated by some predicate ‘relatively and for a time’, following Aristotle’s discussion in Topics V,1 of predicates which are propria of some subject – i. e. predicates in respect of which that subject is unique – ‘relatively and for a time’. 52 Since the Epitome of the Metaphysics was not available in Latin until the Renaissance, it will be important for the Latin reception that much of the Epitome’s argument about unity can be found, if not as clearly developed, in the Great Commentary on Metaphysics Iota 1 – 2 (some parallels have been noted above). But the Great Commentary, sticking closer to Aristotle’s text, seems not to describe unity as ‘delimitation’ [inhiya¯z], or to say that mathematical unity arises from an act of the soul. In addition to the˙ parallels noted above, note that at Tafsı¯r 1282,7 – 12, on Iota 4 1052a13 – 19, Averroes says that Avicenna wrongly thought that if ‘one’ and ‘mawgˇu¯d’ and ‘thing’ signified the same ma na¯ then e. g. ‘the man is one’ would be a tautology or negation like ‘the man is [a] man’. ˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Stephen Menn

95

Averroes against Avicenna on Being and Unity

Bibliography

˘

Averroes, Compendio de Metafisica, ed. C. Quirs Rodrguez, Madrid: E. Maestre, 1919. –– , Commentarium magnum in Aristotelis de anima libros, ed. F.S. Crawford, Cambridge, Massachusetts: The Mediaeval Academy of America, 1953. –– , On Aristotle’s ‘Metaphysics’: An Annotated Translation of the So-called ‘Epitome’, ed. R. Arnzen, Berlin: De Gruyter, 2010. –– , Tafsı¯r ma¯ ba d al-tabı¯ at, ed. M. Bouyges, 4 vols, Beirut: Imprimerie Catholique, ˙ 1938 – 48 [‘Tafsı¯r’]. –– , Tahafot at-Tahafot, ed. M. Bouyges, Beirut: Imprimerie Catholique, 1930 [‘TT’]. Avicenna, Kita¯b al-Nagˇa¯t fı¯ ’l-hikmat al-mantiqiyya wa’l-tabı¯ iyya wa’l-ila¯hiyya, ed. ˙ al-Jadidah, ˙1985. ˙ M. Fakhry, Beirut: Dar al-Afaq –– , The Metaphysics of ‘The Healing’, transl. and ed. M. Marmura, Provo, Utah: Brigham Young University Press, 2005. A. Bertolacci, On the Arabic Translations of Aristotle’s Metaphysics, Arabic Sciences and Philosophy, 15, 2005, pp. 241 – 75. H.A. Davidson, Alfarabi, Avicenna and Averroes on Intellect, Oxford: Oxford University Press, 1992. ˘

˘

˘

Compare also the Great Commentary on Metaphysics C2: ‘and Avicenna erred greatly in this: he thought that “one” and “mawgˇu¯d” signified attributes superadded to the essence of the thing. The astonishing thing about this man is how he committed this error, given that he listened to the Ash arite mutakallimu¯n who mix his metaphysics with their kala¯m, and they say that some attributes [hold] through an [inhering] entity [are ma nawı¯ attributes] and others [hold through] the thing itself [are nafsı¯ attributes], and they say that “one” and “mawgˇu¯d” refer to the essence of the thing to which they are attributed, and they are not attributes signifying something superadded to the essence, as is the case with “white” and “black” and “knowing” and “living”. But this man was compelled to his opinion by his saying that if “one” and “mawgˇu¯d” signified the same ma na¯, then our saying “the mawgˇu¯d is one” would be nugation, like our saying “the mawgˇu¯d is mawgˇu¯d” or “the one is one”. But this would follow only if it were said that our saying about a single thing that it is mawgˇu¯d and that it is one signify a single thing in a single respect and in a single manner; but we say only that they signify a single essence in different manners, not different attributes superadded to it. Thus according to this man there is no difference between significations which signify about a single essence in different manners without signifying ma a¯nı¯ superadded to it, and significations which signify about a single essence attributes superadded to it, i. e. different from it in actuality. And this man was led astray by several things: one was that he found that the noun “one” [wa¯hid] was a paronymous noun and these nouns signify an accident and a substance [i.e.˙they signify some accident present in some substance, as “white” signifies whiteness in a substantial substratum]; another was that he thought that the noun “one” signifies an indivisible ma na¯ in the thing, and that this ma na¯ is not the ma na¯ which is [the thing’s] nature [since, if it’s a material thing, this will be divisible]; another was that he thought that this one which is predicated of all categories was the one which is the principle of number, and number is an accident, and so he concluded that the noun “one” signifies about the mawgˇu¯da¯t an accident. The one which is the principle of numbers is only [one] among the beings of which the noun “one” is said, even if it is the most deserving of this [name], as you will learn in the ninth book of this treatise [= Iota, not Theta]’ (313,6 – 314,11). ˘

˘

˘

˘

˘

˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

96

˘

˘

–– , Proofs for Eternity, Creation and the Existence of God in Medieval Islamic and Jewish Philosophy, Oxford: Oxford University Press, 1987. Master Eckhart, Parisian Questions and Prologues, transl. A. Maurer, Toronto: Pontifical Institute of Mediaeval Studies, 1974. Abu¯ Nasr al-Fa¯ra¯bı¯, Commentary and Short Treatise on Aristotle’s ‘De Interpretatione’, ed. F.W.˙ Zimmermann, Oxford: Oxford University Press, 1981. –– , Kita¯b al-Huru¯f = Alfarabi’s Book of Letters, ed. M. Mahdi, Beirut: Dar El-Mashreq, 1969 – 70.˙ –– , Kita¯b al-Wa¯hid wa’l-wahda = Alfarabi’s On One and Unity, ed. M. Mahdi, ˙ ˙ Casablanca: Editions Toubkal, 1989. –– , On the Perfect State (Maba¯di a¯ra¯ ahl al-madı¯na al-fa¯dila), transl. and ed. R. Walzer, ˙ Oxford: Clarendon Press, 1985. al-Kindı¯, Oeuvres Philosophiques et Scientifiques, vol. 2, Mtaphysique et Cosmologie, eds. R. Rashed and J. Jolivet, Leiden: Brill, 1998. M. Mahdi, Alfarabi and the Foundation of Islamic Political Philosophy, Chicago: University of Chicago Press, 2001. S. Menn, Fa¯ra¯bı¯’s Kita¯b al-Huru¯f and his Analysis of the Senses of Being, Arabic Sciences ˙ pp. 59 – 97. and Philosophy, 18, 2008, –– , Review of Anthony Kenny, Aquinas on Being, Philosophical Review, 115, 2006, pp. 391 – 5. ˙ anı¯ al-Na¯bulusı¯, ¯Ida¯h al-maqsu¯d min wahdat al-wugˇu¯d, Damascus: Matba a Abd al-G ˙ ˙ ˙ ˙ ˙ al- Alam, 1969. J. Puig Montada, Cuanto se encuentra ms all de la naturaleza, introduction to the reprint edition of Averroes, Compendio de metafsica, ed. C. Quirs, Sevilla: Fundacin El Monte, 1998, pp. IX-XXXIV. A.I. Sabra, Avicenna on the Subject Matter of Logic, Journal of Philosophy, 77, 1980, pp. 746 – 64. R. Wisnovsky, Avicenna’s Metaphysics in Context, London: Duckworth, 2003. –– , Notes on Avicenna’s Concept of Thingness (sˇay iyya), Arabic Sciences and Philosophy, 10, 2000, pp. 181 – 221. ˘

˘

˘

˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Stephen Menn

Avicenna and his Commentators on Human and Divine Self-Intellection* Peter Adamson It is a notorious feature of Avicenna’s metaphysics that he begins from the notion of existence, rather than progressing smoothly from natural philosophy to divine science. Avicenna was duly criticised by Averroes for this departure from the Aristotelian tradition.1 And certainly it is true that Avicenna proves the existence of God without premises drawn from natural philosophy, in his famous demonstration of the Necessary Existent. This strategy in turn bears on Avicenna’s view of the subject of metaphysics, since it is well-established Aristotelian doctrine that no science proves the existence of its own subjectmatter. These familiar points are in danger of obscuring a noteworthy aspect of Avicenna’s metaphysics which is rather in tension with the idea that metaphysics is autonomous from physics. When Avicenna is concerned to say not merely that God exists, but what He is like, he does in fact make use of premises established in natural philosophy. In particular, his discussions of God in both the Healing (al-Sˇifa¯ ) and Pointers and Reminders (al-Isˇa¯ra¯t wa-l-tanbı¯ha¯t) depend on points established in the psychological sections of those works (psychology, of course, being a part of natural philosophy both in the previous tradition and in the rubrics of the various sections of the Healing itself ). The dependence of Avicenna’s metaphysical theology on his psychology did not escape the notice of his two most famous commentators, Fahr al-Dı¯n alRa¯zı¯ (d. 606/1210) and Nas¯ır al-Dı¯n al-Tu¯sı¯ (d. 672/1274). Both˘of them say ˙ tanbıh in the Pointers,2 which reads more or less the same thing˙about a crucial ¯ as follows: ˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

*

1 2

This paper has benefited greatly (though, no doubt, not enough) from the input of numerous colleagues. My thanks to audiences at the Villa Vigoni and in Freiburg in July 2008, and to those who kindly read and commented on earlier versions: Amos Bertolacci, Dag Nikolaus Hasse, Heidrun Eichner, MM McCabe, and David C. Reisman. See Bertolacci, Avicenna and Averroes. Avicenna, al-Isˇa¯ra¯t, cited as Pointers followed by section number, with volume and page number from Dunya¯’s edition in brackets. All citations are to part two of the Pointers, so I will not bother to indicate this. There is no English translation of the relevant parts of the Pointers, but for a French translation see Avicenna, Livre des directives et remarques.

98

˘

˘

˘

Pointers § 4.28 [III.53]: The First is self-intelligible (ma qu¯l al-da¯t) of its ¯ materials subsistence.3 For it subsists free from attachments, lacks ( uhad), (mawa¯dd), or anything else that would bestow anything additional that would inhere (ha¯ll za¯ ida) in the essence. And you know that whatever is like this intellects ˙ is intellected by itself. itself and

Now, neither of these two commentators is notable for his concision in dealing with the text of the Pointers; al-Ra¯zı¯’s commentary is especially voluminous.4 So it is all the more striking how little either of the two commentators has to say about the passage just cited. They simply refer us back to the third section or namat of part two of the Pointers: this is the namat which deals with ˙ ˙ 5 psychology. At this critical juncture, where we are establishing the intellective nature of God, the commentators follow Avicenna in allowing the argument to rest on claims proven in the study of psychology. The psychological premise in question is made quite explicit here: anything immaterial must be an intellect, and moreover an intellect that intellects itself. Given the immateriality of God, it follows immediately that God is a selfintellecting intellect. Although the idea of immaterial, self-thinking intellect is a familiar theme from the Neoplatonic tradition, the origins of the crucial premise are already to be found in Aristotle. Of course Metaphysics book Lambda is the source for Avicenna’s idea that God is a self-thinking intellect. But the inference from immateriality to intellection and self-intellection is drawn from De anima III.4. There, Aristotle not only establishes a link between intellection and immateriality, but also discusses the fact that an intellect can be the object of its own thought. So in what follows I will first look briefly at this Aristotelian text. But I will spend most of the paper discussing how Aristotle’s ideas were used and expanded upon by Avicenna and his commentators. In the first section I will discuss the connection between intellection and immateriality, in the second section, the idea that every self-subsisting intelligible is an intellect that can engage in intellection of itself. In the third section, I deal with the claim that God is just such a self-thinking intellect.

The Dunya¯ edition has al-Awwal ma qu¯l al-da¯t qa¯ imiha¯, which Goichon translates ‘Le ¯ Premier a une essence intelligible, il la fait subsister.’ One might instead read qa¯ im bi-ha¯ ; al-Tu¯sı¯ has something similar, since his commentary says instead qa¯ im bi-nafsihi. In that case˙ we could translate more simply ‘the First is self-intellecting and self-subsistent’. My thanks to David Reisman for discussion of this passage. For al-Ra¯zı¯’s commentary I have used the edition in Sˇarhay al-Isˇa¯ra¯t. For al-Tu¯sı¯’s ˙ cited respectively ˙as alcommentary see Dunya¯’s edition in Avicenna, al-Isˇa¯ra¯t. Hereafter Ra¯zı¯, Commentary and al-Tu¯sı¯, Commentary. al-Ra¯zı¯, Commentary I.214˙ lines1 – 2; al-Tu¯sı¯, Commentary III.54 line 10. ˙ ˘

˘

˘

3

˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Peter Adamson

4 5

Avicenna and his Commentators on Human and Divine Self-Intellection

Immateriality and Intellection Turning then to De anima III.4,6 we find Aristotle making the observation, often repeated in the Arabic tradition, that what is extremely intelligible does not incapacitate intellect, the way a bright light or strong scent incapacitates the faculty of sight or smell.7 The reason given for this is that ‘sensation is not without body, whereas intellection is separate from body’ (429b5, p. 73 lines 8 – 9 in the Arabic version). In the next sentence, he explains that a man who has learned an intelligible has that intelligible potentially, but in a higher potentiality than the one prior to learning it, ‘and at that point it can intellectually grasp itself ’ (429b9, p. 73 line 12 Arabic).8 The relationship between self-thinking and immateriality is not yet clear, but it is at least suggestive that the two points follow in such close succession. And later in the same chapter Aristotle does make an explicit link between thinking, selfthinking, and immateriality. This is occasioned by the question whether ‘thought is also intelligible’ (429b26, p. 74 line 7 Arabic). His answer is as follows, translating from the Arabic version: ˘

˘

De anima III.4, 430a2 – 9 (p. 75 lines 13 – 18 Arabic): It [sc. intellect] is also intelligible,9 just like the other intelligible things. And regarding those things that are immaterial, the intellecting and the intelligible (al- a¯qil wa-l-ma qu¯l) are one and the same thing … But I claim that the intelligible in what is material is only potentially intelligible. Therefore material things have no intellect, because intellect in respect of potentiality is not in matter (al- aql min g˘ihat al-quwwa laysa fı¯ hayu¯la¯). But the intelligible belongs to the intellect, being ascribed (mansu¯b) to it. ˘

This passage sets up a clear correlation, if not mutual implication, between immateriality, thinking, and self-thinking. Because it is immaterial, the intellect 6 7

˘

9

˘

8

I have been greatly helped in understanding this chapter by McCabe, Some Conversations. De anima III.4, 429a31 – 429b4, and for the Arabic version, Aristu¯ta¯lis fı¯ l-nafs, ed. ˙ ˙the attribution is Badawı¯, p. 73. This translation is ascribed to Isha¯q b. Hunayn but ˙ ˙ controversial. See Elamrani-Jamal, De anima. The Arabic has ‘wa-yumkinuhu fı¯ da¯lika l-waqt an ya qila nafsahu’ and thus clearly accepts the reading ja· aqt¹r d³ art¹m¯ t|te d}matai moe?m in the Greek manuscripts. Ross follows Bywater in emending d³ art¹m to d¸ artoO, which would yield ‘and at that point it can think through itself.’ A word here about the term ma qu¯l, here translated ‘intelligible’: like other Arabic words of this form it is ambiguous between factive and potential readings. In other words a ma qu¯l might be either that which is in fact being intellected or that which is not being intellected, but could be intellected. I translate the term as ‘intelligible’ throughout, but more often than not Avicenna will mean by it something that is indeed occurently an object of intellection, and the reader should bear this in mind. (For instance, in the passage already translated above, Pointers § 4.28, Avicenna of course means that the First actually intellects itself, not merely that it can do so.) My thanks to Gad Freudenthal for discussion of this point. ˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

99

100

can both think and be thought. Aristotle makes it especially clear that it is materiality that prevents thought (and self-thinking), because he says that some intelligibles do not think precisely because they are in matter. What he has in mind here is that a form in a stone, for instance, may be intelligible – in Avicenna’s terms, the stone could provide me with a basis for abstracting the intelligible form granite – but the stone does not itself think, or think about itself, because it is a material object. Avicenna raises a hypothetical objection in the Pointers that builds on this puzzle: Pointers § 3.20 [II.422]: Perhaps you will say that when the form that subsists materially (al-su¯ra al-ma¯ddiyya fı¯ l-qawa¯m) is abstracted in the intellect, then the ˙ prevents [it from itself engaging in intellection] is removed. So characteristic that what stops us from ascribing intellection to it?

As we just saw, in De anima III.4 Aristotle was worried about two kinds of intelligibles: the intelligible object outside the intellect, and the intellect itself. The former does not think, because it has matter,10 the latter does think, because it has no matter – and for the same reason it is an object of self-thinking. The objection then is that there is a third kind of intelligible: the abstracted intelligible that is in the intellect. If it is no longer enmattered, why will it not think? For example, if I abstract the intelligible form granite from a stone, then should the form granite in my intellect not be able to engage in intellection in its own right? Avicenna responds as follows: ˘

Pointers § 3.20 [II.422 – 4] The answer is: because it is not independently subsisting (mustaqilla bi-qawa¯miha¯), nor receptive of intelligible concepts (ma a¯nı¯) that inhere in it. The representations (amta¯l) of [the forms that subsist materially] are simply ¯ conjoined (yuqa¯rinuha¯) with intelligible concepts, in which they are not inscribed – rather [they are inscribed] in that which receives both of them. Neither is more fitting to be inscribed in the other, instead of the other [being inscribed] in it. The two are not put together the way that form and what has form are conjoined.

Here Avicenna makes a concession to the opponent: suppose that I grasp the form granite and also the form iron. 11 In that case the form granite will be ‘conjoined’ with the form iron, because they will both be objects of the same intellect. That is, they will in some sense be compresent with one another, since they are together in the same subject (so in this context the word ‘conjoin’ just 10 If the external object is itself immaterial, it too will be an intellect. We will see Avicenna’s proof for this in the next section of this paper. 11 For the sake of simplicity, I speak rather loosely of ‘forms’ even though Avicenna is using more varied terminology: he speaks of intelligible ‘concepts’ and also of ‘representations’. The former seems, in this passage, to refer specifically to the form when it is in the intellect, hence my translation of the notoriously difficult ma a¯nı¯. Meanwhile the use of amta¯l (‘representations’) seems to refer to the form in the mind insofar as it is related to the¯ external form in matter. ˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Peter Adamson

Avicenna and his Commentators on Human and Divine Self-Intellection 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

101

indicates that the two things are somehow together). But Avicenna insists that the form granite will not itself grasp the form iron, because it is not a selfsubsisting thing. Their mere compresence is not enough to give either of them an intellectual grasp of the other. This is a question-begging response, insofar as Avicenna is merely stipulating that self-subsistence is a necessary condition for engaging in intellection. But he does add a further argument, which is that granite and iron would have an equal claim to be the subject of intellection. There is no principled reason why granite should think about iron instead of the other way around. But they cannot grasp each other, because intellectual grasping is assumed to be asymmetrical. The sort of asymmetry involved here, as Avicenna says, is the asymmetry between ‘form and what has form’. Al-Ra¯zı¯’s criticism of this tanbı¯h 12 focuses on this aspect of Avicenna’s argument. He does not question the assumption that the intellection relation must be asymmetrical. But he then points out that there could be such an asymmetry, not only between intellect and intelligible, but also between two abstracted forms in my intellect. If the two forms are distinct, then could one form be a fit subject (mahall) for the other, but not vice-versa? For example, it is possible for motion to ˙be slow, but not for slowness to move. So it seems the form slowness in my intellect could be asymmetrically ‘conjoined’ with the form motion in my intellect too. In which case the form motion could intellectually grasp the form slowness. The objection illustrates a feature of Avicenna’s account that al-Ra¯zı¯ frequently exploits in his criticisms, which is that the ‘conjunction (muqa¯rana)’ between intellect and intelligible is an instance of the relation between mahall ˙ and ha¯ll. This pair of terms is difficult to translate with two forms of the same ˙ English word, but the basic idea is that the ha¯ll is something that ‘inheres’ or is present in the mahall. The primary example ˙of the inherence of ha¯ll in mahall is ˙ of form inhering in matter. But the terminology ˙ can be˙ used the ontological case to cover other cases; for instance, as we just saw, al-Ra¯zı¯ speaks of motion as the mahall of slowness. As we have just seen, the inherence relation is always ˙ asymmetrical. This explains why both Avicenna and al-Ra¯zı¯ are happy to assume that two things cannot inhere in one another, as then they would relate symmetrically.13 On the other hand, it is not yet clear what the intellectintelligible relation amounts to, beyond its being asymmetrical. This is the point exploited by al-Ra¯zı¯: insofar as an intelligible that I grasp can be a subject of 12 Al-Ra¯zı¯’s discussion runs from I.174 line 4 to the end of the page. 13 Thus, as al-Ra¯zı¯ makes explicit in his commentary on this tanbı¯h, the only way an intelligible quiddity grasped by my intellect can be the subject of intellection is if that quiddity also exists outside of me as a self-subsisting intellect in its own right. If I grasp the quiddity of a celestial intellect, the quiddity of that intellect qua object of my intellect would not be able to engage in intellection, even though the same quiddity engages in intellection outside of me.

102

inherence for some ha¯ll or other (e. g. motion is the subject of slowness when I predicate the latter of˙ the former), it is hard to see why it could not be a subject of intellection too. What we need is a principled distinction between intellection and other sorts of inherence. We have already seen Avicenna’s answer to this demand, but not his entire rationale for that answer. The answer is that intellection occurs when an intelligible object inheres in a subject that is self-subsisting and immaterial. We can gain some insight into why the subject of intellection must be selfsubsistent, if we consider the following. Suppose that two items, A and B, are conjoined such that A is thinking about B. In this case B subsists in A, by being its object of intellection. As we have seen, this relation is asymmetrical, so we can rule out that A conversely subsists in B. It is the mahall, not the ha¯ll. But ˙ subject could A not subsist in some further subject, C? This would˙ happen if the of intellection (item A) depended on some material substrate (item C). But that is ruled out by the second part of Avicenna’s answer, namely his insistence that the subject of intellection is immaterial. We can easily point to Avicenna’s source for the claim. He is following Aristotle’s assertion in De anima III.4 that it is matter which prevents a form from engaging in intellection, of both other things and itself. (Notice the implication that self-subsistence is necessary, but not sufficient, for being the subject of intellection. After all, physical objects are self-subsistent, and so some things can inhere in them: for instance colour might inhere in a piece of paper. But these things are material, and so cannot be subjects for the special sort of inherence that is intellection.) Of course we will want an argument to show that materiality precludes intellectivity. For such an argument we can look earlier in the same chapter of the De anima (III.4, 429a18 – 27), where Aristotle reasons that because intellect potentially thinks every form, it cannot actually have any given form before it thinks. Hence, it cannot be ‘mixed’ with body, since every body actually has determinate form. Whatever we make of this argument, Avicenna seems to have thought it could be improved upon. He does not make use of it in either the Healing or the Pointers. Instead, he argues in both texts that the intelligible form is ‘undivided’, and thus cannot be received in body, which is ‘divided’ in the sense of actually having discrete parts. (If this is inspired by Aristotle, then the source text is not De anima III.4, but rather III.6, where Aristotle discusses the question of how indivisible objects are grasped by mind.) Here is the version of the argument in the Pointers: Pointers § 3.16 [II.404 – 8]: If you want now to have it shown to you that the intelligible item (ma na¯) is not inscribed in something divided (munqasim)14 or in something that has spatial position, then listen. You know that something undivided may be conjoined with a multiplicity of things without thereby having to become ˘

˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Peter Adamson

14 Or ‘divisible’; cf. my remarks on ma qu¯l, n. 9 above.

Avicenna and his Commentators on Human and Divine Self-Intellection

˘

spatially divided (munqasima fı¯ l-wad ), so long as their multiplicity is one that is ˙ a multicoloured object.15 But something that not spatially divided, as are the parts of is divided according to spatial differentiation cannot be conjoined with something undivided. Inevitably there are undivided concepts in the intelligibles – unless the intelligibles are put together from actually infinite principles. But even then, in every multiplicity, whether finite or infinite, there must be some actual unity. So in the intelligibles there is something actually one, which is grasped intellectually insofar as it is one. Thus it will be grasped intellectually precisely insofar as it is undivided. Therefore it will not be inscribed in what is spatially divided. But every body, and every power (quwwa) in a body, is divided.

In keeping with the design of the Pointers, this passage is rather compressed, but the thrust of the argument is clear. An intelligible form cannot be grasped by dividing it up and sharing it out part-by-part in the parts of a subject. Rather, the subject of intellection as a whole needs to grasp the form as a whole. Al-Ra¯zı¯’s commentary (I.163 – 4) begins by supplementing Avicenna’s argument with a further proof (I.163 lines 8 – 27), which is reminiscent of the Sail passage from the first part of Plato’s Parmenides. 16 The argument goes exhaustively through the different ways an undivided thing can be in a divided thing. It might be wholly present in each part. Or different parts of it might be in different parts of the subject. But in the first case it will appear in the subject an infinite number of times, since the subject is infinitely divisible. Furthermore, what we want is for it to be received by the whole subject, not just each part of the subject. In the second case it will be received as divided, not as undivided – but the goal was to explore how an undivided subject is received. Now al-Ra¯zı¯ shifts to a critical stance. He says that Avicenna’s claim seems to be disproven by the undivided point, which is present in a divided body, and goes on to discuss this possibility at length (I.163 line 27 to end of page). It is interesting to note that the example of the point (for which see also De anima III.6, 430b20), and also the idea of an intelligible being present in parts of a divided subject, can be found in the Psychology section from Avicenna’s Healing. In a chapter devoted to proving that ‘the rational soul does not subsist by being inscribed (muntabi ) upon bodily material’ (V.2),17 Avicenna has a long ˙ ˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

103

15 For bulqa as a ‘multicoloured object’ see Lane, Lexicon, vol. I, p. 253. The term is used for instance of a black-and-white horse. 16 Parmenides 131c-e. If there is a historical connection to the Parmenides it would most likely be via the commentary tradition on Aristotle. As early as al-Kindı¯, the Arabic tradition takes up this question of how a form or species is present – by being divided into parts, or as a whole – in its instances. The passage in al-Kindı¯ is in a text related not to any Platonic work, but the Categories. See Adamson and Pormann, Aristotle’s Categories and the Soul. 17 De anima, ed. Rahman, pp. 209 ff. French translation in Psychologie, ed. Bakosˇ, pp. 148 ff. Cf. the translation at McGinnis and Reisman, Classical Arabic Philosophy, pp. 188 – 92.

104

discussion of the point as the only indivisible thing that inheres in a body, where he argues that the relation between point and magnitude is not the same as the relation between intelligible and intellect.18 He then goes on to consider whether the intelligible could be present in parts of a body, and introduces a dichotomy like that of al-Ra¯zı¯: the intelligible might be present by having either similar or dissimilar parts present in the different parts of the body.19 So it seems likely that al-Ra¯zı¯ consulted the Healing when writing this part of his commentary on the Pointers. However, al-Ra¯zı¯ is less sanguine than Avicenna about the prospects of showing a disanalogy between the point and the intelligible. He also mentions other undivided things that inhere in bodies. For instance unity, relation and existence all inhere in bodies, but it makes no sense to divide something like existence in half (Commentary, I.163 – 4). If all these things inhere in bodies, what rules out that the intellect be a body, yet still have an intelligible object inhering in it? Since al-Ra¯zı¯’s arguments have again turned on analogies between intellection and other sorts of inherence, it is unsurprising that al-Tu¯sı¯’s more ˙ sympathetic commentary strives to distinguish intellection from these other sorts. Again, any such relation is an asymmetrical one between a ha¯ll and a ˙ mahall. Both the subject of inherence and that which inheres can be ‘divided’ in ˙ a number of ways, and many of these ways are consistent with incorporeality. For example black might be divided into genus and specific difference; but this division on the side of the ha¯ll implies no division on the part of the mahall. 20 ˙ fact that body has a genus and a species˙ (e. g. The reverse is also true: the extension and three-dimensional) shows nothing about whether there is some division of what inheres in body. Al-Tu¯sı¯ claims that it is only spatial division of ˙ a mahall which necessitates the division of the ha¯ll, and vice-versa. ˙ ˙ In fact, it is not even the case that everything that inheres in a spatially divided subject will itself be divided. In such a case ‘what inheres in [a subject] does not do so qua this subject (min hayt huwa da¯lika l-mahall), but insofar as a ¯ ¯ different nature belongs to it (min h˙ayt luhu¯q tabı¯ a uhra¯)’˙ (II.406 lines 3 – 4). ¯ ˙ ˙ ˙ This accounts for the cases al-Ra¯zı¯ mentioned, such as˘ the point, relation, and unity: ‘for instance the line: the point is not divided along with [the line], because [the point] does not inhere [in the line] insofar as [the line] is a line, but insofar as [the line] is limited’ (II.406 lines 5 – 6). Or, to use another of his examples, a figure does not inhere in a plane insofar as the plane is a plane, but insofar as the plain is bounded. Analogously, an intelligible form does not inhere in a body insofar as the body is divisible, but insofar as the body instantiates that ˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Peter Adamson

18 The argument runs from p. 210 line 6 to p. 211 line 14. 19 The argument runs from p. 211 line 15 to p. 214 line 3. 20 He later points out (I.408, line 1) that the soul itself has genus and specific difference, but of course we do not want to say that the soul is divided.

105

Avicenna and his Commentators on Human and Divine Self-Intellection

form. Therefore, the same form can inhere in the intellect without the intellect’s being divided:

˘

Commentary II.407, lines 9 – 10: Saying that it is ‘grasped intellectually’ means that it is inscribed in a substance that apprehends it. This inscribing (irtisa¯m) in that substance is not insofar as some other nature (tabı¯ a) belongs to [the thing it grasps], ˙ (bi-da¯tihi). because it apprehends [that thing] in its essence ¯

Here the ‘other nature’ is the divisibility which has accidentally accrued to granite when granite is bodily. Since intellection is a grasp of granite in its essence, rather than with this accidental nature, the grasping can occur in an immaterial subject – in fact, as we have seen, it must do so, if the form is to be grasped as a whole by the whole subject.21 The upshot is that in Avicenna’s conclusion, ‘[the intelligible] will not be inscribed in what is spatially divided’, al-Tu¯sı¯ thinks the emphasis should be placed on the word ‘spatially’ (II.408 lines ˙ 18 – 19).

Self-Intellection

˘

So far, we have seen Avicenna argue that a subject of intellection must be selfsubsisting and immaterial. His success in doing so has been called into question by al-Ra¯zı¯, who has argued that Avicenna fails to distinguish intellection from other cases of inherence. But even if, with al-Tu¯sı¯, we agree that Avicenna has ˙ argued successfully, this would not yet be sufficient. As we saw, later in the Pointers Avicenna will infer that God is a self-thinking intellect, on the basis that God is immaterial. For this inference to go through, the premise that every intellect is immaterial will not do. We need the converse claim, that every selfsubsisting immaterial thing is an intellect and self-intellective to boot. Avicenna is well aware of this, and asserts explicitly that he has provided the necessary arguments by the time we get to Pointers § 3.22 (II.429 – 30). There he says, ‘if you take in what I have established for you, you will know that (1) whatever is such as (min ˇsa nihi) to become an intelligible form and is self-subsistent is also such as to intellect; and from this it follows that (2) it is such as to intellect itself.’22 This raises the question of where in the third namat Avicenna has ˙ established the two claims he mentions here: 21 An interesting intermediate case is the estimative faculty. In estimation (wahm) the intelligible item (ma na¯) is received cognitively, but in a physical object, namely the brain. As al-Tu¯sı¯ points out (II.408, line 9 – 10), this sort of reception of an intelligible ˙ with in the immediately following tanbı¯h, namely Pointers § 3.17. concept is dealt 22 I hereby apologize for the use of the word ‘intellect’ as a verb to translate forms of aqala, rather than ‘intellectually grasp’ as used in the previous section. ‘Intellectually grasp’ is better English but would lead to considerable awkwardness in translating some of the passages to be considered in this section. ˘

˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

106

˘

(1) Every self-subsistent intelligible object is also a subject of intellection (that is, it is able to grasp intelligible objects). (2) Every intellective subject can intellect itself. For claim (2), one might immediately think of the Flying Man thought experiment, which is given a very prominent position in both the Healing and the Pointers. Avicenna uses it to begin the psychology section of the Healing and the third namat on soul in the Pointers. The thought experiment is meant to ˙ show that someone innocent of sensory experience will nonetheless have certainty regarding his own being (tubu¯t anniyyatiha¯).23 Obviously this is not ¯ irrelevant to the topic we are considering, but I will pass over the Flying Man and its treatment by al-Ra¯zı¯ and al-Tu¯sı¯. For one thing, this section of the ˙ Pointers and the two commentaries on it have been discussed in the secondary literature, albeit only in one article by Michael Marmura.24 More importantly, it is not clear that the Flying Man example is really about self-intellection as opposed to something like self-awareness.25 In neither the Healing nor the Pointers version does Avicenna say that the flying man would ‘intellect (ya qilu)’ anything.26 So I would like instead to turn to another tanbı¯h from the Pointers, which explicitly sets out to establish both of the claims just mentioned.27 It reads as follows. 23 Pointers, § 3.1 [II.344 – 5]. 24 See Marmura, Fakhr al-Din ar-Razi’s Critique. My thanks to Dag Nikolaus Hasse for the reference. 25 For self-awareness in Avicenna see Kaukua, Avicenna on Subjectivity. An interesting question here is what faculty is responsible for self-awareness, if self-awareness is not selfintellection. For Avicenna self-awareness seems to be so basic to human cognition that it underlies even sensation and imagination. Thus one might wonder whether it could really be an act of the intellect. (My thanks to Amos Bertolacci for raising this point.) See further Black, Avicenna on Self-Awareness. 26 In the version in the Healing, De anima I.1, he uses terminology similar to that found in the Pointers: the flying man will have itba¯t li-da¯tihi mawg˘u¯da (p. 16, line 7). The verb used repeatedly in the Healing version is¯ ‘affirm¯ (tabata)’. The closest Avicenna comes to ¯ Pointers, that the flying man must be speaking of intellection is when he stipulates, in the ‘of sound mind (sah¯ıha l- aql)’. ˙ ˙ ˙ more deliberate than the Healing in setting out an account of 27 The Pointers is much human self-intellection as a basis for the theory of divine self-intellection. Having said that, there is one intriguing passage in § V.6 of the Psychology in the Healing (239.3 – 10): ‘We say that the soul intellects that it takes into itself the form of the intelligibles, separated from matter. The form is separated either by the intellect’s abstracting it or because that form is in itself separate from matter, in which case the soul has no need to abstract it. The soul conceives (tatasawwara) itself, and its conceiving itself makes it intellect, intellecting and intellected. ˙But its conceiving these forms does not make it like this, because in its substance in the body, it is always potential intellect, even if it becomes actual with respect to some things.’ ˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Peter Adamson

Avicenna and his Commentators on Human and Divine Self-Intellection

˘

˘

Pointers § 3.19 [II.415 – 20]: You know that whatever intellects something intellects – through a potency close to actuality – that it is intellecting28. In so doing (min-hu) it intellects itself. So whatever intellects something may intellect itself (fa-kull ma¯ ya qilu ˇsay an fa-lahu an ya qilu da¯tahu). And whatever is intellected is such that its essence (ma¯hiyya) is conjoined ¯ to another intelligible. Thus it is also intellected along with something else, and inevitably the intellective power intellects it precisely in conjunction [with this other thing]. If [this other thing] is something that subsists through itself, then nothing in its true nature (haqı¯qa) prevents it from conjoining with the intelligible concept. Unless in fact (fı˙¯ wug˘u¯d) it is hindered from this, because it is afflicted by being conjoined with items (umu¯r) that are material (min ma¯dda), or by something else, as the case may be. But if its true nature is unimpaired (musallama), it will not be hindered from conjunction with intellectual forms. And so, [even in the case where it is hindered] it will have this as a possibility, and within this (dimna da¯lika) is the possibility of its intellecting itself. ¯ ˙ ˘

This passage is obviously related to Aristotle’s treatment of self-intellection in De anima III.4, and brings out something important about the original Aristotelian text. As I said above, the end of De anima III.4 discusses two aporiai: why intellects are objects of thought for themselves, and why external intelligibles are not all able to engage in intellection. But there is a third aporia mentioned by Aristotle, which I omitted in my previous translation of the passage: ‘the reason why intellect is not always perceiving (mudrikan) should be considered’ (430a5 – 6, p. 74 line 15 – 16 Arabic). It is not a stretch to think that the infamous following chapter, De anima III.5, is in part an attempt to resolve this aporia, since it includes the claim that the agent intellect always thinks. Nor is it a stretch to think that the explanation for an intellect’s failing always to think is the same explanation that resolved the other two aporiai: matter. In other words, it will be embodiment that results in an intellect’s being only intermittently intellecting.29 It’s clear that Avicenna has the Aristotelian passage in mind in Pointers § 3.19, just as he does in the texts we examined above. As we will see shortly, al-Tu¯sı¯ also thinks that III.4 is relevant to Avicenna’s discussion. ˙ Pointers § 3.19 begins with an explicit endorsement of claim (2): intellection of any intelligible object always brings with it the possibility of self-intellection. If we suppose, as I have just suggested, that self-intellection is not just selfawareness, then what is it? Perhaps Avicenna is making an epistemic point, namely that if I know something, then I must know that I know it. I suspect that Avicenna would accept this internalist constraint on knowledge, which was

˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

107

28 Or ‘that it is intellecting it [sc. that thing]’, if we read ya qiluhu with al-Tu¯sı¯’s lemma; as ˙ we will see this is in any case how he interprets the passage. 29 Some support for this interpretation might be had from Nicomachean Ethics X.8, 1178b33 – 5: ‘[human] nature is not sufficient for contemplation (pq¹r t¹ heyqe?m), but the body must have health, food, and other care.’

108

˘

explicitly set out by al-Fa¯ra¯bı¯ earlier in the Arabic tradition.30 Indeed, Avicenna might even think that the possibility of self-intellection implies the possibility of knowing that one knows. But even if Avicenna has this in mind in § 3.19, he is also talking about something more ambitious, namely having oneself as an intelligible object. This seems to be broader than simply knowing that one’s present judgement is an act of knowledge. Perhaps much broader: it might imply that one must at least potentially have a full grasp of one’s own essence in order to engage in intellection at all. Certainly it is this maximal interpretation that Avicenna seems to employ later on, when he discusses God as a selfintellecting intellect. Avicenna wants, then, to show that intellection implies at least the possibility that one is an intelligible object for oneself. This possibility may or may not be realised in human intellects at any given time, since human intellects are connected to matter. For an immaterial intellect like God, though, it will be a permanent feature of intellection. But why think any of this is true? Let us look at the beginning of the passage in more detail, alongside the discussions from our two commentaries. The tanbı¯h begins with the remark that self-intellection occurs through a potency or power (quwwa) which is ‘close to actuality (qarı¯ba min al-fi l)’ (II.416). As alTu¯sı¯ explains, this is to be understood in terms of Aristotle’s distinction between ˙ levels of actuality: al-Tu¯sı¯, Commentary II.416, lines 4 – 11: [Avicenna] posits three degrees of potency: ˙ distant, which is material intellect; intermediate, which is habitual intellect; and ‘close (qarı¯ba)’, which is actual intellect. This is the one required in order that the subject of intellection may behold its object whenever it wishes. His point is that whatever actually intellects something may, whenever it wishes, intellect that it itself is intellecting this thing. For its intellection of this thing is this thing’s occurring (husu¯l) to it. And its intellection of the fact that it itself is intellecting this thing is ˙ ˙occurring of this occurrence for it (husu¯l da¯lika l-husu¯l la-hu). the ˙ ˙ ¯ ˙ ˙

He seems to be taking ‘close (qarı¯b)’ here to mean something like ‘available’ or ‘ready to hand’, seeing an allusion to De anima III.4. I believe al-Tu¯sı¯ is drawing ˙ an analogy. When one has previously grasped some intelligible and thus has a habitual grasp of it, one has that intelligible ‘close’ or ‘available’. This means that one can grasp it again at will: Aristotle makes the point at De anima III.4, 429b4 – 9. In just the same way, when one is actually grasping that intelligible, one is ‘close’ to grasping that one is grasping it. Or in other words, one may at will intellect that one is intellecting. The reason al-Tu¯sı¯ gives is that ‘knowledge ˙ through assertion (al- ilm bi-l-tasdı¯q)’ involves ‘conceptualizing the subject ˙ [of the assertion] (tasawwur al-mawdu¯ )’ (al-Tu¯sı¯, Commentary, II.217.4 – 5). ˙ ˙ ˙ ˘

˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Peter Adamson

30 As discussed by Black, Knowledge and Certitude. See also Adamson, Knowledge of Universals.

109

Avicenna and his Commentators on Human and Divine Self-Intellection

Despite the technical terms,31 what he means is fairly clear. Suppose I grasp that horse is animal. This requires, indeed involves, my grasping horse. Similarly, if I grasp that my intellect is grasping something, then I also grasp my intellect. So, according to al-Tu¯sı¯, the argument in full goes like this (II.417, lines 8 – 11): ˙ Whatever intellects something may, at will, intellect the fact that it is intellecting that thing. Whatever intellects that it is intellecting something also intellects itself. Therefore whatever intellects something intellects itself. And the conclusion of this syllogism is equivalent to claim (2). But unfortunately, al-Ra¯zı¯ has a clever point which would undermine or at least weaken the second premise of the syllogism: al-Ra¯zı¯, Commentary I.172, lines 26 – 9: [Avicenna’s] saying ‘it thereby intellects itself ’ suggests that a thing’s intellecting that it is intellective of something else is the same as its intellecting itself. But this is not the case. For my knowledge that I am knowing something is knowledge concerning a specific relation between myself and the knowledge of something. And the knowledge of that relation is different from the knowledge of either of the two relata. So how can one say that the knowledge of that relation is knowledge of myself ? Rather, one must say that my knowledge that I am knowing something includes (yatadammana) or comes along with (yaltazimu) ˙ my knowledge of myself.

Al-Ra¯zı¯ may already be thinking ahead to Avicenna’s claim that God’s knowledge of other things is identical to His knowledge of Himself. (This is suggested by the shift of terminology from ‘intellection’ to ‘knowledge’, ilm.) Here he sows a doubt that self-intellection could be like this, because intellection or knowledge is a relation – as we saw above, an asymmetrical relation. Of course it is true that one must grasp X in order to grasp that X is related to Y. But the grasp of X, the grasp of Y, and the grasp of the relation are three different things. One might add, though al-Ra¯zı¯ does not,32 that the knowledge of the relata should really be prior to the knowledge of the relation. If that were so, then I would already have to have self-intellection before intellecting that I intellect some other object. Our two commentators both devote considerably more time to the rest of § 3.19, in which Avicenna seeks to prove claim (1): every self-subsisting intelligible object is also a subject of intellection. This is precisely the converse ˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

31 The contrast between tasdı¯q and tasawwur goes back at least to al-Fa¯ra¯bı¯. See the ˙ ¯ra¯bı¯ on Meno’s Paradox, p. 25, n. 32. discussion of these terms ˙in Black, al-Fa 32 Though he makes a similar point at the end of his commentary on this tanbı¯h: ‘its intellection of itself is not a part of its intellection of something else; and whatever is not part of something is not included within that thing. Rather, one must say that its intellection of things has as an implication the possibility that it may intellect itself, where this implication is from cause to effect’ (al-Ra¯zı¯, Commentary I.173, lines 11 – 18).

110

of a situation we considered above. We saw that intelligibles that subsist in matter, or insofar as they are grasped by an intellect, cannot engage in intellection. Here, we have the converse claim that every self-subsisting intelligible can engage in intellection.33 Avicenna’s proof, as I understand it, is as follows. Suppose that we have an intelligible object; call it O. If O is intelligible, then we know that it can conjoin with other intelligibles in two ways. First, when I grasp O, then O will be conjoined with the other intelligibles that I grasp. As we saw above, Avicenna will say explicitly in the next tanbı¯h, Pointers § 3.20, that this is the case. (The example I used above was that granite is ‘conjoined’ or co-present with iron even though neither granite nor iron grasp each other intellectually.) Second, if I intellectually grasp O then O will also be conjoined with my intellect, and my intellect is intelligible. For we have just seen that my intellect is intelligible to itself. The conclusion to be drawn from this is that there is in principle no obstacle to O’s being conjoined to other intelligible objects. Now suppose that O is self-subsistent. Then it can itself engage in intellection, because intellection is nothing more than a self-subsistent thing’s conjoining with intelligibles. So long, that is, as O is ‘unimpaired’. If it is hindered, for example by coming into a relation with matter,34 then this may prevent it from engaging in intellection. Furthermore, O will be able to engage in self-intellection. For Avicenna just showed with claim (2) that whatever engages in intellection can also intellect itself. But again, matter (or other hindrances, which here remain unspecified) can prevent this from happening. Thus matter accounts for the fact that intelligibles – including our own intellects – do not always intellect or self-intellect. Yet this is always theoretically ‘possible’ for any self-subsisting intelligible, as Avicenna says at the end of the tanbı¯h. 35 33 It stands to reason that these are the only three ways for an intelligible to subsist. It must subsist either by itself, or in something else; and if in something else, then in matter or not in matter, i. e. in an intellect. 34 Note that this relation will not be one of subsisting in matter, because then of course it would not be self-subsistent. What Avicenna has in mind is not an intelligible like granite, which never subsists except in matter or in an intellect, but an intelligible like a human intellect, which is self-subsisting but is sometimes ‘afflicted’ by being related to a body. 35 The two commentators’ remarks on the nature of this possibility show the sophistication with which they handled modal notions. Al-Ra¯zı¯ points out the innate possibility of intellects for self-intellection is also realised by God, who however is necessarily selfintellective. Thus the ‘possibility’ in question is ‘possibility in the general sense (imka¯n a¯mm)’ (al-Ra¯zı¯, Commentary I.172, line 23). I take this to mean ‘one-sided possibility’, i. e. the possibility which includes both the contingent and the necessary. Al-Tu¯sı¯ rejects ˙ include this, because if the possibility in question included the contingent it would things that never engage in self-intellection, though they might have. As he puts it, ‘“possibility in the general sense” applies to the remote possibility [that is present even] when something is always non-existent, but not necessarily so’ (al-Tu¯sı¯, Commentary ˙ capacity for selfII.416, lines 18 – 19). Instead, we should take imka¯n here to refer to the ˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Peter Adamson

Avicenna and his Commentators on Human and Divine Self-Intellection

In objecting to this argument al-Ra¯zı¯, in a move that is by now familiar, exploits the idea that intellection is a conjunction of mahall and ha¯ll. He ˙ ˙ distinguishes between three types of conjunction: (a) the ha¯ll’s conjunction to its ˙ mahall; (b) the mahall’s conjunction to its ha¯ll; (c) the conjunction of two ha¯lls ˙ ˙ ˙ ˙ in a single mahall (al-Ra¯zı¯, Commentary I.171, lines 25 – 6). To put the point in ˙ English, he is envisioning three relations: (a) the relation borne towards a subject by what inheres in it; (b) the converse relation which the subject bears towards what inheres it in; (c) the relation between two things that inhere in a single subject. Al-Ra¯zı¯ then argues that something’s being capable of entering into a conjunction of type (a) or (c) does not mean that it can enter into a conjunction of type (b). In other words, the fact that the intelligible object can conjoin to my intellect and to other intelligibles grasped by my intellect does not mean that it can itself be the subject of intellection. He adds a further reason to think that the inference is unproven. Quiddities have features when they are ‘mental (dihnı¯)’ that they lack when they are ‘externally real (ha¯rig˘¯ı)’. For ¯ ˘ does not. instance when man is a ha¯ll, it requires a mahall, but the real man ˙ ˙ Conversely the real man is capable of sensation, motion, and so on, but man in my mind is not (I.172, lines 1 – 3).36 Thus, ‘from the fact that the mental quiddity is conjoined to the intelligibles, it does not follow that the real quiddity is the same’ (I.172, line 4).37

˘

intellection, i. e. in light of the Aristotelian concept of first and second actuality (lines 19 – 20). 36 In response to the point, al-Tu¯sı¯ quite reasonably distinguishes between man considered ˙ as an essence and man considered as a mental item. When I grasp man intellectually I am grasping the essence mentally, not grasping a mental essence. Al-Tu¯sı¯ does admit that it is ˙ but this would be an possible to consider man precisely insofar as it is a mental item, additional act of the intellect, in which it would reflect on its own concepts (II.421, lines 13 – 16). For a very similar discussion in Yahya¯ b. ‘Adı¯, see Adamson, Knowledge of ˙ Universals, pp. 153 – 4. 37 For an independent discussion discussing proofs that the intelligible must be intellective, see Fahr al-Dı¯n al-Ra¯zı¯, Maba¯hit, I, pp. 369 – 72. In this passage al-Ra¯zı¯ sets out three ˙¯ ˘ arguments to show that every separate substance is self-intellective. The first is explicitly drawn from Avicenna, but from al-Mabda wa-l-ma a¯d (Origin and Return) rather than the Pointers. The passage concludes with a fourth argument which al-Ra¯zı¯ apparently endorses; this looks as if it is inspired by Pointers § 3.19 – 20. To summarize the argument (Maba¯hit I, p. 371, line 13 to p. 372, line 5): any given intelligible object O1 can conjoin with˙ ¯another intelligible O2 in an intellect. But this conjunction cannot presuppose the intellect’s intellection of O1 and O2, because the intellection just is the conjunction of the intellect with O1 and O2. Therefore the objects must be conjoined independently of this intellect which grasps them. In order for this to be the case, O1 must exist outside the intellect in question: ‘the intelligible quiddity always exists among the individuals, subsisting through itself ’ (372, line 2). And if O1 exists as self-subsistent, then its conjunction with other intelligibles will be due to ‘the inscribing of their forms in it’ (p. 372, line 3). ˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

111

112 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Peter Adamson

Al-Tu¯sı¯ is unimpressed by this whole line of objection. Regarding the ˙ between types of conjunction, he says that the proof needs only to distinction show that the intelligible is capable of conjunction at all (mutlaqan; II.421, line 6). This seems to be a disagreement over where the burden˙ of proof lies. AlTu¯sı¯’s point is that if the intelligible object can enter into conjunctions of type ˙ and (c), then some good reason would have to be provided for why it cannot (a) enter into a conjunction of type (b). But I think al-Ra¯zı¯ has the better of the argument here, at least if we assume that the intellect-intelligible relation is asymmetrical (an assumption Avicenna himself seems to make, as we have seen).38 The ability of an intelligible to play the passive role in this relation does not obviously imply any ability to play the active role in a relation of the same type. So there is a good reason for distinguishing between things that enter into conjunctions of type (a) and (c) – that is, things that inhere in something else – and things that enter into conjunction type (b), that is, things that are subjects of inherence. Let us take stock. In the brief compass of Pointers § 3.19, Avicenna has laid two important parts of the foundation for his theory of God. He has shown (2) that every intellect is capable of self-intellection: whenever it grasps an intelligible object it thereby grasps itself or has the ability to do so. Furthermore, Avicenna has shown (1) that any intelligible object is capable of intellection – and hence self-intellection – so long as it is self-subsisting, and so long as it is not hindered by matter or any other factor. Again, al-Ra¯zı¯ and al-Tu¯sı¯ disagree about the success of Avicenna’s arguments. Regarding claim (1), the˙ issue seems to boil down to a question of the burden of proof. Whereas al-Ra¯zı¯ demands to be shown that an object of intellection will be the sort of thing which can itself engage in intellection, al-Tu¯sı¯ seems to think the absence of an argument to the contrary is sufficient. Here˙ al-Ra¯zı¯’s position looks more intuitive. On the other hand, regarding claim (2), al-Ra¯zı¯ suggests that grasping that one knows something is not the same thing as grasping oneself. But al-Tu¯sı¯’s reconstruction ˙ something does of Avicenna’s argument renders it fairly plausible: to know imply at least the possibility of knowing oneself.

38 There is however one reason to question that the intellection relation is always asymmetrical. In self-intellection, the subject and object are the same thing, so it is hard to see how there could be any asymmetry. In the next section of the paper, we will see that al-Tu¯sı¯ uses a similar argument to refute the idea that intellection is a kind of ˙ inherence.

Avicenna and his Commentators on Human and Divine Self-Intellection

God as a Self-Intellective Intellect At any rate, Avicenna takes himself to have established claims (1) and (2): any intelligible object can itself engage in intellection, and whatever can do this can also engage in self-intellection. He now need only add that God is a selfsubsisting, immaterial and unhindered intelligible to reach the desired conclusion that God is an intellect who intellects Himself. Al-Ra¯zı¯ already tries to block this inference in his commentary on Pointers § 3.19, by questioning the basic assumption that God is an intelligible. For, he insists, man cannot have an intellectual grasp of God (I.170, line 31), as Avicenna himself would admit (I.171, line 5 – 6). Of course if God is not intelligible to us, then we have no evidence that God is intelligible at all. In which case we have no reason for applying the argument of § 3.19 to the divine case. This objection leads to an interesting debate between al-Ra¯zı¯ and al-Tu¯sı¯ (cf. the latter’s Commentary, ˙ II.419) as to whether Avicenna’s demonstration of the Necessary Existent constitutes a case of God becoming intelligible to man.39 But I will pass over this issue so as to move on finally to the seventh namat of part two of the Pointers, ˙ which is partially devoted to the topic of God’s self-intellection. Before I do so I should emphasize that I am not going to deal with the problem of God’s knowledge of particulars. I have discussed this elsewhere, albeit not with reference to the commentaries on the Pointers. 40 Instead I want to look at the more fundamental problems that arise even in saying that God grasps intelligibles (never mind particulars) by grasping Himself. The first threat to guard against is that, as we have now seen at length, God is not the only immaterial entity that grasps both intelligibles and itself. Avicenna is thus in danger of making God’s intellection too much like the intellection enjoyed by creatures. Consider for instance the following passage, from the Metaphysics (Ila¯hiyya¯t) of Avicenna’s Healing:

˘

The Necessary Existent is pure intellect because He is an essence dissociated (mufa¯riqa) from matter in every respect. You have known that the cause that prevents a thing from being apprehended intellectually is matter and its attachments, not [the thing’s] existence … Likewise He is a pure intelligible, because that which impedes a thing from being an intelligible is its being in matter and its attachments. This is the impediment preventing [the thing] from being an intellect … Hence, that which is free of matter and [its] attachments [and is] realised through existence separate [from matter] is an intelligible for itself (ma qu¯l li-da¯tihi). ¯ Because it is itself an intellect, being also intellectually apprehended by itself, it [itself ] is the intelligible [belonging] to itself ( aql bi-da¯tihi wa-huwa aydan ma qu¯l ¯ ˙ ˘

˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

113

39 Al-Ra¯zı¯ makes a similar point, and describes a possible response which anticipates alTu¯sı¯’s, at Maba¯hit I, 371 line 8 – 12. ˙ ˙¯ 40 Adamson, On Knowledge of Particulars.

114 ˘

bi-da¯tihi fa-huwa ma qu¯l da¯tihi). Its essence is, hence, [at once] intellect, intellectual ¯ ¯ – not that there are multiple things here.41 apprehender, and intelligible

The striking thing about this passage is that Avicenna is relying solely on God’s immateriality to establish the nature of divine intellection. Everything said here should, then, apply equally well to any immaterial entity. This may be why Avicenna is so emphatic that God is separated from matter in every respect (min kull g˘iha). But it is doubtful whether even this can distinguish God’s mode of intellection from that of humans after their death, or from that of the celestial intellects, which have no direct connection to their celestial bodies. Avicenna responds to this threat in a sequence found in the seventh namat, ˙ where he builds a case for the sui generis character of divine intellection. The first step is to say that God’s intellection precedes the existence of the intelligible outside Him, whereas humans normally acquire their intelligibles from what already exists on the outside: Pointers § 7.13 (III.275 – 6): The intellectual forms may in one way be acquired from external forms, for instance whenever one acquires42 the form of the heavens from the heavens. But it may be that the form first arrives at the intellective power, and thereafter external existence comes to it, for instance when one intellects a figure, and thereafter makes it exist. The Necessary Existent’s intellection of anything must be in the second way.

˘

˘

˘

Both al-Ra¯zı¯ and al-Tu¯sı¯ adopt a terminological contrast between ‘active’ and ˙ ‘passive’ knowledge ( ilm fi lı¯ vs. ilm infi a¯lı¯) for Avicenna’s distinction. As the Avicennan text itself illustrates with its example of the human geometer, ‘active knowledge’ is not the privilege of God alone. But in the case of God all His knowledge is productive, never acquired from ‘outside’ Him. This point is illustrated nicely by a remark in the Metaphysics of the Healing, where Avicenna says that God’s creative activity is like ours would be ‘if the very existence of the intelligible forms occuring in us […] were sufficient for the generation from them of the artificial forms (al-su¯war al-sina¯ iyya).’43 ˙ ˙ But surely Avicenna’s system recognizes other things that produce on the basis of grasping intelligibles? What about the Active Intellect, whose grasp of its intelligibles allows it to bestow the corresponding forms on sublunary matter? Admittedly, there will be a difference here, in that the Active Intellect receives forms and existence from superior entities, and gives form only when matter is suitably disposed. Neither is true of God. Still, Avicenna seems to be ˘

˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Peter Adamson

41 Avicenna, Metaphysics, VIII.6 § 6 – 7. I quote here from Marmura’s translation. 42 As Dag Nikolaus Hasse has pointed out to me, the term ‘acquisition’ here probably indicates that Avicenna is explicitly ruling out the case of first principles, which are not ‘acquired’ from experience but are known innately. On this see Hasse, Avicenna’s ‘De anima’, p. 179, and Adamson, Non-Discursive Thought, pp. 96 – 7. 43 Ibid. VIII.7, § 11, Marmura’s translation.

Avicenna and his Commentators on Human and Divine Self-Intellection 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

115

aware that he is in danger of making God too similar to the celestial intellects. He addresses the issue in the next several sections, Pointers § 7.14 – 16. First, he introduces a regress argument not unlike his famous argument for the existence of God. It cannot, he says, be the case that every intellect receives intellection from outside. Rather, just as something must exist through itself, so must something intellect through itself. And this will be God (§ 7.14). Furthermore, since God’s intellection consists in ‘active’ knowledge, His intellection will be the cause for what comes after it (§ 7.15). The upshot is that there are in fact three types of intellection, as follows: Pointers § 7.16 (III.279 – 80): The First’s perception (idra¯k) of things is from Himself and in Himself, and occurs in the best way that something can perceive and be perceived. Succeeding this is the perception the consequent intellective substances have of the First, through the illumination (isˇra¯q) of the First, and of what is after Him and from His essence. And after these two [kinds of perception] are the perceptions of souls (idra¯ka¯t al-nafsa¯niyya), which are a figure and sketch (naqsˇ wa-rasm) of natures that are intellective and are dispersed as principles and relation.44

Here, then, we have the difference between divine and non-divine selfintellection. All types of intellect are capable of self-intellection, as established by Pointers §3.19. But only God’s intellection consists entirely in selfintellection. The other separate intellects, the lowest of which is the Active Intellect, do grasp intelligibles and grasp themselves in doing so. Since they are immaterial, we can infer that they in fact engage in self-intellection permanently. But their grasp of the intelligibles is nonetheless derived from God, much as their existence is derived from Him, by a kind of ‘illumination’. Furthermore, Avicenna reminds us that these intellects will also have a grasp of God Himself, and this will be neither ‘active knowledge’ (the celestial intellect does not, of course, produce God) nor the consequence of self-intellection. Al-Ra¯zı¯ makes several interesting points in his commentary on these sections of the Pointers. Regarding § 7.15, al-Ra¯zı¯ challenges the idea that God will know His effects just by knowing Himself. For God can conceive of Himself simply as Himself, or as the cause of His effects. In the former case, it is begging the question to say that God’s grasp of Himself will include the knowledge of His effects. But in the latter case, what God is grasping is a relation between Himself and His effects. And, reminiscently of his criticism of Pointers § 3.19, grasping this relation will require a prior grasp of the relata, yielding a vicious circle (al44 This last phrase is difficult to understand. Goichon sees a distinction between two types of knowledge : one knowing the effect by the cause, the other vice-versa (Avicenna, Livre des directives et remarques, ed. Goichon, p. 452). Here she is following al-Tu¯sı¯’s commentary (III.281) quite closely. Though the details are obscure, it’s clear ˙ that Avicenna has in mind the sort of intellection that is accompanied by discursive reasoning, i. e. human intellection. On this see Adamson, Non-Discursive Thought.

116 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Peter Adamson

Ra¯zı¯, Commentary II, p. 69, line 35-p. 70, line 2). But the commentary is not always critical. Regarding the taxonomy of three types of intellection in § 7.16, al-Ra¯zı¯ gives a rather elegant explanation of why Avicenna is right to say that a separate intellect’s grasp of itself will not entail a grasp of God. By considering its own essence, the separate intellect will realize that it is possible-in-itself, and therefore know that its own existence required a cause. But any cause will do. It will not, by considering its own essence, come to an understanding of the specific cause that has brought it into existence, namely God (al-Ra¯zı¯, Commentary II.70 lines 8 – 30). This argument on Avicenna’s behalf meets with approval from al-Tu¯sı¯, who repeats the point in his own commentary ˙ (III.280). There remains one other problem to deal with before Avicenna can turn his attention to the more famous difficulty about God’s knowledge of particulars. The problem is already signalled in the citation from the Metaphysics of the Healing given above. There, after Avicenna describes God as intellect, intellecting, and intelligible, he immediately adds, ‘not that there are multiple things here’. In fact, as any student of Plotinus will know, there are two potential sources of multiplicity that arise when we describe God as an intellect. First, there will be the duality implied by His being both a subject and an object of intellection. This is apparently the kind of multiplicity Avicenna has in mind in this passage from the Healing. Avicenna’s position on this differs from that of Plotinus, in that he takes self-intellection to imply no multiplicity. (As we will see al-Tu¯sı¯ tries to explain why, in the case of self-intellection, knowledge is not ˙ something distinct from the knower.) But there is also the multiplicity of intelligibles that God must grasp, if He is to grasp His effects by grasping Himself. Let me stress again that this is not a worry about knowing particulars. Before we can concern ourselves with whether a simple God can grasp changing particulars like Zayd and ‘Amr, we need to understand how a simple God could grasp both man and horse. There is an irony here. When Avicenna begins to describe God in the Healing, almost the first thing he says is that God is not multiple (VIII.4, § 2). Among the consequences of His simplicity is His immateriality. And as we have seen, God’s immateriality, together with His intelligibility, implies that He is an intellect. But as an intellect, God will grasp a multiplicity of objects. And this suggests that God is Himself multiple after all. In other words, Avicenna has begun from the premise that God is not multiple and on that basis apparently argued to the conclusion that God is multiple. For simplicity entails immateriality; immateriality entails intellectivity; intellectivity entails having a multiplicity of intelligible objects; and having a multiplicity of intellective objects implies a lack of simplicity. Clearly, something has gone wrong here. Avicenna’s diagnosis of the problem is that the last inference is fallacious:

Avicenna and his Commentators on Human and Divine Self-Intellection 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

117

Pointers § 7.17 [III.281 – 5]: Perhaps you will say, ‘the intelligibles are unified neither by the subject of intellection nor by each other, according to what you have said. Furthermore, you have admitted that the Necessary Existent intellects everything. So He will not be truly one, but, in that case, multiple.’ We say that, because He knows Himself through Himself, the fact that He makes things subsist (qayyu¯miyyatuhu), as an intellect of Himself through Himself, furthermore implies that He intellects multiplicity. Multiplicity arises as a posterior consequence, not as something internal to the essence which makes it subsist. And it arises too in a certain order. A multiplicity of consequences from the essence – whether these are distinct or not – does not sully the unity [of the essence]. A multiplicity of consequences belongs to the First, both relational and non-relational, and a multiplicity of negations, and for this reason a multiplicity of names. But from this there is no effect on the unity of His essence.

Al-Ra¯zı¯’s commentary on this tanbı¯h begins with a rather polemical restating of Avicenna’s view: for Avicenna ‘the essence of God is the subject (mahall) for all ˙ those many forms’ (II.71 line 7). Yet again, al-Ra¯zı¯ deploys the idea that intellect relates to intelligible as a mahall relates to its ha¯ll. His motive in doing so ˙ ˙ becomes clear as he proceeds: he is trying to place Avicenna’s position within the context of the theological dispute over divine attributes. He says that Avicenna here departs from the views of the philosophers (fala¯sifa) on two points (II.71 line 30). Firstly, the philosophers say that nothing simple can be both active and passive in the same respect. But Avicenna is saying that God both produces and receives the intelligibles. Secondly, al-Ra¯zı¯ finds it quite striking that Avicenna allows for ‘a multiplicity of consequences, both relational and non-relational’. This implies that God has ‘fixed attributes (sifa¯t tubu¯tiyya)’, which the ¯ ˙ philosophers deny, according to al-Ra¯zı¯. He adds, ‘and how could he avoid conceding this, given that according to him God knows the quiddities, and [given that] knowledge, according to him, consists in the occurrence of the forms of [those quiddities] in the knower?’ (II.71, lines 33 – 4). This is not so much a criticism as an attempt to foist on Avicenna a view on the notorious kala¯m problematic regarding divine unity and attributes. Specifically, he claims that Avicenna’s position is closer to that of the Ash‘arites than to that of the socalled fala¯sifa. While al-Ra¯zı¯’s remarks here might seem tendentious, it must be said that Avicenna has invited this treatment by referring to God’s names.45 Al-Tu¯sı¯’s commentary on this section consists largely in a scornful response ˙ to al-Ra¯zı¯. First, he elaborates on Avicenna’s solution, insisting that the multiplicity involved in divine intellection is posterior to that intellection: ‘the multiplication of consequences and effects does not nullify the unity of their cause which gives rise to them as consequences, regardless of whether the 45 Indeed al-Ra¯zı¯’s interpretation here is quite plausible. For Avicenna’s response to the kala¯m tradition of disputes over divine attributes, see also Wisnovsky, Avicenna’s Metaphysics.

118 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Peter Adamson

consequences are established in the essence of the cause or are distinct from [that essence]’ (III.282, lines 13 – 14). Then he moves on to rebutting al-Ra¯zı¯’s claims (III.282, lines 21 ff.). He flatly denies that God is in any sense ‘passive’ as the object of His own intellection. He then rejects the idea that the intelligible is present in God as a distinct attribute. His argument here takes us back to the mechanics of human intellection: al-Tu¯sı¯, Commentary III.284, lines 7 – 11: You do not think that your being a subject ˙ all) for that form is a condition for your intellecting it. After all you intellect (mah ˙ yourself, without being a subject for yourself! Rather, your being a subject for that form is simply a condition for the occurrence (husu¯l) of that form for you. And this ˙ is a condition for your intellecting it. So if that˙ form occurs to you in another way besides its inhering in you, then intellection happens without inhering in you (hulu¯l ˙ fı¯-ka).

In passages examined above, al-Tu¯sı¯ has subtly distanced himself from the ˙ assumption so often deployed by al-Ra ¯zı¯, that intellection is just an example of inherence. Here, he finally criticizes this assumption explicitly. For al-Tu¯sı¯, when ˙ in me, I grasp an intelligible form, that form does not strictly speaking ‘inhere’ 46 but rather ‘occurs’ to me. The proof lies in self-intellection, for it is absurd to think of something’s inhering in itself, but perfectly reasonable to think of something’s being ‘occurrent’ to itself. Al-Ra¯zı¯’s view remains seductive, because in the human case a form must indeed inhere in an intellect in order for it to occur to that intellect. Presumably this is because the form comes to the human intellect from outside. But we are capable also of self-intellection, and this is akin to God’s intellection which is entirely self-directed. Humans are thus never so much like God as when they intellect themselves. In this case, there is no inherence involved, but simply the occurrence of one’s intellect to itself. One might say that Avicenna, at least on al-Tu¯sı¯’s reading, has brought together the Delphic imperative ‘know thyself ’ with˙ the Platonic idea of imitating God.47 For, if humans come closest to divine thought by thinking about themselves, then self-thinking and imitation of the divine turn out to be the same thing. On the other hand, al-Tu¯sı¯ sees significant differences between human ˙ intellection and divine intellection. Not only is our intellection partial and intermittent, but it requires that we receive a form which then ‘occurs’ in our intellects. In this respect human thought is normally ‘passive’ and posterior to an external object. Both self-intellection and ‘active intellection’ (which precedes its external object) are for us the exception, not the rule. Still, as al-Tu¯sı¯ says ˙ 46 In his commentary on Pointers § 3.19, al-Tu¯sı¯ likewise spoke of ‘occurrence’ rather than ˙ inherence; see e. g. II.416, lines 4 – 11, quoted above. 47 In fact, as Heidrun Eichner has suggested to me, Avicenna may have developed his interest in human self-intellection precisely because he was already interested in the question of divine self-intellection. In which case it is no surprise that al-Tu¯sı¯ is able to make a convincing case for resonances between the two types of thinking.˙

Avicenna and his Commentators on Human and Divine Self-Intellection

(III.284, lines 5 – 6), if humans are capable of a kind of intellection where subject and object are identical, how much more will it be the case for God? Indeed, he argues (III, p. 284, line 16-p. 285 line 12), God is as it were the limit case of this intimate relation, because there is no difference (tag˙a¯yur) at all between God and God’s intellection of Himself. Or rather, there is a difference, but only a conceptual (i tiba¯rı¯) one. Since God and His self-intellection are in reality one and the same, the effects of these two things are also one and the same. That is, ‘the existence of the first effect is the same as the First’s intellection of it, without needing any additional form that would inhere in the essence of the First.’ (By ‘the existence of the first effect’ I take al-Tu¯sı¯ to mean ˙ God’s production of the first emanated intellect, and not the intellect itself.) According to this argument, the production of God’s effects and His intellection of those effects is posterior to God’s essence. But God’s self-intellection is not posterior to His essence – rather, the two are really identical, and only conceptually distinct.48 To recapitulate this section, we have seen Avicenna argue that God’s immateriality implies that He engages in self-intellection, but also that there are principled reasons for distinguishing between His intellection and that performed by other immaterial beings. Rather polemically, al-Ra¯zı¯ suggests that Avicenna’s view here is much like theological views which accept attributes distinct from the divine essence. He also ascribes to Avicenna the view that God is in some sense passive, since He receives intelligibles. Of course al-Tu¯sı¯ takes ˙ issue with this, and in doing so makes some interesting distinctions between the standard case of grasping an intelligible object, and the special case of selfintellection – which is only one possible act of intellection for humans, but the very nature of thought in the case of God. ˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

119

Conclusion The thread of argument I have just followed, taking us from the third to the seventh namat of the Pointers, is a classic example of Avicenna’s method. He is undoubtedly ˙inspired by Aristotle, to the point that we can specify the Aristotelian passages he has in mind. In the present case, we can think of Avicenna as building a bridge from the discussion of self-intellection in De anima III.4 to the claims about divine self-intellection found in Metaphysics book Lambda.49 On the other hand, it is equally typical that he adds complex 48 This part of the commentary should be read alongside the Healing, Metaphysics VIII.7, § 4. The Healing also affirms that the intelligibles are not present to God as ‘forms’ (VIII.7, § 1), which lends support to al-Tu¯sı¯’s claim that they do not inhere in God. ˙ 49 See further Bertolacci, The Reception.

120 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Peter Adamson

new arguments for Aristotle’s claims. Thus he gives a much more elaborate rationale for the claim that any intellect can engage in self-intellection, and is more explicit about the inhibiting role of matter that Aristotle mentions in De anima III.4. He also stakes out entirely new positions of his own. For instance, as far as I can see, the view that every self-subsisting intelligible object must be a subject of intellection is not clearly present in Aristotle.50 All of this gives Avicenna a richer psychology to draw on when he comes to describe divine intellection. I hope that the foregoing has also illustrated the value of reading Avicenna alongside his commentators. Al-Ra¯zı¯ is perhaps the more philosophically interesting of the pair, with his ingenious criticisms of Avicenna. These are usually made from within the Avicennan mindset, and thus provide an ‘immanent critique’ that would be difficult for a modern scholar to duplicate. At the same time, al-Ra¯zı¯ occasionally provides arguments that would help to support Avicenna’s position (for instance, when he supplies a reason to think that a celestial intellect will not grasp God just by grasping itself ). Because of the highly dialectical nature of al-Ra¯zı¯’s commentary, it is not always easy to say what al-Ra¯zı¯’s own position on the various issues would be. To establish this would require reading further in his voluminous corpus, something I have not attempted here. As for al-Tu¯sı¯, he provides a useful balance to the commentary ˙ of al-Ra¯zı¯. More often than not, he has powerful counter-arguments to offer on Avicenna’s behalf. In my admittedly limited experience, he also seems to be more helpful in explaining the wording and structure of the Avicennan text. In general, the main value of the commentators is their careful attention to the structure, premises and potential weaknesses of Avicenna’s arguments. This is indispensible in the case of the Pointers, which was written in a deliberately allusive way such that, for instance, additional premises often need to be supplied by the reader. The commentators also expose links between different parts of the Pointers and, in al-Ra¯zı¯’s case, the relationship between Avicennan arguments and parallel disputes in Islamic theology. On the specific issues I have examined here, perhaps the most philosophically significant aspect of the commentaries is their attention to the relationship between an intellect and its object. As we have seen, al-Ra¯zı¯ frequently exploits the potential parallel between this relation and other types of inherence – for instance, the inherence of a property in a physical object. Al-Tu¯sı¯ makes efforts to specify the kind of ˙ inherence involved in intellection. But, in my judgement at least, he is not able 50 Admittedly, one might infer it from Aristotle’s rejection of Platonic Forms and embrace of separate intellects; but Avicenna seems to innovate in arguing for it as a distinct principle. Perhaps De anima III.5 might be thought to endorse the claim in question. But, given the history of contentious debate over the meaning of III.5, it seems fair to say that whatever it is that Aristotle is endorsing in that chapter he is not doing so clearly.

Avicenna and his Commentators on Human and Divine Self-Intellection

to show against al-Ra¯zı¯’s criticisms that any object of intellection must be capable of being a subject of intellection when unimpaired by matter. More successful, I think, is al-Tu¯sı¯’s analysis of self-intellection, in which the subject and object of intellection˙ completely coincide without any need for inherence. This allows him to say precisely how God’s thinking is, and is not, like that of humans. If the topic pursued here is a typical case, there is good reason to have both of these commentaries at hand while reading Avicenna, just as it is becoming increasingly common to have Aristotle’s Greek commentators at hand while reading Aristotle.

Bibliography

˘

P. Adamson, Non-Discursive Thought in Avicenna’s Commentary on the Theology of Aristotle, in Interpreting Avicenna: Science and Philosophy in Medieval Islam, ed. J. McGinnis, Leiden: Brill, 2004, pp. 87 – 111. –– , On Knowledge of Particulars, Proceedings of the Aristotelian Society, 105, 2005, pp. 273 – 94. –– , Knowledge of Universals and Particulars in the Baghdad School, Documenti e Studi sulla Tradizione Filosofica Medievale 18, 2007, pp. 141 – 64. P. Adamson and P.E. Pormann, Aristotle’s Categories and the Soul: an Annotated Translation of al-Kindı¯’s That There Are Separate Substances, in The Afterlife of the Platonic Soul. Reflections of Platonic Psychology in the Monotheistic Religions, eds J.M. Dillon and M. Elkaisy-Freimuth, Leiden: Brill, 2009, pp. 95 – 106. Aristotle, Aristu¯ta¯lis fı¯ l-nafs, ed. A. Badawı¯,Cairo: Maktabat an-Nahd. a al-Mis. riyya, 1954. ˙ ˙ Avicenna, al-Isˇa¯ra¯t wa-l-tanbı¯ha¯t ma a Sˇarh Nas¯ır al-Dı¯n al-Tu¯sı¯, 4 vols, ed. S. Dunya¯, ˙ ˙ ˙ Cairo: Da¯r al-Ma a¯rif, 1377/1957 – 1380/1960. –– , De anima, ed. F. Rahman, London/New York/Toronto: Oxford University Press, 1959. –– , Livre des directives et remarques, transl. and ed. A.-M. Goichon, Beirut/Paris: Commission Internationale pour la traduction des chefs d’œuvres/Vrin, 1951. –– , The Metaphysics of ‘The Healing’, transl. and ed. M.E. Marmura, Provo, Utah: Brigham Young University Press, 2005. –– , Psychologie d’Ibn Sı¯na¯ (Avicenne) d’aprs son oeuvre asˇ-Sˇifa¯ , transl. and ed. J. Bakosˇ, Beirut: Editions de Patrimoine Arabe et Islamique, 1982. A. Bertolacci, The Reception of Aristotle’s ‘Metaphysics’ in Avicenna’s ‘Kita¯b al-Shifa¯ ’, Leiden: Brill, 2006. –– , Avicenna and Averroes on the Proof of God’s Existence and the Subject-Matter of Metaphysics, Medioevo, 32, 2007, pp. 61 – 97. D.L. Black, Knowledge ( ilm) and Certitude (yaqı¯n) in al-Fa¯ra¯bı¯’s Epistemology, Arabic Science and Philosophy, 16, 2006, pp. 1 – 45. –– , Avicenna on Self-Awareness and Knowing that One Knows, in The Unity of Science in the Arabic Tradition, eds S. Rahman, T. Hassan, T. Street, Dordrecht: Springer, 2008, pp. 63 – 87. –– , al-Fa¯ra¯bı¯ on Meno’s Paradox, in In the Age of al-Fa¯ra¯bı¯: Arabic Philosophy in the Fourth/Tenth Century, ed. P. Adamson, London: The Warburg Institute, 2008, pp. 15 – 34.

˘

˘

˘

˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

121

122 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Peter Adamson

A. Elamrani-Jamal, De anima. Tradition arabe, in Dictionnaire des Philosophes antiques, Supplement, ed. R. Goulet, Paris: CNRS, 2003, pp. 346 – 58. D.N. Hasse, Avicenna’s ‘De anima’ in the Latin West. The Formation of a Peripatetic Philosophy of the Soul, 1160 – 1300, London/Turin: The Warburg Institute/Nino Aragno Editore, 2000. J. Kaukua, Avicenna on Subjectivity, Phd Diss., Jyvskyl, 2007. E.W. Lane, An Arabic-English Lexicon, London: Williams and Norgate, 1863 – 93, reprinted Beirut: Libraire du Liban, 1968. M.E. Marmura, Fakhr al-Din ar-Razi’s Critique of an Avicennan Tanbı¯h, in Historia Philosophiae Medii Aevi, eds B. Mojsisch and O. Pluta, Amsterdam/Philadelphia: Grner, 1991, pp. 627 – 37. M.M. McCabe, Some Conversations with Plato: Aristotle, Metaphysics Z.13–16, in Aristotle and the Stoics Reading Plato, eds V. Harte, M.M. McCabe, R.W. Sharples and A. Sheppard, London: Institute of Classical Studies, 2010, pp. 73–100. J. McGinnis and D.C. Reisman, Classical Arabic Philosophy: an Anthology of Sources, Indianapolis: Hackett, 2007. al-Ra¯zı¯, Fahr al-Dı¯n, al-Maba¯hit al-masˇriqiyya Eastern Studies, 2 vols, Tehran: Maktabat ˙¯ ˘¯, 1966. al-Asadı ˇ –– , Sarh al-Isˇa¯ra¯t wa-t-tanbı¯ha¯t, no ed., in Kita¯b Sˇarhay al-Isˇa¯ra¯t li-l-Hwa¯g˘a Nas¯ıraddı¯n ˙ ˙ ˘ al-T˙u¯sı¯ wa-l-Ima¯m Fahraddı¯n al-Ra¯zı¯, Qum, 1983/4. ˙ ˘ R. Wisnovsky, Avicenna’s Metaphysics in Context, London: Duckworth, 2003.

Essence and Existence. Thirteenth-Century Perspectives in Arabic-Islamic Philosophy and Theology Fahr al-Dı¯n al-Ra¯zı¯’s al-Mulahhas fı¯ al-hikma and the Arabic Reception ˙ ˘ ˘˘ ˙ of Avicennian Philosophy

Heidrun Eichner

˘

The al-Mulahhas fı¯ al-hikma, ‘Compendium on Philosophy’, by Fahr al-Dı¯n ˙ identified as one of the most influential works ˘ in the ˘ ˘ ˙ can be al-Ra¯zı¯ (d. 1210) Arabic reception of Avicennian philosophy from the late thirteenth century onwards. Al-Ra¯zı¯’s presentation of Avicenna’s arguments provides the conceptual basis of the majority of later texts stemming both from the philosophical and the theological tradition.1 Fahr al-Dı¯n al-Ra¯zı¯ is an Asˇ arite ˘ between philosophical and theologian whose importance for the interaction theological traditions in the Post-Avicennian period has been pointed out repeatedly.2 However, a detailed evaluation of the chronology, the context, and the intellectual outlook of his various works remains a major challenge. The complex problems surrounding these aspects of al-Ra¯zı¯’s œuvre need to be kept in mind ; however, they will not be addressed in my present analysis. Instead, this analysis will focus on some aspects of the role which al-Ra¯zı¯ has played for shaping the reception and interpretation of Avicennian ontology. In describing the al-Mulahhas fı¯ al-hikma as a presentation of Avicennian ˙ al-Mulahhas al-Ra¯zı¯ always holds ˘ ˙ in the philosophy, I do not mean ˘that ˘˘ ˙ 1

˘

˘

Works in the philosophical tradition heavily influenced by the presentation of the alMulahhas fı¯ al-hikma include Sira¯gˇ al-Dı¯n al-Urmawı¯’s Mata¯li al-anwa¯r, ‘Places of ˙ al-Abharı¯’s Kasˇf al-haqa¯ iq, ‘Unveiling Truths’, ˙ ˘ ˘ ˙ Lights’, Ascendant his Muntaha¯ alafka¯r, ‘Outmost Aim of Thoughts’, ˙and his Tanzı¯l al-afka¯r, ‘Sending down of Thoughts’; al-Ka¯tibı¯’s Hikmat al- ayn, ‘Philosophy [part] of the Core [of logical rules]’. ˙ Theological works include al-Bayda¯wı¯’s Tawa¯li al-anwa¯r, ‘Ascendant Lights’, Sˇams al˙ ˙ ¯ hiyya, Dı¯n al-Samarqandı¯’s al-Saha¯ if al-ila ‘Divine Folios’, al-I¯gˇ¯ı’s K. al-Mawa¯qif, ˙ ˙ ‘Book of Stations’, al-Tafta¯za¯nı¯’s Maqa¯sid al-maqa¯sid, ‘Intentions of Intentions’. On ˙ literary structure ˙ these writings and the adaptation of the of the al-Mulahhas fı¯ al˘ ˘ ˙see my hikma by Muslim theologians in the environment of the Mara¯g˙a-Observatory ˙Philosophical and Theological Summae, pp. 97 – 132 (Chapter IV, Hybrid forms in the philosophical tradition) and 351 – 498 (Part III, Chapters XIII to XVI). For this context and a comprehensive analysis of al-Ra¯zı¯’s ethical theories see Shihadeh, Teleological Ethics, passim. Shihadeh’s study also discusses the role of individual writings of al-Ra¯zı¯ (for a discussion of chronology see ibid. pp. 5 – 13). For this and a discussion of al-Ra¯zı¯’s biography see also Griffel, Patronage, passim. ˘

˘

˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

2

124

˘

˘

Avicennian positions. Rather, at many opportunities, he argues against Avicenna, for example when he denies the concept of ‘mental existence’ or when he holds that in God, as in contingent beings, essence and existence are distinct. Even more serious divergences from Avicenna’s philosophy result from the fact that al-Ra¯zı¯ applies a kind of ‘agenda’ in his presentation of and commentaries on Avicennian teachings. In the course of a philosophical argument (as in other texts) al-Ra¯zı¯ identifies a series of clearly distinct textual units which deal with specific problems and questions (masa¯ il).3 Analyzing these questions, al-Ra¯zı¯ uses a terminology which frequently does not render the conceptual framework of Avicenna’s philosophy faithfully. Discussing Avicenna’s position, possible objections and problems implied in the argument, al-Ra¯zı¯ relies on a unified terminology which partly obscures the conceptual divergences of the positions discussed. On the one hand, this frequently results in considerable simplifications of more complex philosophical problems, simplifications which perhaps explain the overwhelming success which al-Ra¯zı¯’s writings have had in the course of the later reception. On the other hand, these simplifications also open perspectives on new problems and allow solutions that had not been acceptable to the earlier tradition. As its title suggests, the al-Mulahhas fı¯ al-hikma – like al-Ra¯zı¯’s very ˙ ˘ ˘ ˙ ¯ hit al-mas ˇriqiyya ‘Eastern Invessimilar more comprehensive work al-Maba ¯ ˙ topics and arguments. By and tigations’ – is an exposition of philosophical large, these two works cover the issues dealt with in Avicennian works such as the K. al-Sˇifa¯ and the K. al-Nagˇa¯t in the Ila¯hiyya¯t (‘Divine Science, Metaphysics’) and tabı¯ iyya¯t (‘Physics’).4 However, their structure no longer ˙ these two disciplines but rather relies on a classification uses the division into of existents into ‘necessary’ beings (i. e., God, dealt with in kita¯b 3) and ‘contingent’ beings (dealt with in kita¯b 2). The first kita¯b is devoted to a discussion of ‘common things’ (al-umu¯r al- a¯mma), i. e., things which are common both to necessary and contingent beings.5 In this structure, topics covered by Avicenna in the Ila¯hiyya¯t can be found at various places of the al-Mulahhas. Central notions of Avicenna’s ˘ ˘ ˙relating to ‘existence’ and ‘essence’, ontological system, most notably those are discussed by al-Ra¯zı¯ in the context of the ‘common things’. The section ˘

˘

3

˘

4

On this see my Philosophical and Theological Summae, pp. 61 – 80 (Chapter III, Observations on Fakhr al-Dı¯n al-Ra¯zı¯’s Method). See the concluding remark in the al-Maba¯hith al-masˇriqiyya: ‘Now, since God has granted us to collect these problems relating˙ to physics and divine science (al-masa¯ il al-tabı¯ iyya wa-l-ila¯hiyya) according to this order and refinement ( ala¯ ha¯da¯ al-tartı¯b ¯ Fahr al˙ wa-l-tahd ¯ıb) in which we are not preceded by anyone, we conclude this book’, ¯ ˘ ˇ Dı¯n al-Ra¯zı¯, al-Maba¯hith al-masriqiyya 2:557,10 – 12. ˙ On this see my Dissolving the Unity, passim.

5

˘

˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Heidrun Eichner

125

Thirteenth-Century Perspectives in Arabic-Islamic Philosophy and Theology

on ‘common things’ consists of five chapters, devoted to (1) existence, (2) essence, (3) unity and multiplicity (4) contingent and necessary (5) preeternal and temporally originated. Here, I shall discuss mainly how two paragraphs (paragraphs 2 and 3) from the beginning of the chapter on ‘existence’ are reflected in some closely related texts from writings by At¯ır al¯ Dı¯n al-Abharı¯ (d. ca. 1265) and his student al-Ka¯tibı¯ al-Qazwı¯nı¯ (d. ca.1276).6 These two authors represent the earliest phase of the reception of al-Ra¯zı¯’s al-Mulahhas. In the course of the thirteenth century, the ˘˘ ˙ ‘illuminationist’ philosophy of Sˇiha¯b al-Dı¯n al-Suhrawardı¯ emerges as an alternative to Avicenna’s teachings in a way which has not yet been clarified. Although we certainly cannot label al-Abharı¯ as a representative of an Illuminationist tradition in the sense of a well-defined tradition of intellectual affiliation, it will be seen that al-Abharı¯’s discussion is marked by a familiarity with al-Suhrawardı¯’s writings and is partly influenced by his arguments. The interaction between al-Ra¯zı¯’s presentation with al-Suhrawardı¯’s views (much more than the original wording of Avicenna’s writings) motivates al-Abharı¯ and al-Ka¯tibı¯ to elaborate a complex ontological and epistemological discussion.

Al-Ra¯zı¯ and Avicenna’s Ontology : The Discussion on musˇa¯rakat al-wug˘u¯d

˘

˘

A most important feature of al-Ra¯zı¯’s attitude towards Avicenna’s ontology in the al-Mulahhas is that he denies the very concept of ‘mental existence’.7 Further he ˘ ˘ ˙existence is something which is shared by all existents (see text I, f. 2; asserts that § 1), and that this requires that both in contingent beings and in the Necessary of Existence existence is superadded (za¯ id) to essence (see text I, f. 3 and text I, f. 4; § 2 and § 3, the text of the al-Mulahhas contains further arguments regarding the ˘ ˘ ˙ here). According to al-Ra¯zı¯, only two Necessary of Existence not translated possible theories on the relation between essence and existence can be qualified as consistent: (1) Those who assume that ‘existence’ is superadded (za¯ id) to ‘essence’ assume that existence is shared (musˇtarak) – this is the position held by al-Ra¯zı¯; while (2) those who assume that essence and existence are identical have to assume ˘

Works by al-Abharı¯ referred to in this study include the following: Kasˇf al-haqa¯ iq MS ˙ Istanbul, Tehran, Magˇlis-i Sˇu¯ra¯-yi Millı¯ 2752, pp. 1 – 212; Mara¯sid al-maqa¯sid MS ˙ ˙ ˇ ˇ Serez 1963; Muntaha¯ al-afka¯r fı¯ iba¯nat al-asra¯r MS Tehran, Maglis-i Su¯ra¯-yi Millı¯ 2752, pp. 213 – 358; R. al-masa¯ il MS Istanbul, Ragıp Pas¸a 1461. Among al-Ka¯tibı¯’s ˇ a¯mi al-daqa¯ iq is to be mentioned: al-Ka¯tibı¯ al-Qazwı¯nı¯, Alı¯ b. Umar writings, the G ˇ a¯mi al-daqa¯ iq MS Paris, Bibliothque nationale 2370. ˇ Nagm al-Dı¯n, G For a discussion of some of al-Ra¯zı¯’s arguments see my ‘Knowledge by Presence’. ˘

˘

˘

˘

7

˘

˘

6

˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

126

˘

that existence is not shared between existents. This is the case because – according to his reasoning – an identity of essence and existence together with the assumption that existence is being shared would make it impossible that existents can be differentiated from each other: Essence, being identical to existence would have to be shared as well. While earlier theories regarding the status of non-existent entities in Islamic rational theology (kala¯m) provide one additional context in which this discussion is to be placed, Avicennian concepts constitute the framework for al-Ra¯zı¯’s arguments. Al-Ra¯zı¯ interprets the relation between Avicenna’s concepts ‘essence’ and ‘existence’ by a terminology that he also uses in his analysis of the theory of divine attributes. Al-Ra¯zı¯’s notion of musˇa¯rakat al-wug˘u¯d is very much influenced by this: Using the Avicennian essence-existence distinction for describing ‘existence’ in a framework of attribution, al-Ra¯zı¯’s understanding of existence’s being shared does not include gradations of existence. Musˇa¯rakat al-wug˘u¯d is not to be identified with homonomy of existence. The arguments provided by al-Ra¯zı¯ dominate the later tradition as well. However, by distinguishing between the existence of things ‘in the mind’ and ‘in external reality’, other authors use more differentiated argumentative strategies. Al-Ra¯zı¯’s view that not only in contingent beings but also in the Necessary of Existence essence and existence must be distinct has not gained influence in the course of later reception. Moreover, already during the earliest phase of reception of al-Ra¯zı¯’s analysis, al-Abharı¯ returns to accepting ‘mental existence’. By integrating al-Suhrawardı¯’s views regarding the ontological status of mental constructs he establishes an elaborate conceptual framework. Al-Abharı¯, however, is also the first author who – when proposing a classification of positions regarding the relation of essence and existence – neglects more subtle conceptual and terminological considerations. He establishes a pattern which derives from alRa¯zı¯’s schema and which is followed in the later philosophical and theological tradition up to the Ottoman period. The following tables show how al-Abharı¯ and a thirteenth century theological author, Sˇams al-Dı¯n al-Samarqandı¯ describe pertinent positions. As I am going to discuss, philosophical positions interact also with an interpretation of the ontological position of theological schools, i. e. most notably the Asˇ arite and the Mu tazilite positions. The nexus between the notion of musˇa¯rakat al-wugˇu¯d and the essenceexistence distinction remains dominant in the ontological expositions of later authors. Another context in which the discussion of musˇa¯rakat al-wugˇu¯d is placed is whether ‘existence’ encompasses both essentially contingent and essentially necessary beings – in other words, whether we can label God as ‘existent’ or whether he rather stands above existence. ˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Heidrun Eichner

Thirteenth-Century Perspectives in Arabic-Islamic Philosophy and Theology

Table 1: Positions as described by al-Abharı¯:a) essence and existence are identical in contingent beings

essence and mental existence are existence identical in affirmed necessary being

existence is shared

–

–

–

+

– Ibn Sı¯na¯ al-Abharı¯ in the Kasˇf al-haqa¯ iq

+

+

+

al-Abharı¯ in the Muntaha¯ al-afka¯r al-Suhrawardı¯ (as quoted by alAbharı¯) Abu¯ al-Hasan alAsˇ arı¯ ˙

+

+

(+) ‘existence is a mental construct’

existence as a mental construct is shared between things in the mind

˘

–

˘

al-Ra¯zı¯

˘

‘The Mu tazila’ a)

˘

˘

˘

˘

For al-Asˇ arı¯ and the Mu tazila see the R. al-masa¯ il quoted in nn. 15 – 20; for al-Ra¯zı¯ and Ibn Sı¯na¯ see text II, l. 3 [E] (§ 13); for al-Suhrawardı¯ see text II, l. 3 [D] (§ 12); for al-Abharı¯ in the Kasˇf al-haqa¯ iq see text II, l. 3 [E] (§ 12), for the Muntaha¯ al-afka¯r see text III, l. 1, b. 2 (§ 17). Table 2: Positions as described by Sˇams al-Dı¯n al-Samarqandı¯:a) essence and existence are identical in contingent beings

essence and mental existence are existence identical in affirmed necessary being

existence is shared

those who verify things (almuhaqqiqu¯n) ˙ philosophers

–

–

+

+

–

+

al-Asˇ arı¯ Abu¯ al-Husayn alBasrı¯ (a ˙Mu tazilite ˙ theologian)

+

+

[‘some – among the – later ones deny mental existence’]

˘

˘

a)

˘

See al-Samarqandı¯, al-Saha¯ if al-ila¯hiyya (‘The Divine Folios’), pp. 72 – 82. al-Sa˙ influence al-I¯gˇ¯ı’s K. al-Mawa¯qif which in its turn bemarqandı¯’s al-Saha¯ if al-ila˙¯ hiyya ˙ ˙ comes one of the most influential theological summae in the Sunnı¯ world. On alSamarqandı¯ see ‘al-Samarqandı¯’ in EI2 and my Philosophical and Theological Summae, pp. 379 – 424 (Chapter XIV, Shams al-Dı¯n al-Samarqandı¯ and 13th-century kala¯m). ˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

127

128

Breaking up al-Ra¯zı¯’s Matrix of Distinctions

˘

˘

Arguing against the matrix set up by al-Ra¯zı¯ remains an important motivation for later authors to develop a more refined conceptual framework. As pointed out supra, the first major step in this context is the reception of al-Suhrawardı¯’s philosophy by At¯ır al-Dı¯n al-Abharı¯ and al-Ka¯tibı¯ al-Qazwı¯nı¯. An analysis of the relevant passages ¯from works by these two authors reveals that al-Abharı¯ modifies his earlier views (as laid out in the Kasˇf al-haqa¯ iq) under the influence of arguments ˙ is continued by his student al-Katibı¯ associated with al-Suhrawardı¯, a trend which ¯ al-Qazwı¯nı¯. In the Kasˇf, presumably the earliest among al-Abharı¯’s writings introduced here, al-Abharı¯ does not accept two features of al-Ra¯zı¯’s ontology but opts for the ‘Avicennian’ position instead: al-Abharı¯ affirms mental existence, and he assumes an identity of essence and existence in the Necessary of Existence (see text II, l. 3 [E]; § 13). His discussion of musˇa¯rakat al-wugˇu¯d and that existence is superadded (za¯ id) to essence follows al-Ra¯zı¯’s course of argument. At the end of his discussion of essence and existence in contingent beings, al-Abharı¯ reports al-Suhrawardı¯’s argument that a distinctness of essence and existence would require an infinite series of hypostasized existences (see text II, l. 3 [D]; § 12 and text II, l. 3 [E]; § 13). Al-Suhrawardı¯’s argument is dismissed by al-Abharı¯ in the Kasˇf while in the al-Muntaha¯ he adopts this position (on this see infra; see text III, l. 1, b. 2; §§ 17 – 18). Fundamental divergences between the ontological system of al-Ra¯zı¯’s alMulahhas and al-Abharı¯’s Kasˇf are reflected by al-Abharı¯’s discussion of arguments ˘ ˘ ˙from his return to an Avicennian position while retaining the basic layout resulting of the exposition of the al-Mulahhas. Al-Abharı¯ adduces arguments for the concept ˘ ˘ ˙the possibility of conceptualizing impossible of ‘mental existence’, based on 8 entities. In part, these arguments were already known and discussed, though rejected, by al-Ra¯zı¯.9 The relations of essence and existence in contingent beings and in the Necessary of Existence constitute two different contexts in which new arguments are developed and tested. Most notably, the Kasˇf contains a series of eight objections against the very conception of the Necessary of Existence which are refuted afterwards by al-Abharı¯ (see text II, L3 [A] and [B]; §§ 7 – 9). This series of (albeit hypothetical) objections comes very close to arguing against the existence of the Necessary of Existence, to the best of my knowledge an absolutely singular exception in the Arabic-Islamic philosophical reception of Avicenna’s theological ontology. Most of these arguments center on the problem 8 9

See e. g. al-Abharı¯, Kasˇf al-haqa¯ iq, ‘fı¯ itba¯t al-wugˇu¯d al-dihnı¯’, pp. 110 – 11 (not translated ¯ ¯ ˙ here). v r See al-Ra¯zı¯, al-Mulahhas fı¯ al-hikma, fols 78 ,17 – 79 ,21. ˙ ˘˘ ˙ ˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Heidrun Eichner

Thirteenth-Century Perspectives in Arabic-Islamic Philosophy and Theology

of how ‘existence’ can be instantiated in concrete beings when assuming that some of these concrete beings are essentially necessary while others are essentially contingent. Al-Abharı¯ complements the refutation of these objections by an argument which is an elaboration of Avicenna’s ‘ontological’ proof for God’s existence (see text II, l. 3 [C]; § 10, other versions of this proof are not translated here).10 This argument, too, is further elaborated by al-Abharı¯ and alKa¯tibı¯ in later writings. Al-Ka¯tibı¯’s exchange of letters with Nas¯ır al-Dı¯n al-Tu¯sı¯ ˙ 11 on this topic has gained considerable prominence in the later tradition. In˙the passage from the Kasˇf translated here, based on Avicenna’s ‘ontological’ proof, al-Abharı¯ distinguishes the ‘totality of contingent beings’ vs. the ‘essentially necessary’. Al-Abharı¯’s remarks in the Kasˇf show that his understanding of Avicenna’s ontological proof (‘the proof of the philosophers’) is influenced by alRa¯zı¯’s rejection of this proof. As we further hear, al-Ra¯zı¯’s rejection is connected to his rejection of the identity of essence and existence in the Necessary of Existence (see text II, l. 3 [C]; § 11). In later writings by al-Abharı¯, the influence of al-Suhrawardı¯’s criticism of the essence-existence distinction and his insistence on the importance of mental activity for constituting the being-there of mental constructs gains momentum (among the texts translated here these are the Muntaha¯ al-afka¯r, ‘Outmost Aim of Thoughts’ and the K. al-Sˇuku¯k, ‘Book of Doubts’). This development manifests itself on the level of terminology, not so much by al-Abharı¯ and alKa¯tibı¯ adopting al-Suhrawardı¯’s terminology but rather by their readiness to break up the clear-cut distinctions set up by al-Ra¯zı¯. But also apart from alSuhrawardı¯’s influence, returning to the ‘Avicennian’ concept of ‘mental existence’ after it had been rejected by al-Ra¯zı¯, both al-Abharı¯ and al-Ka¯tibı¯ are forced to present more elaborate theories of the constitution of ‘mental existence’ which integrate the contemporary debate critical of this concept. Al-Abharı¯ and al-Ka¯tibı¯ discuss relevant arguments primarily in the context of the relation of essence and existence in contingent beings, because for them in contingent beings – other than in the Necessary of Existence – essence and existence are not simply identical. Following al-Suhrawardı¯, they point out that an identity in external existence can go along with a distinctness in the mind. The Kasˇf al-haqa¯ iq and the Mara¯sid al-maqa¯sid (‘Lookouts of Intentions’; not ˙ cf. note 16) are ˙familiar with ˙ al-Suhrawardı’s criticism of altranslated here, ¯ Ra¯zı¯’s position regarding the relation of essence and existence but do not follow it. The K. al-Sˇuku¯k likewise reports objections against al-Ra¯zı¯’s position. But since the K. al-Sˇuku¯k as a whole is devoted to objections against the al-Mulahhas ˘˘ ˙ ˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

129

10 On Ibn Sı¯na¯’s ‘ontological’ proof and its reception in the 13th century see Mayer, Burha¯n al-Siddı¯qı¯n. ˙ exchange of letters has been edited under the title Muta¯raha¯t falsafiyya bayna Nas¯ır 11 This ˙ ˙ ˙ al-Dı¯n al-Tu¯sı¯ wa-Nagˇm al-Dı¯n al-Ka¯tibı¯. ˙

130

˘

˘

˘

˘

it is not easy to determine with sufficient reliability whether the criticism in the K. al-Sˇuku¯k represents al-Abharı¯’s own philosophical position while writing the work, or whether the arguments are rather motivated by the overall critical outlook of the book. The Muntaha¯ al-afka¯r, however, in the two chapters ‘On Existence’ and ‘On Essence’, accepts the arguments of al-Suhrawardı¯. In the Kasˇf al-haqa¯ iq al-Abharı¯ had reported these arguments and dismissed them. Now, in the˙ Muntaha¯ al-afka¯r he adopts them (see text III, l. 1, b. 1 – 2; §§ 15 – 18). ˇ a¯mi al-daqa¯ iq (‘What This approach is also adopted by al-Ka¯tibı¯ in his G Collects Subtleties’) (see text V, f. 2; §§ 20 – 21). A problem is the evaluation of al-Abharı¯’s position in those of his writings which do not follow the Ra¯zı¯an model. These works contain no comparable discussion of the essence-existence relation, in particular the four works in the MS Kçprl 1618 which al-Ka¯tibı¯ has copied and studied with al-Abharı¯ in the years 628/9 H. (1228/9 AD).12 To some extent, in these works a discussion of ‘universal’ and ‘particular’ seems to replace the discussion of the essenceexistence relation. In some passages, these four works reveal an influence of Suhrawardian arguments in other contexts.13 Therefore I tend to assume that in these works the absence of a discussion of the essence-existence distinction following the Ra¯zı¯an model is the result of al-Abharı¯ emancipating himself from al-Ra¯zı¯’s position. This would also be in accordance with al-Ka¯tibı¯’s familiarity with al-Suhrawardı¯’s criticism, and with his acceptance and further elaboration ˇ a¯mi al-daqa¯ iq (see text V, f. 2; §§ 20 – 21). of al-Suhrawardı¯’s position in his G Following this line of an argument e silentio one might likewise argue that the absence of a discussion of the essence-existence distinction in al-Abharı¯’s Hida¯yat al-Hikma and al-Ka¯tibı¯’s Hikmat al- ayn (a very popular compendium, ˙ here) is to be explained ˙ not translated by their increasing distance from the position of the al-Mulahhas. This would be the case although the Hikmat al- ayn ˙ has. Further traces many features of˘ ˘its˙discussion immediately to the al-Mulah ˘ ˘ ˙ developclarification and a more precise understanding of the chronological ment of al-Abharı¯’s position regarding the constitution of ‘mental existence’ and his analysis of the essence-existence relation is to be expected from an evaluation of his theory of universals. His theory of universals is discussed not only in the context of the Ila¯hiyya¯t but in his logical writings as well. ˘

˘

˘

12 The works contained in the MS bear the titles Baya¯n al-asra¯r, ‘An Explanation of Secrets’, Talh¯ıs al-haqa¯ iq, ‘A Compendium of Truths’, Kita¯b al-Mata¯li , ‘Book of Places of ˘˙ Ascendants’, Zubdat al-haqa¯ iq, ‘Cream of Truths’. The MS˙ Kçprl 1618 contains ˙ by the hand of al-Abharı¯ for al-Ka¯tibı¯ at the end of each of reading certificates signed these works. 13 This is the case, for example, in the discussion of God’s knowledge of particulars: In the Baya¯n al-asra¯r and in the Zubdat al-asra¯r al-Abharı¯ displays a familiarity with Suhrawardian arguments. On the underlying Suhrawardian arguments see my ‘Knowledge by Presence’. ˘

˘

˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Heidrun Eichner

Thirteenth-Century Perspectives in Arabic-Islamic Philosophy and Theology

˘

The explicit quotations from al-Suhrawardı¯’s K. al-Talwı¯ha¯t in the Kasˇf al˙ haqa¯ iq and their acceptance in the Muntaha¯ al-afka¯r establish an immediate ˙ textual link between al-Abharı¯ and al-Suhrawardı¯. In particular the concluding remark in the Kasˇf shows that the positions of Ibn Sı¯na¯, Fahr al-Dı¯n al-Ra¯zı¯ and ˘ ¯. In the Kasˇf, he al-Suhrawardı¯ are the major points of reference for al-Abharı explicitly sides with Ibn Sı¯na¯ (see text II, l. 3 [E]; § 14).

Philosophical Positions and Theological Affiliations

˘

An evaluation of the elaboration of philosophical arguments, however, has to take external factors into consideration as well. Most notably, there is the problem of how these arguments relate to theological positions. As we posses few precise information regarding the chronological order of most writings, it is often not easy to determine whether differences between positions held in various works reflect a development in the thought of an author, or whether varying circumstances and conventions (i. e. whether a work belongs to a theological or a philosophical context) has influenced the author’s exposition. The case of al-Abharı¯’s Mara¯sid al-maqa¯sid and his R. al-masa¯ il can show ˙ ˙ how al-Abharı¯’s adaptation of Suhrawardian arguments is situated in a more complex context of intellectual affiliations. Thus, the Mara¯sid al-maqa¯sid ˙ ˙ describe the arguments for the identity of essence and existence in external reality as one of the arguments of the ‘natural philosophers’ against the ‘basic rules of the divine [philosopher]’ – groups which so far cannot be identified as associated with known individual authors. Possibly this refers rather to the discipline mentioned.14 In addition to philosophical affiliations, theological schools, most notably the mutually antagonistic Asˇ arites and Mu tazilites, are referred to in the debate. Thus, in the R. al-masa¯ il al-Abharı¯ portrays the position of an identity of essence and existence as the position of the founder of the Asˇ arite school Abu¯ al-Hasan al-Asˇ arı¯ (d. ca. 935) – an interpretation which ˙ might have constituted an important motivation for al-Abharı¯ to integrate alSuhrawardı¯’s criticism of the contemporary interpretation of Avicennian philosophy in his philosophical interpretation. He portrays his own ontological position as the position of Abu¯ al-Hasan al-Asˇ arı¯ and contrasts it with a ˙ Mu tazilite position which incidentally is identical to the ontological position laid out by (the Asˇ arite) Fahr al-Dı¯n al-Ra¯zı¯ in the al-Mulahhas. Thus al-Abharı¯ ˘ ˘ ˙the prism of alpresents his interpretation ˘of Avicennian positions through Suhrawardı¯’s arguments. He presents his detour from al-Ra¯zı¯’s ontology not as ˘

˘

˘

˘

˘

˘

˘

˘

˘

˘

˘

˘

˘

14 Mara¯sid al-maqa¯sid fol. 33v : ‘al-marsad al-ha¯mis […] yahtawı¯ ala¯ al-i tira¯da¯t allatı¯ ˙ biha¯ min ˙gˇihat al-hukama¯ al-t˙abı¯ iyyı¯˘n ala¯ qawa¯ id ˙al-ila¯hı¯’. ˙ yuqdah ˙ ˙ ˙ ˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

131

132

˘

˘

an antagonism between a ‘philosophical’ position and a kala¯m-position but as a return to orthodox Asˇ arite teachings set against Mu tazilism. This can be seen from the introductory statements to relevant questions in the Mara¯sid al-maqa¯sid which refer to positions which turn out to be identical ˙ ˙ to the position as depicted in some of al-Abharı¯’s philosophical writings (e. g. in the Muntaha¯ al-afka¯r). Thus, at the beginning of the second question, al-Abharı¯ states: ˘

The Ima¯m Abu¯ al-Hasan al-Asˇ arı¯ teaches that the existence of every contingent being is identical to˙ its external essence. The Mu tazila says that existence is superadded to the external essence.15 ˘

˘

In a similar vein, al-Abharı¯ describes the assumption of ‘mental existence’ as supporting the Asˇ arite position that ‘what is not existent is not a thing’ – otherwise essences would have to be persistent in outward reality with existence later being superadded to them.16 al-Abharı¯ further points out that the assumption of an identity of essence and existence in God is the position of alAsˇ arı¯.17 Applying Avicennian terminology and analytical categories to the interpretation of the position of authors – most notably theologians – predating Avicenna is an anachronism typical of how the amalgamation of theological and philosophical positions in the thirteenth century takes place. As we have seen, in the context of these ontological problems, al-Abharı¯ uses Avicennian positions for interpreting the Asˇ arite position and for distancing himself from what he describes as Mu tazilı¯ positions. In part, al-Abharı¯ regards the Asˇ arite position and arguments as identical to that expounded in his (later) philosophical writings. The complexity of the situation regarding the evaluation of intellectual affiliations in al-Abharı¯’s writings may be highlighted by another very similar case. This is likewise a problem heavily disputed between the philosophers and the theologians, i. e. whether atomism or hylomorphism is to be accepted as basic ontological explanation underlying physical reality. Typically, atomism is described as the position of the mutakallimu¯n who oppose (Aristotelian) hylomorphism. R. al-masa¯ il, question (11) ‘On the atom’ first gives arguments for atomism, then arguments against it. Then, al-Abharı¯ concludes his enumeration of anti-atomist arguments by the statement: ˘

˘

˘

˘

˘

˘

15 al-Abharı¯, R. al-masa¯ il, fol. 366r,2 – 4. 16 See the discussion of mas ala (3) fı¯ anna al-ma du¯m laysa bi-sˇay (what is non-existent is not a thing), R. al-masa¯ il, fols 366v,15 – 367v,5. This runs very much counter to standard expositions of Avicennized Asˇ arism in late 13th-century kala¯m. 17 See the discussion of mas ala (5) fı¯ anna wugˇu¯d al-ba¯rı¯ nafs haqı¯qatihı¯ (the existence of the creator is identical to his core-essence), R. al-masa¯ il, fol.˙ 368v,11 ff. ˘

˘

˘

˘

˘

˘

˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Heidrun Eichner

Thirteenth-Century Perspectives in Arabic-Islamic Philosophy and Theology

This question rightly causes one to hesitate (yastahiqqu al-taqawwuf fı¯ha¯) – or to admit that the atom has to be negated because of ˙these aspects.18

Here, al-Abharı¯ is ready to accept the arguments against atomism, and he does not describe the opposition between pro- and anti-atomist positions as an antagonism between mutakallimu¯n and the philosophers. Nevertheless, in the next question (12) he seems to prefer a rejection of hylomorphism. At least, this is what the title of the question ‘On rejecting hylomorphism’ suggests. In his philosophical writings, however, al-Abharı¯ rather defends hylomorphism.19 He associates hylomorphism with the philosophers (fala¯sifa) and ascribes the rejection of hylomorphism to the mutakallimu¯n. Again, al-Abharı¯ expresses his doubts and hesitation at the end of the question: ˘

This question, too, rightly causes one to hesitate because the arguments bring each other to a standstill (li-ta a¯rud al-adilla). God is the one who provides guidance.20 ˙

The problem of the ontological status of mental constructs and the antagonism between atomism and hylomorphism represent ‘classical’ questions disputed between philosophers and theologians. While by the thirteenth century these problems have a centuries long history, another set of discussions touching upon related basic ontological assumptions begins to emerge only at that time. These are questions centering on how individuation (tasˇahhus) and concretisation ˘ ˘ ˙ mental and external (ta ayyun) take place. The Avicennian distinction between existence changes the role which (corporeal) matter plays for the conception of individuation and concretisation. From this result very complex discussions in later authors. For example, discussions of the ontological status of mental abstractions which are described as concepts (and hence taken as concrete beings existing in the mind) and their relation to other concrete beings, both in external and mental reality, motivate later authors such as Sa d al-Dı¯n al˙ complicated Tafta¯za¯nı¯ (d. 1390) in his Maqa¯sid al-Maqa¯sid to highly ˙ ˙ typically associated with elaborations. Interactions with ontological traditions Islamic ‘mysticism’ are to be considered in this context as well. In addition to the tradition associated with Sˇiha¯b al-Dı¯n al-Suhrawardı¯, this includes the tradition going back to Ibn al- Arabı¯. In the texts translated here, some elements of this tradition are dealt with in the discussions of how God as the Essentially Necessary of Existence relates to the instantiations of contingent beings. ˘

˘

˘

Only a minor percentage of relevant texts has been published so far, and few reliable translations of these highly complex documents exist. Already from this ˘ ˘

18 al-Abharı¯, R. al-masa¯ il, fol. 373v,14 – 15. 19 For the Kasˇf al-haqa¯ iq see for example the first book of the section of physics, pp. 169 – ˙ 73. 20 al-Abharı¯, R. al-masa¯ il, fol. 375r,13 – 14. ˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

133

134

˘

˘

˘

material it can be discerned that the discussion reaches a level of complexity which is mirrored on the level of terminology as well. The texts selected and presented in translation here may serve to document some features of this discussion in Arabic texts of the thirteenth century. The selection of texts centers on the discussion of musˇa¯rakat al-wug˘u¯d and its relation to the essence-existence distinction, including some aspects of ‘ontological’ proofs for the Necessary of Existence. The al-Mulahhas fı¯ al-hikma by Fahr al-Dı¯n al-Ra¯zı¯ (Text I) provides the ˙ ˘ ˙ discussions starting-point for˘ the of later˘authors, al-Abharı¯’s Kasˇf al-haqa¯ iq and ˙ his Muntaha¯ al-afka¯r are important specimens (Texts II and III). Al-Abharı¯’s K. ˇ al-Suku¯k (Text IV) documents how closely he has studied the al-Mulahhas fı¯ al˘˘ ˙ ˇ a¯mi al-daqa¯ iq (Text V) shows how al-Abharı¯’s arguments hikma. Al-Ka¯tibı¯’s G ˙ are continued and refined by his student al-Ka¯tibı¯ al-Qazwı¯nı¯.

Primary Texts – Translations (I) Fahr al-Dı¯n al-Ra¯zı¯, The Compendium on Philosophy (al-Mulahhas fı¯ al˘˘ ˙ hikma,˘ MS Berlin, Staatsbibliothek MS Or. Oct. 629): ˙

˘

(§ 1) fasl (2): On existence being shared (fı¯ anna al-wugˇu¯d musˇtarak) because of ˙ two reasons: wagˇh (1): If we know that something contingent has a cause then we judge that this cause exists. Our doubts (taraddud) whether it [i.e., this cause] is ‘necessary’ or ‘contingent’, or ‘substance’ or ‘accident’ does not affect our first judgement. Then, when we believe ‘that it is necessary’, this belief ceases by our belief ‘that it is contingent’. If existence were not common to all divisions (bayna gˇamı¯ al-aqsa¯m), it would cease when the belief of some specifics ceases ( inda zawa¯l i tiqa¯d al-husu¯siyya¯t), in the same way as the belief about each single one ˙˙ of the specifics ˘ceases by the belief about another specific. wagˇh (2): The referent of negation (mafhu¯m al-salb) is one single referent insofar as it is negation (salb). If what is contrary to it (al-muqa¯bil lahu¯) were affirmations with various referents (ı¯gˇa¯ba¯t muhtalifa al-mafhu¯m) the exhaustive ˘ a unified referent (muttahida lenumeration (hasr) would be false. If they have ˙ ˙ ˙ mafhu¯m): q. e. d. ˘

˘

(§ 2) fasl (3): On existence being additional to the essences of contingent beings ˙ (fı¯ anna al-wugˇu¯d za¯ id ala¯ ma¯hiyya¯t al-mumkina¯t): The existence of blackness (sawa¯d), for example, either is identical to its being blackness (nafs kawnı¯hı¯ sawa¯dan) or intrinsic to it (da¯hilan fı¯hi) or ˘ because of extrinsic to it (ha¯rig˘an anhu). The first two possibilities are false ˘ ˘

˘

˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Heidrun Eichner

Thirteenth-Century Perspectives in Arabic-Islamic Philosophy and Theology

˘

several reasons. Some of these [reasons] pertain to both [possibilities, 1 and 2], some of them pertain only to one. Those which pertain to both of them are two aspects: wagˇh (1): Intellection (ta aqqul) of a heptagon is possible while having doubts about its outward existence ( inda al-sˇakk fı¯ wugˇu¯dihı¯ al-ha¯rig˘¯ı) – as it is obvious – and about its mental existence (al-wugˇu¯d al-dihnı¯) – if˘ it [i.e., mental existence] were to be affirmed (law tabata), (for even¯ if it follows from being aware of it it does not follow that this¯ occurs as part of being aware of it (wa-in ka¯na la¯ziman li-l-sˇu u¯r bihı¯ lakinnahu¯ g˙ayr la¯zim fı¯ al-sˇu u¯r bihı¯). Therefore, for someone who knows a heptagon ( alima al-musabba ) it is possible to deny its mental existence). What is doubted about is not identical to what is not doubted about (al-masˇku¯k fı¯hi laysa nafs g˙ayr al-masˇku¯k fı¯hi), and it is not part of it (da¯hil ˘ fı¯hi). As to doubting about existence (tasˇakkuk fı¯ al-wugˇu¯d): If by this it is meant doubting about existence being affirmed for existence (tasˇakkuk fı¯ tubu¯t al¯ wugˇu¯d li-l-wugˇu¯d): This is impossible because ‘non-existence’ ( adam) and ‘existence’ cannot be predicated of ‘existence’. If doubting about its being present for an essence (fı¯ husu¯lihı¯ li-l-ma¯hiyya) is meant, what we have said is to ˙ ˙ be applied. wagˇh (2): If we take blackness together with existence, under this condition it is not susceptible (qa¯bil) of non-existence; and vice versa. If we take it regardless of ‘existence’ and ‘non-existence’, then it is susceptible of both of them. Hence its ipseity (huwiyya) which is susceptible of both of them is something different (mug˙a¯yir) from both of these opposite stipulations (alqaydayn al-muta a¯nidayn). ˘

˘

˘

˘

˘

˘

˘

˘

˘

(§ 3) Now the specific [aspects]: wagˇh (1): What falsifies existence being identical to essence is: If existence of blackness (sawa¯d) were identical to its being blackness, then whiteness would not be sharing its existence, in the same way as it is not sharing its blackness-ness (sawa¯diyya). wagˇh (2): If we were to say that ‘a substance ( gˇawhar) is existent’, this would take the same rank of bearing no information (fı¯ adam al-fa¯ ida) as if we said that ‘a substance is a substance’. wagˇh (1): What falsifies existence being part of it is: If this were the case, then it would be the most general of [all] shared essential things (a amm alda¯tiyya¯t al-musˇtaraka). Hence it would be a genus ( gˇins), and the species (anwa¯ ) ¯which fall under it (da¯hil tahtahu) would be distinguished from each other by ¯ ˙ the genus would form part of the nature of the ˘ differences that exist. Therefore, differentia, and its being differentiated from the species would require another differentia, and so on to infinity. But the differentia which constitutes the division (al-fasl al-muqassim) has a cause ( illa) for its existence, and thus ˙ ˘

˘

˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

135

136

existence would have a cause for its existence. Hence, existence would have another existence – this according to what Ibn Sı¯na¯ says. Then, the differentiation of ‘the necessary’ from ‘the contingent’ would be by a differentia which constitutes it, and hence ‘the necessary’ would be composed. Existence would be constituting the things which fall under it (al-umu¯r al-mundarigˇa fı¯hi). If it were in itself independent from a substrate, it would be a substance, and it would be part of an accident. Then, an accident would be a substance. This is absurd.

˘

˘

(II) At¯ır al-Dı¯n al-Abharı¯, Unveiling Truths regarding Settling Subtleties (Kasˇf al¯ haqa¯ iq fı¯ tahrı¯r al-daqa¯ iq, MS Tehran, Magˇlis-i Sˇu¯ra¯-yi Millı¯ 2752): ˙ ˙ (§ 4) la¯mi (1): On its [= existence] being shared between existents (almawgˇu¯da¯t). This may be clarified by various aspects: wagˇh (1): We conceive the notion (natasawwar ma na¯) of existence ˙ intuitively (bi-l-badı¯ha), and we judge that it is true (nagˇzam bi-sidqihı¯) for ˙ each of the existing things. If this what is conceptualized (al-mutasawwar) ˙ would not be common between them, the judgement that it is true (al- gˇazm bi21 sidqihı¯) for each single one would be impossible. ˙ wagˇh (2): We conceive the referent of existence and non-existence (mafhu¯m al-wugˇu¯d wa-l- adam) intuitively, and we pass the judgement that if the second is false for something, then the first is true. If existence were not something shared between existents, then from the falseness of the second for a thing there would not follow that the intellect judges that the first is true for it (gˇazm al- aql bi-sidq al-awwal alayhi), because then it would be possible that both of them ˙ are false. wagˇh (3): If the judgement that something is present among concrete beings, then the judgement about its existence is there (hasala). If existence were ˙ ˙ to have two referents, then from the judgement of the mere fact that something is among concrete beings would not follow that existence is true for it. For it would be possible that another referent [of ‘existence’] would be false for it, when there is the mere judgement that it is among concrete beings. ˘

˘

˘

˘

˘

21 Read: la-staha¯la instead of li-stiha¯la. ˙ ˙

˘

(§ 5) la¯mi (2): On existence of contingent beings being additional (za¯ id) to their essences, because if it were not additional, then it would either be identical to the essence or intrinsic to it. Both possibilities (qisma¯n) are false. The first [option is false] because of several aspects: ˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Heidrun Eichner

Thirteenth-Century Perspectives in Arabic-Islamic Philosophy and Theology

˘

wagˇh (1): Existence is shared between all existents, and no one of the contingent essences is so. wagˇh (2): Under the condition of existence, the existence of a contingent essence is necessary, but insofar as it is what it is (min haytu hiya hiya) its ˙ ¯ of existence’ is existence is not necessary. Hence ‘essence under the condition different (mug˙a¯yir) from ‘essence insofar as it is what it is’. Hence ‘existence’ is different (mug˙a¯yir) from ‘essence insofar as it is what it is’. wagˇh (3): If existence were identical to the contingent essence, if we say that ‘blackness is existent’, for example, and that ‘blackness is blackness’ and that ‘an existent is existent’, this would have one meaning. The consequent22 is false because the judgement of the truth of the last two [examples] realizes only the conception of that about which the predication is made ( unwa¯n al-qadiyya), ˙ other than the first.

˘

(§ 6) The second [option is false] because of several aspects: If existence were part of essence, then – if the essence were simple – the composition of simple things would follow. This is absurd. If it were composed, the existence of an essence would be preceding it because it is necessary that the part precedes the whole. If someone says: ‘If existence were additional to the contingent essence, it would have a ipseity (huwiyya) among concrete beings (a ya¯n) behind the ipseity of the essence in which it inheres. Then, its substrate (mahall) would have an ˙ existence which precedes it by existence, and essence would have existence before its existence. This is absurd. For if it were additional, it would follow that existence would be subsisting (qiya¯m) by something which is not existent’ – we say: ‘We do not concede that it follows that the substrate of the ipseity of existence has another existence. Why is it not possible that it precedes it by its self (bi-nafs da¯tiha¯), not by yet another existence?’ As to his claim that the ¯ ¯ m) of existence by something which is not existent would subsistence (qiya follow, we say: ‘We do not concede this. What follows is the subsistence of existence by the essence while essence is existent. Yet, its existence is different from it’. ˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

137

(§ 7) la¯mi (3): On establishing the Essentially Necessary. If none of the existences were essentially necessary, then the nature of existence insofar as it is what it is either would be essentially independent from something else, or it would be in need of it. The second is impossible, because if it [i.e., the nature of existence] were essentially in need of something else it would be essentially contingent. Inevitably it would need a cause which precedes it by existence. Then, existence would precede it [i.e., the nature of existence]. If 22 Read: al-ta¯lı¯ instead of al-ta¯nı¯. ¯

138

˘

it [i.e., the nature of existence] were identical to existence, something would precede itself. This is absurd. If it [i.e., the nature of existence] were a specific existence (wugˇu¯d ha¯ss), a concretised existence (wugˇu¯d mu ayyan) would precede ˘ ˙˙ This is absurd. The first [option], too, is absurd because the nature of existence. if existence were essentially not in need of something else it would either follow for it an existence which is concretised by itself or by something else, or nothing at all would follow. In the first case, the species of existence would be restricted to one individual (naw al-wugˇu¯d munhasir fı¯ ˇsahsihı¯). In the second case, and if ˙ ˙ ˘ ˙ nature, then, is not in need of they are not two things caused by a cause (for the something else) follows that it is possible that the nature can be split from all existences. This is absurd. ˘

(§ 8) [A] It might be said: ‘Nothing is essentially necessary because of several aspects’: (1) If some among the existences were necessary, the nature of existence either would be essentially not in need of something else, or it would be in need of it. The second is absurd, as you have explained. The first, too, is absurd, because if existence were essentially not in need of something else then from it would follow either an existence which is concretised by itself, or by something else, or nothing at all would follow. The first is absurd, as you have explained. The second, too, is absurd. Otherwise it would follow that it is possible that the nature of existence is split from both these existences. This is absurd. (2) If some one among the existences were essentially necessary, its concretisation would be either because of the nature of existence or not. The first is absurd; otherwise the species of existence would be restricted to one individual so that existence would not be shared. This is absurd. The second is absurd; otherwise it would follow that the Necessary of Existence needs for its concretisation something else. This is absurd. (3) If concretised existence were essentially necessary its concretisation would be either ‘something relating to existence’ or ‘something relating to nonexistence’ (wugˇu¯diyyan aw adamiyyan). The second is absurd because concretisation is a part of what is concretised and what is concretised is existent, so that its part is existent. The first is absurd because if it were ‘something relating to existence’ it would be so because of a cause which precedes it. If this cause were this existence itself it would be concretised before its concretisation. If it were something else, the essentially necessary would be in need of something else in its concretisation. This is absurd. (4) If concretised existence were essentially necessary, then ‘existence insofar as it is existence’ would be either essentially in need of something else, or not. The second is absurd; otherwise no need would accede to it so that the existence of contingent beings would not be in need. This is absurd. The first is absurd ˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Heidrun Eichner

Thirteenth-Century Perspectives in Arabic-Islamic Philosophy and Theology 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

139

because otherwise the essentially necessary would be ‘in need’ because of its essence. This is absurd. (5) Existence insofar as it is what it is either requires that it is connected to an essence or it requires that it is not connected or it does not require one of these two things. The second is absurd; otherwise the existence of contingent beings would not be connected to their essences. This is absurd. The third is absurd; otherwise both ‘connection’ and ‘being-abstracted from anything else’ (muqa¯rana wa-tagˇarrud) would be because of a cause, and then the essentially necessary would be in need of something else in its abstraction. Hence the first has to be taken into consideration, and so no one among the existences is essentially necessary. (6) If the concretised existence were necessary it would be self-subsistent. If its subsistence by itself were by existence itself or by something which follows from it, every existence would be abstract. This is absurd. If it were by something which follows from it, its existence would be connected to some essence while the opposite has been postulated. If it were by something separate (muba¯yin), the existence of the Necessary [of Existence] (al-wugˇu¯d al-wa¯gˇibı¯) would in its abstraction be in need of a cause which is distinct. This is absurd. (7) If the essentially necessary were existence then every contingent being would be attributed with something identical to the necessary existence (bi-mitl ¯ al-wugˇu¯d al-wa¯gˇib). This is absurd. (8) If the essentially necessary were abstract existence, then the essentially necessary would be composed out of existence and non-existence, i. e. abstraction in the sense of ‘non-existence of its being connected with an essence’. This is absurd. (§ 9) [B] The answer: (1) We say: We do not concede that if nothing of the existences follows from existence, it follows necessarily that it is possible that the nature can be split from both existences. This follows only if no concretised existence is essentially necessary. Its necessity, however, is evident because we argue based on this assumption. (2) We say: We do not concede that if its concretisation were not because of the nature of existence it would be because of a separate cause. This follows only if concretisation were ‘something relating to existence’. Why do you say that it is ‘relating to existence’? For according to us, to the nature of existence accede different concretisations, some of them because of the essence which is susceptible of them, and some of them because of the absence of something susceptible. Those which accede to it because of the absence of something susceptible are something ‘relating to non-existence’ which is ‘abstraction from a substrate of inherence’ and ‘not-being-mingled with contingent beings’.

140

(3) We say: Why is it not possible that the concretisation is ‘something relating to non-existence’? As to the statement, that ‘what is concretised is existent, and concretisation is a part of what is concretised, so it must be existent’, we say: If you mean by your statement that ‘what is concretised’ is two existences and that existence accedes to what is concretised insofar as it is concretised, then it is not permissible. For existence is predicated of this existence like the species is predicated of its individuals. If you mean by it that this existence is concretised existence we concede this. But why do you say that the concretisation of what is judged to be a concretised existence is something existent? This, because – according to us – the pure [nature] of existence (sira¯fat al-wugˇu¯d) is something by which this existence is concretised, and˙ it is ‘something relating to non-existence’. It is true for pure existence that it is essentially necessary according to its essence which is existence. (4) We say: Why do you say that the need accedes to existence insofar as it is what it is? It accedes only to specific existences. This, because the contingent essences have the dispositions for specific existences so that on them emanate these existences from the necessary existence. ‘What is in need of ’ is only the concretised existence. ‘Existence insofar as it is what it is’ is essentially not in need of something else. (5) We say: Why do you say that the third alternative is absurd? Regarding the statement that both connection and abstraction are because of a distinct cause, we say: We do not concede this. This would follow necessarily only if abstraction were ‘something relating to existence’ which is in need of a cause. This is not the case. Rather this is a ‘stipulation relating to non-existence’ (qayd adamı¯) and it is not caused by a cause. (6) From this results the answer to the sixth argument. (7) We say: Why is it not possible that each of the contingent beings is attributed by one of the individuals of the nature (mawsu¯f bi-fard min afra¯d al˙ tabı¯ a)? This individual is according to its specificity essentially contingent. The ˙claim that it is absurd is not permissible, because there is no demonstration. Rather it is a mere assumption that it is unlikely (istib a¯d), nothing else. (8) We say: We do not concede that composition must follow. This, because ‘the essentially necessary’ is existence which is acceded by the non-existence ( adam) of an essence. ‘Non-existence’ is not a part of it but an accident, so that it does not lead to composition.

˘

˘

˘

˘

˘

(§ 10) [C] You must know that the philosophers (al-hukama¯ ) have explained that some existent is essentially necessary by another ˙method. They say: The totality of contingent beings must have something which exerts an influence (mu attir) from outside. What is outside of the totality of contingent beings is ¯¯ necessary. The existence of the essentially necessary must be identical essentially to its essence because if it were additional it would be in need of the essence. ˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Heidrun Eichner

Thirteenth-Century Perspectives in Arabic-Islamic Philosophy and Theology

Then it would be essentially contingent, so that it would need something which exerts an influence. If what exerts an influence on it were identical to this essence, it follows that it precedes existence by existence because what exerts an influence on existence must precede it by existence. If it were not identical to it [i.e., the essence], the essentially necessary in its existence would be in need of something else. This has to be deliberated [critically], for when they say that the totality of contingent beings must have something which exerts an influence outside of it, this is not permissible. This would need to be demonstrated.

˘

˘

(§ 11) The Ima¯m [Fahr al-Dı¯n al-Ra¯zı¯] does not accept what they say that if ˘ what exerts the influence on it is identical to this essence it follows that it precedes existence by existence. What the rejection rests on is that a contingent essence insofar as it is what it is accepts existence, but it does not precede it by existence. It is possible that the agent (fa¯ il) is like that. The philosophers say that this is a sophistry (muka¯bara). We know by necessity that what exerts an influence (mu attir) on the existence of something must precede it by existence. ¯¯ is susceptible [of existence] does not exert an influence on The essence which existence, and therefore it does not have to precede existence by existence. He has already explained that some concretised existence is necessary of existence, and that it is not connected to an essence. Otherwise it would be in need of something else, and what is essentially necessary would be essentially contingent. This is absurd. (§ 12) [D] The author of The [Philosophy of] Illumination (sa¯hib al-isˇra¯q) [al˙ the existence Suhrawardı¯] has another method regarding existence. He says˙ that of contingent beings in external [reality] is identical to the essences, for were they distinguished from each other, the essence would have another ipseity, and so both of them would be existent in external [reality]. Then, essence would have another existence and existence would have another existence, and essence would have yet another ipseity. Thus, both of them would be existent in external [reality]. Hence it would follow that the essence has infinite existences. Therefore, essence in concrete beings is one thing, and only the intellect splits external essence into two things, ‘essence’ and ‘existence’, so that two forms are there in intellect which correspond (muta¯biq) to external essence. ‘Common ˙ place in concrete beings but it is existence’ (al-wugˇu¯d al- a¯mm) does not take only in the mind. The existence which takes place is the existence pertaining to the necessary (al-wugˇu¯d al-wa¯gˇibı¯) which is abstract from essence. When it is there in the intellect, intellect does not split it into two things, ‘essence’ and ‘existence’, rather only existence results from it. Its existence among concrete beings is not connected to some kind of essence. [This is so] because if it were connected to some kind of essence this essence would be a universal essence so that it would have particulars in intellect (gˇuz iyya¯t aqliyya). Then, the relation ˘

˘

˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

141

142

of essence to [all of ] them would be identical, and it would not require the existence of one of them because this would be a preponderance without an agent (targˇ¯ıh bi-la¯ muragˇgˇih). The essentially necessary would not be existent ˙ ˙ because of its essence, and the essentially necessary would not be essentially necessary. This is absurd.

˘

(§ 13) [E] This must be deliberated [critically] for we say: We do not concede that if essence is distinct from existence and both of them exist, essence must have another existence. Why is it not possible that the essence is existent by an existence which is distinct from it while the existence of existence is identical to itself (al-ma¯hiyya mawgˇu¯da bi-l-wugˇu¯d al-mutamayyiz anha¯ wa-yakunu wugˇu¯d al-wugˇu¯d nafs da¯tihı¯)? As to the statement that ‘the intellect splits the essence in ¯ external reality into two things, “essence” and “existence”, and so two forms are there in intellect which correspond to one thing, and that this is something absurd because to one thing cannot correspond two different forms’, and further as to the statement that ‘“common existence” does not take place in concrete beings’, this means that existence is a mental consideration (i tiba¯r dihnı¯) which ¯ is not there in concrete beings. This is not true. Otherwise, the nature of existence would be ‘something relating to non-existence’ in concrete beings. This is absurd. The statement that ‘if existence pertaining to the necessary (al-wugˇu¯d alwa¯gˇibı¯) were connected to some essence this essence would be a universal essence which has intellectual particulars whose relations to it are identical and the essentially necessary would not be existent by its essence’ is not permissible. This would follow only if the relation of this essence to what provides its external individualization (al-musˇahhis al-ha¯rigˇ¯ı) were identical to its relation to ˙ ˘ ˘ ˘ you the intellectual particulars. Why do say that this is like this? This would need to be demonstrated. ˘

(§ 14) The truth is what the Sˇayh [Ibn Sı¯na¯] teaches, i. e. that existence is shared among existents, that existence˘ in contingent beings is additional to their essences, and that the existence of the essentially necessary is not connected to some essence, as we have established. The Ima¯m [Fahr al-Dı¯n al-Ra¯zı¯] disagrees with him regarding his abstraction from essence and˘claims that it is connected to it and that this essence is a cause for it and that it does not precede it by existence. (III) Al-Abharı¯, The Outmost Aim of Thoughts on Clarifying Secrets (Muntaha¯ alafka¯r fı¯ iba¯nat al-asra¯r, MS Tehran, Magˇlis-i Sˇu¯ra¯-yi Millı¯ 2752): (§ 15) la¯mi (1): On existence. ˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Heidrun Eichner

143

Thirteenth-Century Perspectives in Arabic-Islamic Philosophy and Theology

baht (1): It is widespread (masˇhu¯r) that existence is common between all ˙ ¯ They argue for this by several aspects: existents. wagˇh (1): The intellect divides existence into ‘necessary’ and ‘contingent’. That on which the division is carried out (mawrid al-taqsı¯m) must be common between the parts. wagˇh (2): ‘Non-existence insofar as it is non-existence’ is one referent. Therefore, its annihilation (raf ) is one single referent. ‘Annihilation of nonexistence’ is ‘existence’. Hence, existence is one referent. wagˇh (3): If the judgement of the existence of an existent is there ( inda husu¯l al- gˇazm bi-wugˇu¯d mawgˇu¯d), and if we are believed that ‘it is necessary’, ˙then ˙ this belief ceases by the belief that ‘it is contingent’. The same [is the case] if we are believed that ‘it is substance’. This ceases by our belief that ‘it is accident’, but the belief ‘that it is existent’ does not cease. If existence were not common between all parts, then the belief of the existence would cease when the belief of the specifics ceases. All this is weak. ˘

˘

(§ 16) As to (1): We do not concede that the intellect divides existence into ‘necessary’ and ‘contingent’. Rather, the intellect passes a judgement that every existing essence is either ‘essentially necessary’ (wa¯gˇiba bi-l-da¯t) or ‘essentially contingent’. This does not point to existence being common¯ between the two things (al-amrayn). As to (2): We do not concede that ‘non-existence insofar as it is nonexistence’ is one referent. Who is believed that the existence of every thing is identical to its essence is believed that non-existence has various referents. Even if we were to concede this, it is not possible that the annihilation of nonexistence as such (nafsuhu¯) is existence as such. Otherwise, all existents would be realized when there is one existence. This would be necessary because existence as such is realized which is the negation (raf ) of non-existence as such which necessitates (mustalzim) the realization of all existents. As to (3): We do not concede that, if existence is not common, then the belief of all existences would cease if the belief of the specifics ceases. This follows only if predication of existence over the existents is not by ‘sharing in expression’ (isˇtira¯k lafz¯ı). If it is like this, then the belief of one of the referents ˙ of existence would remain there (ha¯sil) when the belief of one of the specifics ˙ ˙the belief of existence ceases by the fact that ceases. Hence it does not follow that the belief of specifics ceases. The truth is that in concrete beings (fı¯ al-a ya¯n) existence is not one common nature between ‘the necessary’ and ‘the contingent’. Otherwise, it would either be ‘essentially necessary’ or ‘essentially contingent’. The first is absurd, because – if its ipseity were because of its essence (da¯t) – its species would be restricted to one individual, and it would ¯ that it is common. If it were because of something else (g˙ayr), then it impossible would need it, and what needs something else is essentially contingent. The ˘

˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

144

˘

second is impossible, otherwise each individual (fard) of it would be essentially contingent because it needs something contingent. Therefore, the existence which is essentially necessary would be essentially contingent. This is absurd. Whether or not that what is conceived (mutasawwar) from each existence is ˙ – this is something which one thing not considering (ma a hadf ) the accidents ¯ ˙ neither of them proves. (§ 17) baht (2): The opinion is widespread that the existence of contingent ˙ ¯ concrete beings is additional to their external essence. They argue beings among that existence is either identical to essence or intrinsic to it or extrinsic to it. The first two [options] are false: The first [option is false], because of several aspects: (1) If we intellect (na qul) blackness while having doubts about its outward existence then ‘what is known’ is different from ‘what is doubted about’. (2) ‘Blackness insofar as it is what it is’ is susceptible of non-existence. ‘Existent blackness’ is not susceptible of non-existence. Hence ‘blackness insofar as it is what it is’ is different from ‘existent blackness’. Therefore, existence is different from essence. (3) Existence is common to all existents. Blackness is not common. Hence existence is different from essence. (4) If existence were identical to blackness, if we say that ‘blackness is existent’ this would have the same status as the claim that ‘blackness is blackness’. This is not the case. ˘

˘

(§ 18) The second [option is false], because if existence were intrinsic to the essences it would be the most general essential [property] (a amm al-da¯tiyya¯t) ¯ which is common to them. So it would be genus for them, and the essentially necessary would be composed out of genus and differentia. This is absurd. Further, because if it were intrinsic, it would be either substance or accident. If it were substance, then a substance would be intrinsic to an accident. If it were accident, then the contrary (bi-l- aks) [would be the case]. If these both possibilities are false, then the third is to be considered true – q.e.d. All this is weak. As to the argument that one can know blackness while doubting about its existence, we say: We do not concede that from this follows that the existence of blackness in concrete beings is different from the external essence. Rather it follows that ‘the referent of blackness in the intellect’ is different from ‘existence’. As to the argument that ‘blackness insofar as it is what it is’ is susceptible of non-existence and ‘the existent blackness’ is not susceptible of non-existence, we say: If you mean by its ‘being susceptible of non-existence’ that it is possible that it can be annihilated (irtifa¯ ) from the outward, then we do not concede that ‘existent blackness’ is not susceptible of non-existence according to this terminology (tafsı¯r). If you mean by it that blackness in the state (ha¯la) of non˙ ˘

˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Heidrun Eichner

Thirteenth-Century Perspectives in Arabic-Islamic Philosophy and Theology

˘

existence is there (ha¯sil) in the outward and attributed with non-existence, this is ˙¯ )˙ because it is impossible that blackness has a ipseity in the impossible (mamnu outward during the state of non-existence. As to the argument that existence is common between all existent things: You already know that it is weak. As to the argument that if existence were identical to blackness, if we say that ‘blackness is existent’ this would have the same status as the claim that ‘blackness is blackness’, we say: If you make the substrates of both judgements (mawdu¯ al-qadiyyatayn) external blackness, we do not concede the difference ˙ both ˙judgements. If you make its substrate the referent of blackness, between from this does not follow that what is aimed at results, because then from this follows their being different in the intellect, not in the outward [reality]. As to the argument that if existence were part of an essence it would be the most general essential [property], we say: We do not concede this. This would only follow if existence were common to all existents, and that this is weak has just been shown. From the assumption (taqdı¯r) of its being common follows its being the most general essential [property] only if it is essential for the essence of the Necessary of Existence. Its being essential only follows if it has an essence beyond existence. Why do you say that this is the case? As to the argument: If it were intrinsic, it would be either substance or accident etc., we say: If you mean by ‘substance’ the essence which, if it is existent among concrete beings, is not in a substrate, and by ‘accident’ ‘existent in a substrate’, then we do not accept this enumeration [to be exhaustive] (hasr), ˙ ˙ because a third part can be realized. This is, that – if it exists in concrete beings – then it is in a substrate. If you mean by ‘substance’ the essence which if it exists in concrete beings then it is in a substrate, then existence is neither a substance nor an accident, because it does not have an essence beyond existence. If you mean by ‘substance’ ‘what is independent (g˙anı¯) from a substrate’ and by ‘accident’ ‘what needs a substrate’ – why is it not possible that according to this definition (tafsı¯r) substance makes the accident subsist (muqawwim) because it is possible that the whole (magˇmu¯ ) needs a substrate while its parts do not need it? The truth is that in concrete things existence is identical to external essence because otherwise existence would either be a part (gˇuz ) of it, or an attribute (sifa) of it. The first is impossible, otherwise the existence of a thing ˙ would precede it because it is necessary that the part precedes the whole (alkull). This is absurd. And also, because, if existence were intrinsic to the external essences, all simple essences would be composed. This is absurd. The second is also absurd because, if existence were an attribute of essence, it would need it (ka¯na muftaqiran ilayha¯). What is needed by something has to have a ipseity which comes before its ipseity. Then, essence would have a ipseity in outward [reality] before existence. This doctrine is what one of the eminent scholars reports from Aristotle (al-mu allim al-awwal). ˘

˘

˘

˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

145

146

(IV) At¯ır al-Dı¯n al-Abharı¯, The Book of Doubts (Kita¯b al-Sˇuku¯k, MS Istanbul, ¯ 2319): Ayasofya

˘

(§ 19) He says: ‘If we know that a contingent being has a cause we judge that this cause is existent etc.’ We say: Why do you say that if existence were not shared this belief would cease? This, because we have a belief of some existence which is concretised in itself (wugˇu¯d mu ayyan fı¯ nafsihı¯) whose self is unknown (magˇhu¯l nafsahu¯). According to us, the ‘concretised existence in itself ’ is identical to the essence of this existence (nafs ma¯hiyyat da¯lika al-wugˇu¯d). Why ¯ of an existence which do you say that if existence were not shared, then the belief is concretised in itself would cease because ‘considering other specifics’ ceases? Don’t you see that if we know one of the referents of the expression ayn [source, eye, concrete being] is existence, we are convinced of some concretised ayn as such? The belief of the specifics of its concretised referent leads to the disappearance of all others. Yet, given this (ma a ha¯da¯), why is it necessary that ¯ referents? the referent of ayn is shared by the remainder of its He says: ‘The referent of negation is one referent insofar as it is negation’. We say: We do not concede this, but according to us, negation (salb) is a shared expression (lafz musˇtarak) which is predicated of different referents in a way like existence, too,˙ is a shared expression. He says: ‘It is possible to intellect a triangle while doubting its external existence’. We say: This points to the fact that external existence is different from the essence which is intellected from the triangle. But why does it follow from this that external existence is different from the external essence? We concede that external existence is different from the intellected essence and also from essence itself (nafs al-ma¯hiyya). We say that external existence is in (fı¯) the external essence. What you have mentioned does not point to this. We say the same about the remaining aspects which point towards existence being not identical to essence. Our answer is that we do not claim that existence is different from outward essence. Rather we claim that outward existence is different from the essence itself insofar as it is what it is, according to what we have mentioned. We say that existence is identical to the outward essence because, if it were additional to it in the outward, the essence would have a ipseity in the outward other than the ipseity of existence in the outward. Both personalities would be following each other (mutala¯zima¯ni) in the outward, and each of the two things following each other would become substratum for the other. So it would be necessary that one would be the cause ( illa) for the other. Otherwise, there would not be resulting from them one composed thing. If existence were the cause for essence it would be preceding essence, and then existence which is accidental for something would be preceding this thing. This is absurd. If essence were the cause for existence the existence of a contingent being would be caused by the essence. The cause must precede the caused by ˘

˘

˘

˘

˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Heidrun Eichner

Thirteenth-Century Perspectives in Arabic-Islamic Philosophy and Theology

existence. Therefore, the existence of essence before its existence would follow. This is absurd. What results from the teachings (madhab) of the philosophers is ¯ that external existence is different from the essence itself, and that it is identical ( ayn) to the external essences. Both are different in the intellect. Many things are united in the outward and different in the intellect, as the author himself admits in the case of the difference between mental and external composition. If someone were to say: What is known from the teachings of the philosophers is that all existents share in existence. If existence were identical to the external essence, how can it be shared? We answer: What is meant when they say that existence is shared between all existing things is that in intellect they share in existence and the general (al- a¯mm) which has no occurence (wuqu¯ ) in concrete beings.

˘

˘

˘

˘

ˇ a¯mi al-daqa¯ iq (MS Paris, Biblioth(V) Nagˇm al-Dı¯n al-Ka¯tibı¯ al-Qazwı¯nı¯, G que nationale 2370): ˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

147

(§ 20) fasl (2): On existence of contingent beings being not identical to the ˙ part of it. essence nor We conceptualize a triangle while doubting about its external existence. So, in this state, one passes the judgement about the triangle in the intellect that it is a triangle, but one does not pass the judgement that it is existent in outward [reality]. If its external existence were identical to its being a triangle or part of it, then the judgement would be impossible that it is a triangle without the judgement that it is existent in outward [reality]. Therefore, existence is not identical to the triangle and not part of it. Likewise it is possible for the remainder of essences to intellect them while neglecting their external existence. So, existence is not identical to contingent essences, nor is it part of them. (§ 21) If someone were to say: ‘If existence were not identical to the contingent essence, the external essence would be attributed with it, and existence would be an attribute of it. An attribute needs “something which is attributed”, and what is needed must have priority. Hence, the essence must have priority by [its] existence before existence and then it would have another existence whose existence must have priority over this existence, too. Thus, between essence and existence would be an infinite number of existences, and something infinite would be included between two ends, and this is absurd.’ We say: ‘We do not concede that if existence were not identical to essence, then the external essence would be attributed with existence in outward [reality]. This would follow only if the external essence were different from existence. Why do you say that it follows from existence being different from “essence as such” (nafs al-ma¯hiyya) that it is different from the “external essence”?’

148 ˘

This is so because in outward [reality] essence and existence are one thing (sˇay wa¯hid), and when they are in intellect, intellect splits (fassala) them into ˙ ‘essence’ and ‘existence’. ˙˙ two things: If you say: ‘If it is affirmed that existence is different from essence in intellect it follows that they are different in the outward [reality], too. Otherwise, the judgement of intellect that they are different would not correspond to what actually is the case (hukm al- aql bi-l-mug˙a¯yara g˙ayr muta¯biq ˙ ˙ li-ma¯ fı¯ nafs al-amr).’ We say: ‘We do not concede this because intellect passes the judgement about them that they are different in intellect and united in outward [reality]. That this judgement corresponds to what actually is the case rests only on their being different in intellect, but not in outward [reality]’. If you say: ‘If they are different in intellect and unified in outward [reality] from one thing in outward [reality] would result two representations (mita¯la¯n) in intellect, one of them being “essence”, one of them being “existence”’.¯ We say: ‘Why is it not possible that from one thing in outward [reality] result two representations in intellect? This, because from an isosceles triangle result two representations in intellect, one of them being the representation of “triangle absolutely”, and one of them being the representation of “isosceles triangle”. In a similar vein, from “blackness” result two representations in intellect while it is one thing in outward [reality] which does not have two aspects so that to one of them would correspond “triangle” and to the other “isosceles triangle”. Likewise, from blackness result two representations in intellect, one of them being the representation of “colour-ness absolutely” (allawniyya al-mutlaqa) and the other being the representation of “blackness-ness” (al-sawa¯diyya) ˙while in outward [reality] they are one thing. There are many similar things which will follow for you – God willing so.’ ˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Heidrun Eichner

(§ 22) fasl (3): The existence of the essentially necessary is not different from its ˙ essence, neither in external [reality] nor in the intellect. In external [reality] this is so because if its existence were different from its essence it would either be part of it or extrinsic to it. Both options are false. The first, because if it were part of it its essence would be composed and everything which is composed is in need of its parts which are something other than it. What needs something else is essentially contingent. Then, the essentially necessary would be essentially contingent. This is absurd. The second [is false] because if it were extrinsic to it, it would be an attribute which subsists in an essence. Then it would need it, and what needs something else is essentially contingent. Hence this existence insofar as it is existence would be essentially contingent and it would have a cause. If its cause were not this essence, then the essentially necessary would in its existence be in need of something else. This is absurd. If it were identical to this essence (and the cause must precede by

Thirteenth-Century Perspectives in Arabic-Islamic Philosophy and Theology

existence what is caused), then this essence precedes existence by existence. Then it would have another existence before existence and it follows that it precedes by existence this existence, too. Thus it follows that between the essence and its existence are infinite existences. This is absurd. In intellect [this is absurd] for if intellect were to split the essentially necessary into essence and existence it would have a universal essence whose very referent would not prevent that it has infinite particulars. Then no one of these particulars would exist by this essence itself because if some particulars were to exist because of the essence itself but others not, this would be a preponderance without an agent (targˇ¯ıh bi-la¯ muragˇgˇih). If no one of its particulars were to exist ˙ existence of˙ the particular which is realized among by this essence itself, the concrete beings would be there not because of the essence itself but because of a distinct cause. Then, the Necessary of Existence would be existent because of a separate cause. This is absurd. So, in the intellect the essentially necessary is not split into two things, essence and existence. Rather its existence is identical to its essence both in the intellect and in external [reality]. If it were said that if the existence of the Necessary of Existence were not different from Its essence in concrete beings then its abstraction either would be because of the nature of existence and its concomitants or because of a distinct cause. The second is false because if its abstraction were due to a cause which is distinct from the nature of existence and its concomitants, this would be due to a cause which is something else than the essence of the Necessary of Existence and its concomitants. Then, the Essentially Necessary in its abstraction would be in need of something else. This is absurd. The first is also false because if its abstraction were because of the nature of existence, anything which is there among concrete beings would be abstract. The consequent is not true, and so neither is the premise. We say: We do not concede that if its abstraction were because of the nature of existence everything which is there among concrete beings would be permanently abstracted from existence. This would follow only if existence among concrete beings were one thing. Why do you say that it is like this? This, because existence of the contingents among concrete beings (wugˇu¯d al-mumkina¯t fı¯ al-a ya¯n) is identical to the essence. It [i.e., this essence] is different from the existence pertaining to the Necessary [of Existence] absolutely (al-wugˇu¯d al-wa¯gˇibı¯ al-mugˇarrad). What results among concrete beings from the singular instantiations (afra¯d) is existence which is predicated of the existences which result in intellect and are individualized in the Necessary of Existence. The existence of contingents does not have a ipseity among concrete beings, but what results (al-ha¯sil) among concrete beings are their essences. If essences which ˙ ¯ hiyya¯t al- ayniyya) are there (hasala) in intellect, intellect are concretised ˙(al-ma ˙ ˙ splits them into ‘essence’ and ‘existence’. Then the personalities of the existences result only in mind (fa-l-wugˇu¯da¯t innama¯ tahsulu huwiyya¯tuha¯ fı¯ al-adha¯n). ¯ ˙˙ ˘

˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

149

150

˘

What is meant when the philosophers say that existence is shared among ‘the necessary’ and ‘the contingent’ is that if contingent essences are there in intellect and if intellect splits them into ‘essence’ and ‘existence’, – if one considers their existence – there is a sharing with the existence of the Necessary [of existence] (al-wugˇu¯d al-wa¯gˇibı¯) in what is called ‘existence’ (fı¯ musamma¯ al-wugˇu¯d). This is ‘common existence’ (al-wugˇu¯d al- a¯mm) which takes place only in mind (la¯ wuqu¯ a lahu¯ illa¯ fı¯ al-dihn). ¯ By what we mention is refuted what our Ima¯m and the Ima¯m of all researchers, the recreation of the souls of those who seek for knowledge, says against the philosophers (al-hukama¯ ) in the Kita¯b al-Mulahhas where he says: ˙ nature of the ˙ They agree that it is impossible that some individuals of˘ ˘the species (al-tabı¯ a al-naw iyya) are abstracted from matter and others are connected. ˙So, existence, too, is one nature. If it is not in need of connection with an essence then it is like that absolutely. If it is possible that it is sometimes abstract and sometimes connected–why is it not possible in one nature of a species that it is sometimes abstract from matter, and sometimes connected? Because we say: Why do you say that existence is not independent from being connected to an essence absolutely? It is only then not independent if some of the existences are there among concrete beings as inherent (ha¯ll) in an essence. ˙ This is not the case. The existence of contingent beings takes place only in mind, not in concrete beings. The existence of the Necessary of Existence is abstract from essence. The result is: Existence in concrete beings is identical to the concretised essence (al-ma¯hiyya al- ayniyya), and it is not identical to the essence in intellect. Existence which is predicated of the Necessary of Existence has among concrete beings only the form of the existence of the Necessary [of Existence]. Intellect splits the contingent which takes place in concrete beings into ‘essence’ and ‘existence’ after this. The individual instantiation (ifra¯d) of existence takes place in intellect, and intellect divides the form of absolute existence into existence of the Necessary [of Existence] and into these existences which are assumed (mutaqaddira) in intellect. That on which the division is carried out must be shared among the parts. So, existence is shared between the individual instantiations (afra¯d) of the existents which are there in intellect. ˘

˘

˘

˘

˘

Bibliography al-Abharı¯, al-Mufaddal b. ‘Umar At¯ır al-Dı¯n, Kasˇf al-haqa¯ iq, MS Tehran, Magˇlis-i ˇsu¯ra¯˙ ˙ 1 – 212. ¯ yi millı¯ 2752, ˙pp. –– , Mara¯sid al-maqa¯sid, MS Istanbul, Serez 1963. ˙ ¯ al-afka ˙¯ r fı¯ iba¯nat al-asra¯r, MS Tehran, Magˇlis-i Sˇu¯ra¯-yi Millı¯ 2752, –– , Muntaha pp. 213 – 358. ˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Heidrun Eichner

Thirteenth-Century Perspectives in Arabic-Islamic Philosophy and Theology

˘

˘

˘

˘

˘

–– , R. al-masa¯ il, MS Istanbul, Ragıp Pas¸a 1461. H. Eichner, Dissolving the Unity of Metaphysics: From Fakhr al-Din al-Razi to Mulla Sadra al-Shirazi, Medioevo 32, 2007, pp. 139 – 97. –– , ‘Knowledge by Presence’, Apperception and the Mind-Body Relationship: alSuhrawardı¯ and Fakhr al-Dı¯n al-Ra¯zı¯ as Representatives of a 13th Century Discussion, in In the Age of Averroes, ed. P. Adamson, London: The Warburg Institute, 2011, pp. 117–40. –– , The Post-Avicennian Philosophical Tradition and Islamic Orthodoxy: Philosophical and Theological Summae in Context, unpublished professorial dissertation, submitted Halle 3/2009. F. Griffel, On Fakhr al-Dı¯n al-Ra¯zı¯’s Life and the Patronage he Received, Journal of Islamic Studies 18, 2007, pp. 313 – 44. T. Mayer, Ibn Sı¯na¯’s Burha¯n al-Siddiqı¯n, Journal of Islamic Studies 12, 2001, pp. 18 – 39. ˇ a¯mi al-daqa¯ iq, MS Paris, al-Ka¯tibı¯ al-Qazwı¯nı¯, Alı¯ b. ˙ Umar Nag˘m al-Dı¯n, G Bibliothque nationale 2370. –– , Muta¯raha¯t falsafiyya bayna Nas¯ır al-Dı¯n al-Tu¯sı¯ wa-Nagˇm al-Dı¯n al-Ka¯tibı¯, ed. ˙ ˙ Hasan A¯l-Yası¯n, Baghdad: ˙ Muh˙ammad Maktabat al-ma a¯rif, 1956. ˙ b. Umar Abu¯ Abd Alla¯h Ibn al-Hat¯ıb Fahr al-Dı¯n, al-Maba¯hit alal-Ra¯zı¯, ˙Muh. ammad ˙¯b¯ al˘ ˙ ˙ da-dı¯˘. Beirut: Da¯r al-kita masˇriqiyya, ed. Muh. ammad al-Mu‘tas. im bi-lla-h al-Bag arabı¯, 1990. –– , al-Mulahhas fı¯ al-hikma, MS Berlin Or. Oct. 623. ˙ ammad ˙ ˘¯˘, Muh al-Samarqandı b. Asˇraf al-Hasanı¯ Sˇams al-Dı¯n, al-Saha¯ if al-ila¯hiyya, ed. ˙ ˙ ˙ ˙ 1985. Ahmad Abd al-Rahma¯n al-Sˇarı¯f, Kuwait, ˙ ˙ A. Shihadeh, The Teleological Ethics of Fakhr al-Dı¯n al-Ra¯zı¯, Leiden: Brill, 2006. ˘

˘

˘

˘

˘

˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

151

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Avicenna’s Metaphysics in the Medieval Hebrew Philosophical Tradition A Short Historical Sketch of its Evident Traces

Mauro Zonta Avicenna’s metaphysics and, in particular, Avicennian works in which metaphysical themes are studied in detail had a limited and mostly indirect impact on medieval Jewish philosophy.1 Previous studies on this subject have shown that some major Jewish philosophers probably knew and hinted at various points of Avicenna’s metaphysical thought without mentioning the author’s name. As a matter of fact, they might have not read Avicenna directly, ˙ aza¯lı¯’s The Intentions of the Philosophers but through the intermediation of al-G (Maqa¯sid al-fala¯sifa), which was very well known by late medieval Jewish ˙ 2 philosophers. This might be suggested, e. g., from an examination of Abraham Ibn Daud’s The Exhalted Faith (Kita¯b al-‘aqı¯da al-rafı¯‘a), whose original Arabic text, written in Spain in 1161, is not extant, but was twice translated into Hebrew in the second half of the fourteenth century.3 Moses Maimonides, too, certainly knew and employed some aspects of Avicenna’s metaphysics, though ˙ aza¯lı¯’s work.4 Later on, in 1354, the Spanish possibly through a perusal of al-G ˙ aza¯lı¯’s work as the main Jewish philosopher Moses ben Judah employed al-G source for the structure and contents of his encyclopedia, Love for Pleasures

1

2

3 4

See Zonta, Linee del pensiero islamico nella storia della filosofia ebraica medievale, pp. 450 – 62. This particular fact does not imply that Avicenna’s philosophical thought as a whole had no impact on late medieval Jewish philosophy; on the contrary, as observed there (p. 462), ‘il quadro storico che abbiamo cercato di tracciare induce a pensare che, diversamente da quanto generalmente si crede, nel quadro delle influenze islamiche sulla filosofia ebraica medievale, ad Avicenna spetti un ruolo pressoch pari a quello di Averro’. See also the historical sketches in Zonta, The Role of Avicenna and of Islamic ‘Avicennism’; Zonta, Avicenna in Medieval Jewish Philosophy. This fact was apparently not limited to physics (as pointed out by Harvey, Why Did Fourteenth-Century Jews Turn to Alghazali’s Account of Natural Science?), but probably ˙ aza¯lı¯’s work was translated into Hebrew involved metaphysics too. The whole text of al-G three times in the first half of the 14th century. See in particular ‘Eran, Me-’emunah tame-ah le-’emunah ramah, pp. 27 and 76. See the conclusion suggested in Zonta, Maimonides’ Knowledge of Avicenna, pp. 221 – 2.

154 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Mauro Zonta

(Ahavah ba-ta‘anugim); by this way, he shows to follow many aspects of Avicenna’s metaphysical doctrines.5 There are really very few clear traces of a direct and acknowledged use of Avicenna’s metaphysics, and of that of The Cure (al-Sˇifa¯’) in particular, as a source.6 In this short historical sketch, I will give a general summary of the results of research on this subject, according to the chronological order of the Jewish authors. Probably the first such author who evinces a direct knowledge of the contents of Avicenna’s metaphysics was Moses ha-Levi (Mu¯sa¯ al-La¯wı¯), a neglected thirteenth-century Spanish Jewish philosopher. In his major extant philosophical treatise devoted to metaphysical and theological matters, the Divine Treatise (Maqa¯la ’ila¯hiyya), he displays a familiarity with Avicenna’s thought about those subjects, as Georges Vajda has shown.7 However, a number of direct quotations of Avicenna’s metaphysics are found in some of the works of another Jewish author, probably living in the same time and milieu of Moses ha-Levi: Shem Tov Ibn Falaquera (1225 – 95 ca.). In those quotations, Ibn Falaquera openly admits his dependence on Avicenna, although the direct sources of these quotations cannot be identified in all cases. I have listed elsewhere Avicenna’s philosophical works quoted by Ibn Falaquera.8 The most interesting cases are the passages of the metaphysical section of The Cure found in one of Ibn Falaquera’s main philosophical works, his commentary on Maimonides’s Guide of the Perplexed: The Guide to the Guide (Moreh ha-Moreh). Most of them were first identified by the Israeli scholar Ya’ir Shiffman, who has recently published a critical edition of Ibn Falaquera’s work.9 In the Appendix here below, I will give a list of them. As pointed out by Shiffman, another important source of Avicenna’s metaphysics in Ibn Falaquera’s Guide to the Guide is the metaphysical section of The Salvation (al-Nag˘a¯t). 10 However, there was a partial Hebrew translation of this text, made by the ProvenÅal Jewish philosopher Todros ben Meshullam ben David Todrosi of Arles only later, in the period 1330 – 40. This translation, 5 6 7 8 9

See Zonta, The Role of Avicenna and of Islamic Avicennism, pp. 657 – 9. See Zonta, La tradizione ebraica medievale dei testi arabi, pp. 553 – 4. See Vajda, Un champion de l’avicennisme. Zonta, Hebrew Transmission of Arabic Philosophy and Science, pp. 128 – 9. Shem Tov ben Joseph Ibn Falaquera, Moreh ha-Moreh; about the influence of Avicenna’s metaphysics on Ibn Falaquera’s Guide to the Guide, see in particular Shiffman’s introduction to his edition of this work, on pp. 56 – 8. 10 See the list of quotations of Avicenna’s works in Shem Tov ben Joseph Ibn Falaquera, Moreh ha-Moreh, p. 381. As pointed out by Shiffman, in some cases Ibn Falaquera’s Avicennian quotations seem to result from a mixture of passages of The Cure with the corresponding passages of The Salvation: see, e. g., Shiffman’s notes to his edition on pp. 238 and 291. About this fact, see Bertolacci, The Reception of Aristotle’s ‘Metaphysics’ in Avicenna’s Kita¯b al-Sˇifa¯’, p. 587.

Avicenna’s Metaphysics in the Medieval Hebrew Philosophical Tradition 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

155

bearing the title The Soul’s Salvation (Hasalat ha-nefesˇ ) and being still ˙ 11 it includes the complete unpublished, is found in two handwritten copies; physical section and some of the first chapters of the metaphysical section of Avicenna’s Salvation,12 so showing the interest of the author, one of the main representatives of a sort of fourteenth-century ‘Jewish Avicennism’, for Avicenna’s natural and metaphysical philosophy in general, and for his theological doctrines in particular. In the same period (second quarter of the fourteenth century), some explicit quotations of Avicenna, allegedly drawn from the lost metaphysical section of an Avicennian philosophical work, The Oriental Philosophy (al-Hikma al-masˇri˙ qiyya), can be found in four Medieval Hebrew texts by a Castilian Jewish (and 13 subsequently Christian) author, Avner of Burgos. Finally, it should be pointed out that at least one literal quotation of the metaphysical section of The Cure has been found by Vajda in a mid-fourteenth-century work by another Spanish Jewish philosopher, Joseph Ibn Waqqar: The Treatise Reconciling Philosophy and Religious Law (al-Maqa¯la al-g˘amı¯‘a bayna al-falsafa wa-l-sˇarı¯‘a).14 Ibn Waqqar’s apparent ‘Avicennism’ in this theological-philosophical work might justify the explicit employment of Avicenna’s metaphysics as one of its sources. From the above data, it seems that there were two phases of the reception of Avicenna’s metaphysics among Medieval Jewish philosophers. The former is found in the twelfth-century Jewish Aristotelians like Abraham Ibn Daud and Maimonides, and in some of their later followers, where Avicenna’s metaphys˙ aza¯lı¯’s ical doctrines were apparently transmitted in an indirect way, through al-G 11 These copies are found in fols 208r–301v of the manuscript of London, British Library, Add. 27559 (Margoliouth 890), probably written by a Sephardic hand in the 15th century; and in fols 87r–159v of the manuscript of Paris, Bibliothque National de France, hbreu 1023, probably written by a Sephardic hand in the 15th-16th centuries. The former was the only copy known to Moritz Steinschneider: see Steinschneider, Die hebrischen bersetzungen, p. 285. A microfilm copy of both manuscripts is found in the Institute of Microfilmed Hebrew Manuscripts of the National Library of Israel in Jerusalem, under the signatures F 6094 and F 15717; in that place, there is also a CD copy of the former, under the signature CD 177. See now Berzin, The Medieval Hebrew Version. 12 The first chapters of the metaphysical section of The Salvation are found in fols 277v–301v of the London manuscript, and in fols 141r–159v of the Paris manuscript. 13 For an identification, edition and English translation of these quotations, see here below, Zonta, Possible Hebrew Quotations of the Metaphysical Section of Avicenna’s Oriental Philosophy and Their Historical Meaning. 14 See Vajda, Recherches sur la philosophie et la kabbale, pp. 115 – 297, and p. 132 in particular, where Ibn Waqqar’s quotation of Avicenna’s Cure, Metaphysics, treatise 9, c. 6 (on providence), is mentioned. Ibn Waqqar probably knew the contents of Avicenna’s metaphysics through Fahr al-Dı¯n al-Ra¯zı¯’s commentary on Avicenna’s Remarks and ˘ Admonitions (al-Isˇa¯ra¯t wa-l-tanbı ¯ha¯t) too, The Cores of the Remarks (Luba¯b al-isˇa¯ra¯t): see Vajda, Recherches sur la philosophie et la kabbale, p. 125 – 6.

156 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Mauro Zonta

interpretation of them in particular. The latter is found in some thirteenth- and fourteenth-centuries Jewish authors, most of whom are Spanish: they show to have read Avicenna’s metaphysics and often admit their dependence on it. Among them, a paramount role was played by Shem Tov Ibn Falaquera, as it results from the detailed examination of some of his works.

Appendix Quotations of the Metaphysics of Avicenna’s Cure (al-Sˇifa¯’) in Shem Tov Ibn Falaquera’s Guide to the Guide (Moreh ha-Moreh)15 * Shem Tov ben Joseph Ibn Falaquera, Moreh ha-Moreh, p. 136, lines 50 – 53 (on GP, I, 34) = Avicenna, Metaphysics, p. 14, line 36–p. 15, line 4 of the English translation; p. 14, line 19–p. 15, line 2 of the Arabic text (about the objects of mathematical sciences and their relationship to the contents of metaphysics). Shem Tov ben Joseph Ibn Falaquera, Moreh ha-Moreh, p. 162, lines 57 – 8 (on GP, I, 58) = Avicenna, Metaphysics, p. 296, lines 14 – 17 of the English translation; p. 296, lines 9 – 10 of the Arabic text (about the definition of God as ‘intellect, and subject and object of intellection’). * Shem Tov ben Joseph Ibn Falaquera, Moreh ha-Moreh, p. 220, lines 249 – 55 (on GP, II, Introduction) = Avicenna, Metaphysics, p. 30, lines 7 – 8, 19 – 26 of the English translation; p. 30, lines 4, 11 – 14 of the Arabic text (on the nature of the Necessary Existent, i. e. God). Shem Tov ben Joseph Ibn Falaquera, Moreh ha-Moreh, p. 223, lines 320 – 23 (on GP, II, Introduction) = Avicenna, Metaphysics, p. 131, line 10–p. 132, line 8 of the English translation; p. 131, line 6–p. 132, line 4 of the Arabic text (summarized; about the philosophical meaning of the Arabic term qu¯wa, ‘potency’). Shem Tov ben Joseph Ibn Falaquera, Moreh ha-Moreh, p. 237, lines 143 – 53 (on GP, II, 4) = Avicenna, Metaphysics, p. 311, line 27–p. 312, line 12 of the English translation; p. 311, line 15–p. 312, line 7 of the Arabic text (partially altered; about the characteristic of the soul, which is the proximate mover of the heavens). Shem Tov ben Joseph Ibn Falaquera, Moreh ha-Moreh, p. 238, lines 159 – 62 (on GP, II, 4) = Avicenna, Metaphysics, p. 314, lines 11 – 17 of the English translation; p. 314, lines 5 – 8 of the Arabic text (about the characteristic of the infinite power which moves the heavens). 15 The quotations that escaped Shiffman’s notice are signed as *; GP is Maimonides’s Guide of the Perplexed.

Avicenna’s Metaphysics in the Medieval Hebrew Philosophical Tradition 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

157

Shem Tov ben Joseph Ibn Falaquera, Moreh ha-Moreh, p. 243, lines 17 – 26 (on GP, II, 6) = Avicenna, Metaphysics, p. 243, line 29–p. 244, line 10 of the English translation; p. 243, line 15–p. 244, line 5 of the Arabic text (about the opinion that two things should exist in one thing).16 Shem Tov ben Joseph Ibn Falaquera, Moreh ha-Moreh, p. 289, line 32-p. 290, line 44 (on GP, II, 40) = Avicenna, Metaphysics, p. 364, line 27–p. 365, line 25 of the English translation; p. 364, line 16–p. 365, line 13 of the Arabic text (about the necessity of a law, given by a prophet).17 Shem Tov ben Joseph Ibn Falaquera, Moreh ha-Moreh, p. 291, lines 11 – 16 (on GP, II, 48) = Avicenna, Metaphysics, p. 362, line 28–p. 363, line 2 of the English translation; p. 362, line 15–p. 363, line 1 of the Arabic text (about the fact that the principles of all matters descend from God). Shem Tov ben Joseph Ibn Falaquera, Moreh ha-Moreh, p. 316, lines 9 – 11 (on GP, III, 19) = Avicenna, Metaphysics, p. 288, lines 3 – 6 of the English translation; p. 288, lines 2 – 3 of the Arabic text (about God’s omniscience). Shem Tov ben Joseph Ibn Falaquera, Moreh ha-Moreh, p. 334, line 133–p. 335, line 150 (Appendix 1) = Avicenna, Metaphysics, p. 350, line 15–p. 351, line 26 of the English translation; p. 350, line 8–p. 351, line 13 of the Arabic text (about the perfection of the man’s natural soul).18

Bibliography Avicenna, al-Nag˘a¯t fı¯ al-mantiq wa-l-’ila¯hiyya¯t ta’lı¯f li-sˇayh al-ra’ı¯s … Ibn Sı¯na¯, ed. ˘ ‘A. ‘Amı¯ra, Beirut: Wa¯lid ˙al-mag˘¯ıd, 1992/1412. –– , The Metaphysics of ‘The Healing’, transl. and ed. M. Marmura, Provo, Utah: Brigham Young University Press, 2005. A. Bertolacci, The Reception of Aristotle’s ‘Metaphysics’ in Avicenna’s ‘Kita¯b al-Sˇifa¯’’. A Milestone of Western Metaphysical Thought, Leiden/Boston: Brill, 2006. G. Berzin, The Medieval Hebrew Version of Psychology in Avicenna’s Salvation (Al-Naja¯t), Ph.D. Thesis, Harvard University, 2010.

16 In this case too, the first sentence of Avicenna’s text is not entirely identical with that translated by Ibn Falaquera. Avicenna writes: fa-zanna qawmun ’anna al-qismata tu¯g˘ibu ˙ wug˘u¯da ˇsay’ayni fı¯ kulli ˇsay’in, ‘(One) group thought that the division necessitates the existence of two things in each thing’. Ibn Falaquera seems to translate here a partially different text, as follows: fa-zanna qawmun ’anna al-’ila¯ha yu¯g˘ibu wug˘u¯da ˇsay’ayni fı¯ kulli ˇsay’in, ‘(One) group thought˙ that God necessitates the existence of two things in each thing’. 17 It should be pointed out that an identical passage is found in al-Nag˘a¯t fı¯ al-mantiq wa-l˙ ’ila¯hiyya¯t … li- … Ibn Sı¯na¯, vol. II, p. 165, line 24-p. 166, line 16. 18 An identical passage is found in al-Nag˘a¯t fı¯ al-mantiq wa-l-’ila¯hiyya¯t … li- … Ibn Sı¯na¯, ˙ vol. II, p. 153, line 3–p. 155, line 6.

158 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Mauro Zonta

E. Eisenmann, Ahabah ba-Ta‘anugim: A Fourteenth-Century Maimonidean Encyclopedia, in Traditions of Maimonideanism, ed. C. Fraenkel, Leiden: Brill, 2009, pp. 213 – 22. A. ‘Eran, Me-’emunah tame-ah le-’emunah ramah. Haguto ha-qdam-Maymonit ˇsel haRabad (From Simple Faith to Sublime Faith. Ibn Daud’s Pre-Maimonidean Thought), Tel Aviv: ha-Qibbus ha-me’uhad, 1998. ˙ ˙ S. Harvey, Why Did Fourteenth-Century Jews Turn to Alghazali’s Account of Natural Science?, Jewish Quarterly Review 91, 2001, pp. 359 – 76. Shem Tov ben Joseph Ibn Falaquera, Moreh ha-Moreh, ed. Y. Shiffman, Jerusalem: The World Union of Jewish Studies, 2001. M. Steinschneider, Die hebrischen bersetzungen des Mittelalters und die Juden als Dolmetscher, 2 vols, Berlin: Kommissionsverlag des Bibliographischen Bureaus, 1893. G. Vajda, Recherches sur la philosophie et la kabbale dans la pense juive du Moyen ffge, Paris/Le Haye: Mouton & Co., 1962. –– , Un champion de l’avicennisme. Le problme de l’identit de Dieu et du premier moteur d’aprs un opuscule judo-arabe indit du XIIIe sicle, Revue tomiste 56, 1948, pp. 480 – 508. M. Zonta, Avicenna in Medieval Jewish Philosophy, in Avicenna and His Heritage, eds J. Janssens and D. De Smet, Leuven: Leuven University Press, 2002, pp. 267 – 79. –– , Hebrew Transmission of Arabic Philosophy and Science: A Reconstruction of Shem Tov Ibn Falaquera’s Arabic Library, in L’interculturalit dell’Ebraismo, ed. M. Perani, Ravenna: Longo Editore, 2004, pp. 121 – 37. –– , La tradizione ebraica medievale dei testi arabi, in Lo spazio letterario del Medioevo. 3. Le culture circostanti, vol. II, eds M. Capaldo, F. Cardini, G. Cavallo and B. Scarcia Amoretti, Rome: Salerno Editrice, 2003, pp. 531 – 67. –– , Linee del pensiero islamico nella storia della filosofia ebraica medievale, Annali dell’Istituto Universitario Orientale di Napoli, 57, 1997, pp. 101 – 44, 450 – 83. –– , Maimonides’ Knowledge of Avicenna. Some Tentative Conclusions About a Debated Question, in The Trias of Maimonides/Die Trias des Maimonides. Jewish, Arabic, and Ancient Culture of Knowledge/Jdische, arabische und antike Wissenkultur, ed. G. Tamer, Berlin/New York: Walter de Gruyter, 2005, pp. 211 – 22. –– , The Role of Avicenna and of Islamic ‘Avicennism’ in the 14th-Century Jewish Debate Around Philosophy and Religion, Oriente moderno, 80, n.s. 19, 2000, pp. 647 – 60.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

‘Happy is he whose children are boys’: Abraham Ibn Daud and Avicenna on Evil Resianne Fontaine 1. Introduction The title of this paper is the first part of a quotation from the Babylonian Talmud, which reads: ‘Happy is he whose children are boys, unhappy is he whose children are girls.’1 As we shall see, the twelfth-century philosopher Abraham Ibn Daud, the main protagonist of my paper, adduces this Rabbinic statement in his treatment of the problem of evil. Ibn Daud’s approach to the problem of evil differs markedly from that of other pre-Maimonidean Jewish thinkers, and this is mainly due to his use of Avicenna. In order to explain this, I will first provide some background information on Ibn Daud and then present a brief outline of Avicenna’s discussion of evil. Following this I will turn to the main topic of this paper: the reception of Avicenna’s views on evil by Ibn Daud. Born in Andalusia somewhere around 1110, Abraham Ibn Daud received an education both in religious learning and ‘Greek wisdom’, that is, non-Jewish science and philosophy. Later in life, and presumably in the wake of the Almohad invasion in Andalusia, he moved to Toledo, where he died ca. 1180. In Toledo he composed two works, a chronicle entitled the Book of Tradition, and his philosophical The Exalted Faith, which is the subject of this paper. Written around 1160, both were intended as a defense of Judaism. While the former became a classic in Jewish literature, The Exalted Faith, subtitled ‘[The book] that brings harmony between philosophy and religion’, was soon overshadowed by Maimonides’ masterpiece The Guide of the Perplexed, which was written some thirty years later. Ibn Daud wrote the Exalted Faith (Hebrew title: Ha-Emunah ha-Ramah, henceforth: ER) in Arabic (al-‘Aqı¯da al-rafı¯‘a). The Arabic original is now lost, but the work is extant in two Hebrew translations, both dating from the end of the fourteenth century.2 The book has long been neglected in modern scholarship. However, the past two decades have 1 2

Babylonian Talmud, Qidd. 82b; Pes 65b. For more detailed information on Abraham Ibn Daud and his thought see the surveys in Encyclopedia Judaica, ed. 2006, under Ibn Daud, Abraham; The Stanford Encylopedia of Philosophy under Abraham Ibn Daud; and the monographs Eran, From Simple Faith, and Fontaine, In Defence of Judaism.

160 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Resianne Fontaine

borne witness to a new interest in Ibn Daud’s thought, and he continues to attract the attention of scholars of Latin philosophy too, since there are indications that our Ibn Daud is the Avendauth who cooperated with Dominicus Gundissalinus in translating philosophical texts from Arabic into Latin.3 In his introduction to The Exalted Faith Ibn Daud tells us that he was prompted to write his book to address a query about the problem of human freedom. Some Biblical verses seem to teach that man has free will, whereas others, such as: ‘I will harden Pharaoh’s heart’ (Ex 7:3), seem to imply that God determines human action. According to Ibn Daud, the problem can only be solved through a correct interpretation of the verses that contradict reason when interpreted literally. In other words, reason should be the ultimate arbiter between conflicting biblical texts. Regrettably, Ibn Daud continues, many people of his generation fail to investigate the principles of the Jewish religion philosophically and, as a result, they are confused when it comes to solving theological problems, such as that of free will. This is why he seeks ‘to open the eyes of the learned men of our nation regarding the principles of faith by providing proofs from biblical verses and demonstrations from true philosophy.’4 By the term ‘true philosophy’ Ibn Daud means the philosophy as taught by Alfarabi and, in particular, Avicenna. Although some earlier Jewish thinkers display familiarity with views of the fala¯sifa, Ibn Daud is the first Jewish philosopher to make extensive and systematic use of them. In fact, his Exalted Faith can be seen as an attempt to transplant the thought of the fala¯sifa to Jewish soil. Therefore he is commonly considered to be the first Jewish Aristotelian, although the question to what extent his Aristotelianism is mixed with Neoplatonism is a matter of debate. What is undisputed, however, is that with Ibn Daud a ‘wind of change’ enters medieval Jewish thought, primarily because of his use of Avicenna whose views are present throughout Ibn Daud’s book, even though the Muslim philosopher is never mentioned by name. For example, Avicenna’s notions about the soul pervade Ibn Daud’s presentation of the subject, and the same applies to his conception of God as the Necessary Existent.5 Ibn Daud’s approach to the problem of evil is another case in point. To substantiate this claim, I will now highlight the main points of Avicenna’s 3 4 5

For the most recent information on this issue, see Hasse, Avicenna’s ‘De Anima’, pp. 4 – 7, and Bertolacci, Albert the Great, pp. 133 – 7. See also Essudri, R. Avraham bn Daud, R. Efraim al-Naqawa we-sefer ‘al-‘Aqı¯da al-rafi‘a’. References to ER are to ed. Weil 1851 with references to ed. Samuelson/Weiss 1986 (henceforth: S) given in parentheses. ER Intro, ed. Weil, p. 3.7 – 8 (S 5b11 – 13). On Ibn Daud’s use of Avicenna’s concept of the Necessary Existent see Fontaine and S. Harvey, Jewish Philosophy on the Eve of the Age of Averroism.

Abraham Ibn Daud and Avicenna on Evil 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

161

approach to the subject and then proceed to discuss how Ibn Daud responds to it.

2. Avicenna on Evil Avicenna addresses the problem of evil in various works, for example in Kita¯b al-Sˇifa¯’, Kita¯b al-Nag˘a¯t, Kita¯b al-Isˇa¯ra¯t wa’l-tanbı¯ha¯t and Sirr al-qada¯’. ¯ Unfortunately, our information about the circulation of his writings in alAndalus and their use by Jewish philosophers is as yet incomplete.6 Yet, there are several good indications that Ibn Daud derived his knowledge of Avicennian views on evil from the discussion in chapter IX.6 of the Ila¯hiyya¯t of his Sˇifa¯’.7 To begin with, the Sˇifa¯’ was available in Toledo, as it was translated into Latin there in Ibn Daud’s day. Furthermore, Ibn Daud displays familiarity with Sˇifa¯’ and/or Nag˘a¯t in various other sections of ER. Finally, a comparison of the account of evil in the four aforementioned Avicennian texts shows that Ibn Daud’s discussion of the issue has much in common with how the topic is dealt with in both Sˇifa¯’ and Nag˘a¯t. Kita¯b al-Isˇa¯ra¯t wa’l-tanbı¯ha¯t and Sirr al-qada¯’ do not ¯ contain all the issues that appear in Ibn Daud.8 The account in Nag˘a¯t is virtually identical to that in Sˇifa¯’, Ila¯hiyya¯t IX.6.9 Another likely source of Ibn Daud is al˙ aza-lı¯’s rendering of Avicenna’s views in Maqa¯sid al-fala¯sifa.10 G ˙ Chapter IX.6 of the Ila¯hiyya¯t in Sˇifa¯’ is entitled: ‘On providence and how 11 evil enters divine predetermination’. The title thus makes it immediately clear that Avicenna treats the problem of evil as related to his conception of 6 See Zonta, Avicenna in Medieval Jewish Philosophy, and S. Harvey, Avicenna’s Influence on Jewish Thought. For the Andalusian background, see Gutas, What was there for the Latins to Receive?. 7 Avicenna, Sˇifa¯’, Ila¯hiyya¯t IX.6, References to the text are to the Cairo edition, 1960, with the corresponding pages in the text published by Marmura, 2005, given in parentheses. 8 For the latter, see Hourani, Ibn Sı¯na¯’s Essay on the Secret of Destiny, pp. 27 – 31. For Kita¯b al-Isˇa¯ra¯t wa’l-tanbı¯ha¯t, see ed. S. Dunya¯, pp. 729 – 46. For the Nag˘a¯t I have consulted ed. M. Fakhry 1985, pp. 320.5 – 326. 9 This is consistent with Bertolacci’s findings with respect to Sˇifa¯’, Ila¯hiyya¯t VIII.6–IX.3, see Bertolacci, Reception, p. 486. 10 Cf. Eran, From Simple Faith, index under al-Ghazali, and Fontaine, In Defence of Judaism, index, under (al-)Ghazali. See Maqa¯sid, ed. S. Dunya¯, p. 296 – 9. ˙ of Avicenna’s treatment of evil and its 11 Ila¯h. IX.6, p. 414.16 (339.3). For a full study Greek sources, see Inati, The Problem of Evil. This book is based primarily on Sˇifa¯’, but also adduces other Avicennian texts. A very useful overview is Steel, Avicenna and Thomas Aquinas, based on the Latin translation of the Sˇifa¯’, see p. 172, n. 6. See also Janssens, Problem of Human Freedom, who treats the issue on the basis of various Avicennian texts; Belo, Chance and Determinism, especially pp. 38 – 53 and 113 – 20 where the problem of evil is discussed within the framework of an examination of Avicenna’s views on chance, and Michot, La destine de l’homme, pp. 59 – 68.

162 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Resianne Fontaine

providence. He first explains how he understands providence, which comes down to this: the First Cause knows Itself as absolutely good and as the cause of all goodness and perfection of the lower beings. From this apprehension of goodness and order the universe emanates from the First Cause in the best possible way, and this is what is meant by providence.12 Thus, according to Avicenna, God’s knowledge lies at the basis of providence. But if God is the cause of the good and perfect order in the universe, how are we to explain evil? Avicenna starts his theodicy by clarifying the various meanings of the term ‘evil.’ We call evil that which is a deficiency (naqs), like ignorance, weakness, or ˙ a deformity. We also use the term for things like pain and grief.13 This last kind of evil is not just privation like the first kind, for pain is perceived as an existing reality, but it follows upon deficiency and privation.14 Later on Avicenna provides a somewhat different ‘classification’ of evil, where he explains that ‘evil’ is said of: 1) reprehensible acts, which have their principles in 2) moral dispositions; 3) pain and distress; and 4) a defect (nuqsa¯n) or a loss (fuqda¯n) of a ˙ natural perfection.15 Reprehensible acts are evil in so far as they cause privation to the person who performs such acts, as in the case of someone who commits an act of injustice. Moral dispositions are not evil in themselves, but only because of the evil acts that proceed from them.16 What emerges from both passages is that evil is privation and lack or loss of perfection. Furthermore, Avicenna distinguishes between essential evil (al-sˇarr bi’l-da¯t) and accidental evil ¯ (al-sˇarr bi’l-‘arad). Essential evil is privation (‘adam), that is, not just any ˙ privation, but the privation of a perfection that is required by the nature of the thing. Accidental evil is that which produces privation17 or that which withholds perfection from that which deserves it.18 The difference is that essential evil is nothing but evil, while accidental evil is not evil in itself. For example, clouds that withhold sunshine from plants that need the sun for their growth are not in themselves evil. The effect, however, is privation. 12 Ila¯h. IX.6, pp. 414.17 – 415.7 (339.4 – 12); see Steel, Avicenna and Thomas Aquinas, p. 174. 13 Ila¯h. IX.6, p. 415.8 – 10 (339.13 – 15). 14 Ibid. p. 419.6 – 7 (343.16 – 17). 15 Ibid. p. 419.4 – 6 (343.14 – 16). 16 Ibid. p. 419.9 – 10 (344.1 – 2). 17 That which produces privation: this rendition presupposes the reading al-mu‘dim as adopted by Bertolacci, Reception, p. 549. The Cairo edition has al-ma‘du¯m, p. 416.7 (340.12). The Latin translation renders the Arabic passive participle, cf. Steel, Avicenna and Thomas Aquinas, p. 176. For an evaluation of the two different readings, see Inati, The Problem of Evil, pp. 86 – 7. Unfortunately, Ibn Daud’s ER cannot shed light on the issue as he does not reproduce this sentence, which has to do with his silence on the notion of accidental evil, cf. below. 18 Ila¯h. IX.6, pp. 415.18 – 416.5 (340.6 – 13).

Abraham Ibn Daud and Avicenna on Evil 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

163

Since evils are privations they can only occur in things that have potentiality and this is because of matter.19 Matter may or may not receive its ultimate form, but if it does not, this is due to the lack of receptivity of matter, or because of some external impediment (such as the clouds that withhold sunshine from that which needs the sun for its growth), and not because of an agent who denies the perfection. However, Avicenna also concedes that a more active external agent may be at work, for example, an opposing element, as when cold strikes plants and thus removes their perfection.20 Yet since evil is connected with matter, it is found only in the sublunary world, that is, the world of generation and corruption. This implies that evil is in fact insignificant as the sublunary world is only a small part of the universe.21 Avicenna has more good news: evil afflicts only individuals, not the species, and only at certain times. In fact, it is rather the exception than the rule. His stock-example is the effect of fire: fire can indeed cause great harm to individuals, but on the whole it is beneficial. Evil is thus a necessary concomitant of the good. For Avicenna a good can only be a good if evil also can ensue from it and with it.22 It is the good thing that is willed by God primarily and this is what happens for the most part. Yet even if good far outweighs evil, why should there be evil at all? Could not the First Governor have brought into existence a world that is entirely free from evil? This would be possible, Avicenna agrees, in ‘absolute existence’, but such a world would not be our world. By ‘absolute existence’ Avicenna means the intellectual, psychological and celestial things that emanate from God, that is the superior, good causes. However, if emanation stopped at the level of these causes, there would be no need for the activity of the higher causes, and the universe would be less complete than it is now. Withholding evil, Avicenna concludes, would be the greatest fault in the universal order of the good.23 Finally, he argues that even though evils may be numerous, like illnesses, they are not preponderant. Indeed there is a kind of evil that is preponderant, like ignorance in a certain field (for example, geometry) or a lack of radiant beauty, but these fall under the category of superabundance and excess. Being an outstanding geometer or extremely beautiful does not belong to the category of primary perfections. Again, such evils have to do with a deficiency in the recipient, and are not due to action on the part of an agent, but rather to inaction.24 19 20 21 22 23

Ibid. p. 416.9 – 10 (340.16). Ibid. pp. 416.1 – 417.4 (340.18 – 341.7). Ibid. p. 417.4 – 5 (341.8 – 9). Ibid. p. 418.5 – 6 (342.9). Ibid. p. 418.12 – 20 (342.17 – 343.8). As Steel, Avicenna and Thomas Aquinas, p. 185 puts it: ‘the universe would be less perfect than it is now. For a possible level of existence would be lacking.’ 24 Ibid. p. 422.9 – 18 (347.3 – 12).

164 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Resianne Fontaine

In sum, the key-notions in Avicenna’s theodicy are the following: evil is privation, and is associated with potentiality and matter. As for its presence in this world, Avicenna justifies it by stating that: 1) the sublunary world is only a small part of the universe; 2) it is rare since it only strikes individuals; 3) it is a necessary concomitant of the good; 4) without it the universe would be less perfect. In view of these ‘solutions’ Carlos Steel has characterized Avicenna’s metaphysics as ‘optimistic.’25 If Avicenna is indeed an optimist, then Ibn Daud also qualifies as one, as I will try to show in what follows.

3. Ibn Daud’s Use of Avicenna’s Account of Evil It can immediately be noted that Ibn Daud’s discussion of evil bears a great similarity to that of Avicenna as articulated in the Sˇifa¯’.26 I shall briefly mention its main features before addressing them in greater detail. Like Avicenna, Ibn Daud treats the problem of evil in connection with the question of God’s knowledge and providence, and as in the Sˇifa¯’, the discussion is found towards the end of his book.27 The two expositions display roughly the same order of treatment and share the same basic views: evils are privations; evil is due to matter, and thus found only in the sublunary world. Ibn Daud also accepts Avicenna’s ‘justifications’ for the presence of evil in this world and they constitute in fact the main ingredients of his theodicy. Such a theodicy is not found in the writings of earlier Jewish philosophers. One of Ibn Daud’s predecessors, Abraham bar Hiyya (d. 1136?) of Barcelona, ˙ mentions the theory of privation in his ethical-philosophical treatise The Meditation of the Sad Soul (Hegyon ha-nefesˇ ha-‘asuvah) as a solution put forward by ‘the philosophers’ to account for the presence˙of evil. He does not refer to the term ‘privation’, however, and he rejects the theory.28 Another ‘Spanish’ Abraham, the many-sided scholar Abraham Ibn Ezra, who was an older contemporary of Ibn Daud, views the imperfection of the recipient as the origin of evil, as does Avicenna, and also holds that it is not fitting for superior wisdom to preclude the greater good because of small evil.29 However, the combination 25 Steel, Avicenna and Thomas Aquinas, p. 181. 26 For Ibn Daud’s discussion of evil, see also Eran, From Simple Faith, pp. 236 – 51 and Fontaine, In Defence of Judaism, pp. 196 – 204. These monographs treat Ibn Daud’s account of evil within the framework of an overall analysis of his thought, and in relation to other sections of his book. The present paper focuses more specifically on Ibn Daud’s use of Ibn Sina. 27 Abraham ibn Daud, Ha-Emunah ha-Ramah II.6.2, ed. Weil 1851, pp. 93 – 8 (S 201a12 – 208b11). 28 Hegyon ha-nefesˇ, ed. Wigoder, pp. 123 – 4. 29 Comm. on Ecclesiastes, Introd.

Abraham Ibn Daud and Avicenna on Evil 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

165

of views on evil as found in The Exalted Faith is new to Jewish philosophy, and therefore, we can safely assume that Ibn Daud drew on Avicenna’s exposition. In fact, this conclusion is wholly consistent with Ibn Daud’s stated declaration, referred to above, that he needed ‘true philosophy’, that is, the philosophy of Avicenna, to solve the problem of free will. After all, this problem is closely connected to that of evil, and Avicenna is the first Muslim philosopher to provide a systematic treatment of the problem. However, as we will see, Ibn Daud’s treatment of the topic is far from a slavish copy of that of Avicenna. Ibn Daud addresses the issue in the penultimate chapter of his book under the heading ‘On the origin of good and evil’,30 which is subtitled: ‘On the order and number of causes, providence, and the secret of [God’s] power.’ To this he adds: ‘This is the chapter because of which we have composed the book.’ Ibn Daud does not provide an explicit formulation of what he understands by ‘the problem of evil’, nor does he offer anything that resembles a ‘classification’ of evils. However, it may be inferred from his account that like Avicenna he stands in the tradition of the Stoics and Neoplatonists to whom the question was: ‘how can evil exist in a world that proceeds from the One who is perfect?’31 His starting point is the thesis that God knows His own essence. He then goes on to sketch the structure of the universe: the farther a thing is removed from matter, that is, potentiality, the more complete is its knowledge, since ‘Every deficiency and every evil originates in what has something in potentiality’, a statement that is a close rendering of Avicenna’s aforementioned view: ‘Evil attaches only to that which has in its nature what is potential.’32 God, who is furthest removed from potentiality, knows His essence in the most perfect way and He also knows that His perfection is not intended towards Himself alone, but that it emanates to others in progressive degradation. Thus, existence has two extremes: at the one end that which is perfect and does not admit of any deficiency, and at the other that which we call ‘matter’, which is removed from perfection to the utmost degree.33 All this is more or less a summary of his views as developed earlier in his book, views that are taken from the fala¯sifa. Yet at the same time this point of departure with its emphasis on God’s knowledge, perfection, emanation and the hierarchical order of the universe displays all the ingredients of Avicenna’s conception of providence. Unlike Avicenna, however, Ibn Daud, is careful not to present this position as entailing providence, because, as we shall see, he has a different understanding of what providence is. 30 Or, according to some mss, ‘on free will’. 31 For a survey of the treatment of this dilemma by Greek and Latin thinkers, see J. Opsomer and C. Steel, Evil without a cause; Proclus’ doctrine on the origin of evil, and its antecedents in Hellenistic philosophy, in Zur Rezeption der hellenistischen Philosophie in der Sptantike, eds T. Fuhrer et al., Stuttgart: Franz Steiner Verlag, 1999, pp. 229 – 61. 32 ER p. 93.28 (S 201b3 – 4), cf. Ila¯h. IX.6, p. 416.9 – 10 (340.16). 33 ER p. 93.25 – 41 (S 201a16 – 202a2).

166 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Resianne Fontaine

Ibn Daud’s next step is to emphasize that according to reason, Scripture, and the Jewish exegetical tradition, it is impossible that evil or deficiency proceed from God.34 Despite Ibn Daud’s categorical assertion, it should be noted that the position that God produces evil alongside the good was in fact defended by one of his compatriots, the aforementioned Abraham bar Hiyya, in an exegetical ˙ context. This philosopher based himself on Isaiah 45:7 where God is said to be ‘the maker of peace and the creator of evil’, accepting the literal meaning of the text.35 Given Ibn Daud’s overall aim to reject the literal meaning of biblical verses when they contradict reason he obviously could not adopt this position.36 As he argues, God would be composite and not One if two opposites were united in him. God is not like man, from whom can proceed both good (on account of his rational soul) and evil (on account of his lower faculties).37 Since it is likewise impossible that only evil proceeds from God, the question of the origin of evil imposes itself. Thus far we have heard that deficiency and evil are due to matter and potentiality, and to the lower faculties of the soul. Ibn Daud then introduces rather abruptly the notion of ‘privation’ with the statement that ‘among the attributes of deficiencies some are privative (he‘ederiyyim) and some are positive (hiyyuviyim).’38 This statement is puzzling, since Ibn Daud does not make clear ˙ what he means by ‘positive’ attributes of deficiencies, nor does he provide any examples of them. In my view, this distinction can be taken to reflect the distinction made by Avicenna between evils like ignorance and deformities that are privations of being, and evils like pain and distress, which are apprehended as very present and real, that is, as existing things.39 If this interpretation is correct, perhaps a better translation of hiyyuviyim would be ‘existing.’40 Ibn ˙ Daud shows himself to be more interested in the privative evils, of which darkness is an example. As he explains: ‘light is something in the air that is produced (mithaddesˇ ) when a luminous body shines on it, and when the ˙ luminous thing is removed, it remains dark, and nothing is produced in it, rather the produced thing departs from it. Thus, darkness is the privation of a

34 35 36 37 38 39

ER p. 93.42 – 3 (S 202a3 – 5). Abraham bar Hiyya, Hegyon ha-nefesˇ, pp. 119 – 22. ˙ – 4 (S 2b22 – 5) ER intro, p. 1.31 ER pp. 93.45 – 94.3 (S 202a6 – 11). ER p. 94.4 – 5 (S 202a11 – 12). Ila¯h. IX.6, p. 419.6 – 7 (343.16 – 17), cf. also p. 415.12 – 13 (340.4 – 5). Cf. Steel, Avicenna and Thomas Aquinas, pp. 174 – 6. 40 The second Hebrew translation, Ha-Emunah ha-Nisa’ah, reads mehuyyavim (necessary). The Arabic has: wugˇu¯diyya, which seems to underlie the reading of ˙ER. The term hiyyuvi ˙ also occurs in ER I.3 in the sense of ‘actualization’, ‘realization’.

Abraham Ibn Daud and Avicenna on Evil 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

167

thing, but it is not a thing [itself ].’41 Privations are not made by God, or by any other agent, because they are not beings that can be produced. Thus, one cannot say that God made the privation of elephants in Spain, but rather that He did not make an elephant in Spain. In other words, there is inaction on the part of God, not action. If this absence is a deficiency in Spain, God did not produce it; He just did not make the perfection that removes the deficiency, and left it undetermined whether or not there would be elephants in Spain.42 From darkness and absence of elephants, Ibn Daud moves on to ignorance. The same applies here: when a person’s intellect remains intellect in potentiality, it cannot be claimed that God ‘produces’ ignorance. God merely did not decree that this intellect would become intellect in actuality, so that it remains on the level of matter. Hence, since most vices and defects have their origin in a lack of intellect, God cannot be their maker.43 At this point, however, Ibn Daud anticipates an objection. It could be argued, he says, that the two cases are not comparable, ‘because it does not belong to the nature of Spain to have elephants walking around there, and their absence is not viewed as a deficiency, whereas it certainly belongs to the nature of man to possess intellect and perfection, and their absence is seen as a deficiency.’44 Such a privation seems incompatible with God’s goodness, for withholding perfection may be seen as evil. Ibn Daud’s response to the objection is based on Avicenna’s contention that imperfection is to be ascribed to matter. It is found only in that which is composed of the elements, in other words, in the sublunary world. This forces him to account for the occurrence of evil in our world, which is why he argues that the reception of form by matter may be prevented by some external impediment, so that a thing will not attain its natural end. However, instead of Avicenna’s example of clouds that withhold sunshine, Ibn Daud evokes another image: When a sower sows seed on the land, a part will fall in fertile moist ground, wellpositioned with respect to the sun, so that the mixture of the seed can benefit from it and lo, it will spring up and grow well. But the part that falls unto dry barren ground, or ground that is not positioned thusly that the mixture can benefit from the sun45 will grow up only weakly. The part that will fall onto stony ground, will 41 ER p. 94.7 – 8 (S 202a13 – 16). Another example is poverty, which is the privation of the opposite, namely the presence of money. The Hebrew term used for privation (he‘der) renders the Arabic ‘adam, and like the Arabic it can also be translated as non-being or absence. 42 ER p. 94.8 – 16 (S 202a16 – 202b8). 43 ER p. 94.16 – 24 (S 202b8 – 203a2). 44 ER p. 94.25 – 9 (S 203a2 – 5). 45 The reading ‘sun’ (ha-sˇemesˇ ) is supported by various manuscripts. Weil’s printed text has sadeh (field).

168 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Resianne Fontaine

not spring up at all, it will be blown away by the storm scattered or eaten by birds or it will become foul.46

The ultimate source of this image is of course the parable of the sower of the evangelists.47 As Amira Eran has shown, however, Ibn Daud’s direct sources were ˙ aza-lı¯’s Maqa¯sid and the collection of stories entitled Kita¯b al-malik wa’lal-G ˙ and The Ascetic) in Arabic.48 We can add that in another na¯sik (The Prince context Avicenna himself employs the example of seed that does not fall on this particular spot of land always or for the most part, namely in his expositions on chance in the Physics of his Sˇifa¯’.49 Next he brings up Avicenna’s argument according to which evil is small when compared to the good, which is far more widespread. This contention introduces Ibn Daud’s attempt to minimize and even legitimize evil. In so doing, he follows in the footsteps of the Muslim philosopher, for Avicenna’s various strategies to this purpose appear in The Exalted Faith. Ibn Daud, however, gives them a different flavor by choosing his own examples, which are drawn from Rabbinic literature. Here we come across Avicenna’s question whether the First Governor could not have brought into being a world that is entirely free from evil. Ibn Daud’s wording, however, is slightly different: the Creator is under no obligation to create a world in which everything reaches its final perfection, and he goes on to say: Whosoever says that this should have been the case and that the Creator should have chosen the best is saying something that is absurd. If the Creator had chosen the best for everything, then necessarily all plants would be like animals, all animals like man, and all man like Moses, our Rabbi, and Moses our Rabbi, like the most sublime of the angels.50

In such a perfect world there would be no order, or hierarchy. All existence would be uniform for everything would have reached its final perfection. Like Avicenna, Ibn Daud thus believes that a world in which there is hierarchy and order is better than a world in which there is only good. As Ibn Daud puts it rather poetically: ‘existence would come to an end by the very power of its perfection, like something that is cut in its prime’, a phrase that is inspired by a verse from Job.51 He adds that in such a world God’s perfection would emanate only on few existents while God’s goodness is great and is intended to emanate on many things. 46 47 48 49 50 51

ER p. 94.34 – 9 (S 202b8 – 203a2). Cf. Matt. 13:3 – 9; Mk. 4:3 – 9; Lk 8:5. Eran, The Parable of the ‘Sower and the Seeds’, pp. 139 – 51. See Belo, Chance and Determinism, p. 43. ER pp. 94.42 – 5. ER pp. 94.ult.–95.1 (S 203b11 – 12); cf. Job 8:12.

Abraham Ibn Daud and Avicenna on Evil 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

169

At this juncture Avicenna’s thesis that evil is a necessary concomitant of the good makes its appearance in Ibn Daud as well. The Jewish thinker refers to tiqqun ha-kelal (improvement of the whole), which renders Avicenna’s nizza¯m ˙ ˙ by al-hayr al-kullı¯ (universal order of the good).52 Ibn Daud supports the thesis ˘ means of Rabbinic dicta. It is in this context that the Talmudic quotation of the title of this paper is found. The full quotation is: The world cannot do without perfume-makers and tanners; blessed is he whose occupation is that of perfume-maker and unhappy he whose occupation is that of tanner. The world cannot do without men and women. Blessed is he whose children are boys and unhappy he whose children are girls.53

To this he adds that every individual wishes to have boys, for this is generally regarded as perfection, but luckily God considers the good of the whole. For the same reason, he continues, the High Priest used to implore God not to heed the prayer of travelers who always ask for fine weather, for without rain all harvesting would come to an end.54 This particular example may also have been ˙ aza-lı¯’s rendering of Avicenna’s views on evil in the Maqa¯sid, for inspired by al-G ˙ he likewise writes that without rain all sowing would come to an end.55 In his elaboration of tiqqun ha-kelal, Ibn Daud goes a long way in agreeing with Avicenna. As was to be expected, he also employs the stock-example of fire, claiming that damage wrought by fire or wind is a necessary concomitant of the good, and, again, that these evils are not predominant.56 Moreover, like Avicenna he accepts that universal order necessitates the existence of individuals with some kind of evil in their bodies or souls. Ibn Daud, however, goes into more detail with respect to what he calls ‘deficient species’. These species include people whose lack of intellect is counterbalanced by a strong physical constitution, which makes such people fit to do hard labor, like animals, which is beneficial to society as a whole.57 Ibn Daud thus refers both to universal order and to order in society, which may explain the use of the term

52 Avicenna means by al-nizza¯m al-kullı¯: ‘that the whole [is such] that the active and passive ˙˙ powers, celestial and terrestrial, natural and psychological, have been organized in it in such a way only so that they would lead to the universal order’ (tr. Marmura, p. 346). 53 Cf. n. 1. ER p. 95.5 – 10 (S 204a4 – 8). The existence of women is thus not an evil in itself, as it serves a good end, but individuals may view the birth of daughters as evil, since all people wish to have sons. Cf. Aristotle’s notion that the formation of the female is in fact a deviation, even though nature requires the existence of females, see his Generation of Animals XVIII, 767b8 – 10. 54 ER p. 95.13 – 16 (S 204a810 – 13), cf. b Yoma 53b and Ta‘anit 24b. 55 Maqa¯sid, ed. S. Dunya¯, p. 298. 56 ER p.˙ 95.26 – 33 (S 204b8 – 13); cf. Ila¯h. IX.6, p. 418.3 – 4 (342.6 – 8). 57 ER p. 95.16 – 26 (S 204a13 – 204b8).

170 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Resianne Fontaine

tiqqun (improvement) of the whole here for Avicenna’s nizza¯m.58 Ibn Daud even ˙ goes so far as to say that people whose temperament is ˙such that they quarrel and kill due to an excess of heat in their bodily constitution have their own place in creation just like bears and dogs: In the matter out of which the wicked sinful person comes to be there is an excess of heat, be it in the constitution of the seminal fluid or in the womb, or both, and following the heat that has assembled in [this person’s] heart in an unbalanced proportion, a certain disposition will occur, which is called anger. As a result, [such a person] will fight with people and be ready to bite them, as dogs do, and it may happen that he will slay and kill them, like wild animals do. But just as it is not strange that there should exist dogs and wild animals, it is not strange that there should be among the species man a species which resembles man, but whose nature is like that of animals.59

This position may appear surprising at first sight, for it seems to leave little hope for those who are the unfortunate possessors of such a constitution. A possible explanation would be that Ibn Daud’s intention is to draw a parallel between metaphysical and moral evil, and to finally state his view on human freedom. His problem is again, or still, the absence of intellect. It is this particular issue on which Ibn Daud’s account of evil focuses and to which it gravitates. In the passage discussed above (cf. above, text to n. 44) we learned that lack of intellect, viewed on the physical level, does not come from God; instead it represents a case of a potentiality in matter that has failed to become actualized. In the present passage Ibn Daud argues that moral evil cannot come from God either: evil actions result from a deficiency in one’s physical constitution, in other words, they also go back to matter. However, he hastens to add that people who unfortunately lack knowledge and cannot distinguish right from wrong are not left to their own devices, for God has sent people who are entirely free from evil, that is, the prophets who brought commandments and warnings.60 It is at this point that Ibn Daud turns to his defense of free will. Here his understanding of the ‘possible’ plays a crucial role, the possible being a class of existents alongside necessary and impossible things.61 According to Ibn Daud, it is not impossible for man to oppose his bad character traits, for God does not command the impossible. Nor is it necessary, like breathing. Instead it is possible. Just as God left it undetermined whether or not there would be 58 The more literal rendering siddur (order) also occurs in the Hebrew text. Of course we do not know what the Arabic original had. 59 ER p. 95.33 – 6 (S 205a4 – 8). I adopt the reading zar (strange) here, with most of the manuscripts. Weil’s printed text has zeh (this), which does not make sense. (S 205a5 – 6 also reads zeh, but translates ‘strange’). 60 ER pp. 95 ult.–96.3 (S 205a15 – 16). 61 Earlier on in ER Ibn Daud had already argued that the impossible in fact does not exist, ER p. 47.41 – 4 (S 126b4 – 5).

Abraham Ibn Daud and Avicenna on Evil 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

171

elephants in Spain, He left it undetermined which course of action man will choose for himself. Freedom of action is guaranteed since God created the possible as possible and knows it only as possible.62 At this point of his discussion, Ibn Daud introduces the notion of causality. Advancing a fourfold division of causes into divine, natural, accidental and free causes, he holds that it is thanks to the existence of the last type of cause that man is free to act.63 While Ibn Daud’s focus in this section is on free will, the problem of evil also plays a part here, for he contends that man can protect himself against the evil results of natural and accidental causes, for example by refraining from certain kinds of food and lethal drugs. The introduction of accidental causes at this point is not without difficulty in view of Ibn Daud’s earlier account of evil. As has been shown above, there are many points of contact between his and Avicenna’s treatment of the problem. There is, however, a significant difference in that Ibn Daud nowhere mentions Avicenna’s aforementioned distinction between essential and accidental evil, a distinction that is crucial for Avicenna. Similarly, Ibn Daud bypasses Avicenna’s position that evil was willed by God in an accidental way.64 I suspect that his silence has to do with his insistence that evil has no agent and can by no means proceed from God. To be sure, both thinkers hold that privations are not due to an action on the part of God, but rather to inaction, but nonetheless Avicenna seems to designate the agent as a cause of evils that are attached to good things.65 Ibn Daud is more emphatic with regard to inaction by the agent, especially when it comes to accounting for the absence of intellect in certain individuals, where he underlines that God is not the cause of ignorance. Avicenna does not single out this particular point for discussion, but for Ibn Daud it is most relevant, since it bears on his defense of free will. Stating that evil proceeds from God’s will, albeit accidentally, would imply that to some extent God is the agent of ignorance either by action or inaction, and hence, that He may ultimately be held responsible for any bad actions that result from ignorance. It is therefore possible that it is Ibn Daud’s eagerness to avoid this notion that led him to ignore Avicenna’s distinction between essential and accidental evil.66 In this regard it is also noteworthy that Avicenna’s classification of evils has no counterpart in Ibn Daud: Ibn Daud seems primarily interested in one particular evil: lack of intellect. However, the notion of accidental evil, or of evil brought about by an external agent is not absent from ER, as is clear from the aforementioned parable 62 63 64 65 66

ER p. 96.21 – 4 (S 206a5 – 8). ER 96.33 – 97.8. (S 206b1 – 207a7). Ila¯h. IX.6, pp. 420.19 – 421.1 (345.11). Ibid., lines 11 – 15, Cf. Inati, The Problem of Evil, pp. 91 – 2. Belo, Chance and Determinism, p. 117 concludes that, according to Avicenna’s Nagˇa¯t, ‘evil too is willed and created by God’.

172 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Resianne Fontaine

of the sower. Ibn Daud obviously admits that some effects do not attain their intended goal. Though he ascribes this failure to lack of receptivity of matter, it is difficult not to view the storm or the birds of Ibn Daud’s example as external agents. Moreover, in the present passage on causality the implication that evil ultimately comes from God, albeit indirectly, or, as Ibn Daud puts it not ‘by primary intention’, because God entrusted some causes to nature, seems unavoidable. Our author, however, does attempt to avoid it by claiming that it is given to man to be on his guard against evil effects and that he is helped in this by revealed law. At any event, in his defense of free will Ibn Daud cannot follow Avicenna any longer. Of course it can be and has in fact been argued that Avicenna also seeks to absolve God from any blame for moral evil and even that he allows for some human freedom, but this does not come to the fore in the Sˇifa¯’, Ibn Daud’s most likely source.67 In the relevant chapter of the Sˇifa¯’ Avicenna quotes with approval prophetic traditions like ‘I created these for the fire and I care not; and I have created these for paradise and I care not’,68 and, given Ibn Daud’s overall aim, the Jewish philosopher cannot but reject Avicenna’s position. He has to defend human freedom forcefully because it is on free will that man’s acceptance of the biblical commandments hinges, and for Ibn Daud the end of ‘theoretical philosophy’ is practice, that is, religious practice. At the end of his discussion Ibn Daud offers his own view of providence, and here he definitely departs from Avicenna. While for Avicenna, as may be recalled, providence consists in God knowing Himself as the cause of the good, for Ibn Daud providence lies in the assistance on the part of God and the angels that is granted to man following man’s own initial choice. When man chooses the good, he will be helped to pursue his goal. However, when one opts for evil, one will be directed on that path. In this manner, the biblical verse ‘I will harden Pharaoh’s heart’ must be understood, for Pharaoh, by his own free will, refused to let the Israelites go, and he was helped to persist in his choice accordingly.69 We can thus conclude that despite the obvious similarities between Avicenna’s and Ibn Daud’s treatments of evil, there are important differences between the two philosophers. Ibn Daud treats the topic of evil in a wider conceptual framework, since his ultimate goal is the defense of free will, the key67 Avicenna’s expositions on determinism have given rise to divergent interpretations: while scholars like G. Hourani and C. Belo view Avicenna’s position as precluding human freedom, A. Ivry and J. Janssens have argued that Avicenna’s system leaves some room for free human action. See Hourani, Ibn Sı¯na¯’s Essay on the Secret of Destiny, pp. 40 – 41; Belo, Chance and Determinism, pp. 52 – 3 and pp. 116 – 20; Ivry, Destiny Revisited, pp. 167 – 70; Janssens, Problem of Human Freedom, pp. 112 – 18. 68 Ila¯h. IX.6, p. 422.7 – 8 (347.1 – 2). 69 ER p. 98.4 – 7 (S 208b2 – 6).

Abraham Ibn Daud and Avicenna on Evil 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

173

theme of The Exalted Faith. This circumstance explains his focus on the lack of intellect, for this problem constitutes a major obstacle in upholding human freedom. Yet at the same time his discussion is narrower and less detailed than that of his source precisely because of his selective use. In other words, Abraham Ibn Daud follows Avicenna only to the extent that the viewpoints of the Muslim philosopher suit his own purposes, while at the same time he displays a considerable independence from his source. In struggling with the dilemma evil versus providence Avicenna and Ibn Daud put forth views and arguments that are well known from Neoplatonic and Stoic sources.70 It is difficult to determine to which tradition exactly Ibn Daud belongs. His account reveals a mixture of different and sometimes conflicting views: with Plotinus he maintains that evil is to be ascribed to matter and privation; the notion that evil is a side effect, some kind of collateral damage that has no agent echoes Proclus’ position, while Stoic thought is the source for the view that evil when viewed from a universal perspective is only apparent evil, and Ibn Daud adduces biblical passages in support of all these viewpoints. In view of his use of these stock arguments it can be claimed that to some extent Ibn Daud’s explanation of evil is ‘unoriginal’. Yet, from the perspective of the history of medieval Jewish philosophy, it can be called innovative, precisely because of his use of Avicenna and the manner in which he adapts his doctrine. As we have seen no Jewish author before him had attempted to provide a philosophical systematic discussion of the issue. The question whether Ibn Daud’s use of Avicenna has left traces in the works of later medieval Jewish philosophers remains to be explored. A few decades later, Maimonides addressed the problem of evil in his Guide, taking issue with several of the views discussed above, such as the existence of physical and moral evil, divine providence and omniscience. Much has been written already on Maimonides’ positions on these themes.71 However, what is still lacking is a comprehensive study that takes into consideration Maimonides’ use of Avicenna’s theodicy in relation to Ibn Daud’s use. Such a study, it is expected, will throw more light on the reception of Avicenna and his theodicy in medieval Jewish philosophy.

70 For discussions of Ibn Sı¯na¯’s sources, see Hourani, Ibn Sı¯na¯’s Essay on the Secret of Destiny, and Inati, The Problem of Evil, pp. 54 – 65. The question to what extent the ‘solutions’ that the Muslim and the Jewish philosopher put forth were acceptable to their respective audiences is intriguing, but does not fall within the scope of this paper. 71 See Burrell, Maimonides, Aquinas and Gersonides; Rosenberg, Good and Evil, pp. 24 – 31; W.Z. Harvey, Maimonides and Spinoza on the Knowledge of Good and Evil; Pines, Truth and Falsehood; Dobbs-Weinstein, Matter as Creature; Motzkin, Maimonides and Spinoza on Good and Evil; Leaman, Evil and Suffering, pp. 64 – 101.

174

Bibliography Primary Sources

˘

Abraham bar Hiyya, Hegyon ha-nefesˇ ha-‘asuvah, ed. G. Wigoder, Jerusalem: Mosad ˙ ˙ Bialiq, 1971. Abraham Ibn Daud, Ha-Emunah ha-Ramah, ed. S. Weil, Frankfurt a. M., 1852. –– , Ha-Emunah ha-Ramah, The Exalted Faith, eds N.M. Samuelson and G. Weiss, London/Toronto: Associated University Presses, 1986. ˙ aza-lı¯, Metaphysica, Maqa¯sid al-fala¯sifa, ed. S. Dunya¯, Cairo: Da-r al-Ma a-rif, 1960. al-G ˙ ¯ t, vol. 2, eds M.Y. Mu¯sa¯, S. Dunya¯, Cairo: al-Hay’a alAvicenna, Kita¯b al-Sˇifa¯’, Ila¯hiyya ‘a¯mma li-sˇu‘u¯n al-mata¯bi‘ al-amı¯riyya, 1960. –– , The Metaphysics of ˙‘The Healing’, transl. and ed. M. Marmura. Provo, Utah: Brigham Young University Press, 2005. –– , Kita¯b al-Nag˘a¯t fı¯ al-hikma al-mantiqiyya wa-al-tabı¯ iyya wa-al-ila¯hiyya, ed. ˙ ˙ 1985. ˙ M. Fakhry, Beirut: Dar al-Afaq al-Jadidah, –– , Kita¯b al-Isˇa¯ra¯t wa’l-tanbı¯ha¯t, ed. S. Dunya¯, Cairo: Da¯r al-Ma a¯rif, 1958. ˘

˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Resianne Fontaine

Secondary Sources C. Belo, Chance and Determinism in Avicenna and Averroes, Leiden/Boston: Brill, 2007. A. Bertolacci, Albert the Great and the Preface of Avicenna’s Kita¯b al-Sˇifa¯’, in Avicenna and his Heritage, eds J. Janssens and D. De Smet, Leuven: Leuven University Press, 2002, pp. 131 – 52. –– , The Reception of Aristotle’s ‘Metaphysics’ in Avicenna’s Kita¯b al-Sˇifa¯’: a Milestone of Western Metaphysical Thought, Leiden: Brill, 2006. D.B. Burrell, Maimonides, Aquinas and Gersonides on Providence and Evil, Religious Studies, 20/3, 1984, pp. 335 – 51. I. Dobbs-Weinstein, Matter as Creature and Matter as the Source of Evil: Maimonides and Aquinas, in Neoplatonism and Jewish Thought, ed. L.E. Goodman, Albany, New York: State University of New York Press, 1992, pp. 217 – 35. A. Eran, From Simple Faith to Sublime Faith. Ibn Daud’s Pre-Maimonidean Thought, Tel Aviv: Ha-kibbutz ha-me’uhad, 1998 (in Hebrew). –– , The Parable of the ‘Sower and the Seeds’ in the Exalted Faith, Tarbiz, 73, 2003, pp. 139 – 51(in Hebrew). Y. Essudri, R. Avraham bn Daud, R. Efraim al-Naqawa we-sefer ‘al-‘Aqı¯da al-rafi‘a’ (in Hebrew, forthcoming). R. Fontaine, Abraham Ibn Daud, Encyclopedia Judaica, vol. 9, 2nd edn, Jerusalem: Gale Cengage Learning, 2006, pp. 662 – 5. –– , Abraham Ibn Daud, The Stanford Encylopedia of Philosophy, at http:// plato.stanford.edu/entries/abraham-daud. –– , In Defence of Judaism: Abraham Ibn Daud. Sources and Structure of ‘ha-Emunah haRamah’, Assen: Van Gorcum, 1990. R. Fontaine and S. Harvey, Jewish Philosophy on the Eve of the Age of Averroism: Ibn Daud’s Necessary Existent and his Use of Avicennian Science, in The Age of Averroism, ed. P. Adamson, London: The Warburg Institute, 2011, pp. 215–27.

Abraham Ibn Daud and Avicenna on Evil 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

175

D. Gutas, What was there for the Latins to Receive? Remarks on the Modalities of the Twelfth-Century Translation Movement in Spain, in Wissen ber Grenzen, Miscellanea Mediaevalia 33, ed. A. Speer and L. Wegener, Berlin: de Gruyter, 2006, pp. 3 – 21. S. Harvey, Avicenna’s Influence on Jewish Thought: Some Reflections, in Avicenna and his Legacy: A Golden Age of Science and Philosophy, ed. Y.T. Langermann, Turnhout: Brepols, 2009, pp. 327 – 40. W.Z. Harvey, Maimonides and Spinoza on the Knowledge of Good and Evil, Binah 2, 1989, pp. 131 – 46. D.N. Hasse, Avicenna’s ‘De anima’ in the Latin West. The Formation of a Peripatetic Philosophy of the Soul, 1160 – 1300, London/Turin: The Warburg Institute/Nino Aragno Editore, 2000. G.F. Hourani, Ibn Sı¯na¯’s Essay on the Secret of Destiny, Bulletin of the School of Oriental and African Studies, 29, 1966, pp. 25 – 48. S. Inati, The Problem of Evil: Ibn Sı¯na¯’s Theodicy, Binghamton, New York: Global Publications, 2000. A. Ivry, Destiny Revisited: Avicenna’s Concept of Determinism, in Islamic Theology and Philosophy: Studies in Honor of George F. Hourani, ed. M. E. Marmura, Albany: SUNY Press, 1984, pp. 160 – 71. J. Janssens, The Problem of Human Freedom in Ibn Sı¯na¯, in Ibn Sı¯na¯ and his Influence on the Arabic and Latin World, ed. J. Janssens, Aldershot: Ashgate, 2006, pp. 112 – 18. O. Leaman, Evil and Suffering in Jewish Philosophy, Cambridge: Cambridge University Press, 1995. J. Michot, La destine de l’homme selon Avicenne: le retour  Dieu (ma’ad) et l’imagination, Leuven: Peeters, 1986. A.L. Motzkin, Maimonides and Spinoza on Good and Evil, Da‘at 24, 1990, pp. v-xxiii. S. Pines, Truth and Falsehood Versus Good and Evil: a Study in Jewish and General Philosophy in Connection with the Guide of the Perplexed I,2, in Studies in Maimonides, ed. I. Twersky, Cambridge, Massachusetts/London: Harvard University Press, 1990, pp. 95 – 157. S. Rosenberg, Good and Evil in Jewish Thought (Hebr.), Tel Aviv: Ministry of Defence Publishers, 1985. C. Steel, Avicenna and Thomas Aquinas on Evil, in Avicenna and his Heritage, eds J. Janssens and D. De Smet, Leuven: Leuven University Press, 2002, pp.171 – 96. M. Zonta, Avicenna in Medieval Jewish Philosophy, in Avicenna and his Heritage, eds J. Janssens and D. De Smet, Leuven: Leuven University Press, 2002, pp. 267 – 80.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Possible Hebrew Quotations of the Metaphysical Section of Avicenna’s Oriental Philosophy and Their Historical Meaning* Mauro Zonta Introduction Among Avicenna’s philosophical works that have the general structure of an encyclopedia, there is The Oriental Philosophy (al-Hikma al-masˇriqiyya), whose ˙ ways: either as ‘Eastern title has been usually interpreted in two different Philosophy’, or as ‘Illuminative/Enlightening Philosophy’.1 Although the former interpretation of the title has often been criticized, a number of considerations militate in its favour. Some of these might become clear from the examination of some Hebrew quotations of apparently lost passages of the work, neglected until now. These quotations are found in some fourteenth-century Hebrew philosophic-theological works, whose texts have been published only recently (mostly in medieval Spanish translations), and whose contents and sources have been examined in detail very recently by Ryan Szpiech in a very interesting article.2 According to Szpiech, these quotations were incorrectly ascribed to Avicenna, but were in reality taken from the Hebrew version of Ibn Tufayl’s Living Son of the Watchful (Hayy ibn Yaqza¯n), made in Provence ˙at the ˙ beginning of the fourteenth century. However,˙ Szpiech does not point out that the quotations even show some surprising similarities with corresponding ideas of one of the many Indian philosophies and of Hinduism as well, as these latter traditions might have been known in Avicenna’s time and milieu. I will reexamine all these fragments, and will try to suggest some tentative hypotheses about them, which differ in part from those suggested by Szpiech. The general

*

1 2

Many thanks to Amos Bertolacci and Dag Nikolaus Hasse for their useful observations, which have helped me in improving some points of the article, which is a partially revised version of the original paper What Remains of the Metaphysical Section of Avicenna’s al-Hikma al-masˇriqiyya? read at the International Conference on ‘The Arabic, ˙ Hebrew and Latin Reception of Avicenna’s Metaphysics’ (Villa Vigoni, Loveno di Menaggio, 3 – 5 July 2008). These two interpretations were first discussed in Nallino, Filosofia ‘orientale’ od ‘illuminativa’. See Szpiech, In Search of Ibn Sı¯na¯’s Oriental Philosophy.

178 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Mauro Zonta

structure and contents of The Oriental Philosophy, as well as what is probably the correct meaning of its title, have been recently and excellently examined in detail by Dimitri Gutas.3 I will suggest some possible improvements to a part of Gutas’s study: the contents of the metaphysical section of The Oriental Philosophy, which, as far as I know, have not yet been studied. What are the contents of Avicenna’s Oriental Philosophy? It seems that in this work Avicenna himself wanted not to give a summary of Aristotle’s philosophy but to examine some of the most discussed points of it. Judging from the already published parts of Avicenna’s work as well as from what Avicenna himself states in the introduction to it,4 The Oriental Philosophy should have included a number of sections, as follows: 1. A section on logic (mantiq), including the contents of Porphyry’s Eisagoge ˙ and of Aristotle’s De interpretatione, Prior Analytics and Posterior Analytics (the last part being no longer extant). According to Gutas, also such parts of logic as dialectics, sophistics, rhetoric and poetry should have been found in the complete original version of this section; 2. A section on metaphysics, which according to Avicenna’s intention should have included two parts: one on the so-called ‘universal science’ (‘ilm kullı¯) and the other one on theology (‘ilm ’ila¯hı¯). According to Amos Bertolacci, this division seems to reflect the substantial originality of the metaphysics of The Oriental Philosophy with respect to other metaphysical works by Avicenna, which are more adherent to Aristotle’s views;5 3. A section on physics (‘ilm tabı¯‘ı¯), including the contents of natural ˙ sciences according to medieval Arab-Islamic philosophical tradition, i. e. physics proper, the nature of heavens, generation and corruption of things, actions and passions of natural things, meteorology, psychology, and probably zoology; 4. A section on ethics, called ‘practical science’ (‘ilm ‘amalı¯). According to Gutas, this part of the work included neither ethics proper, nor economics, but only what concerns ‘prophetic legislation’, i. e. man’s salvation in the afterlife. The Oriental Philosophy, then, was divided into four sections. But Avicenna’s work is apparently not preserved as a whole. How much of each of these sections is definitely extant, and what has already been published? As for the logical section, most of it has been preserved in some manuscripts and has been published a number of times: the first edition of it, under the fictitious title Mantiq al-masˇriqiyyı¯n (The Logic of the Oriental People), appeared ˙ 3

4 5

See Gutas, Avicenna’s Eastern (‘Oriental’) Philosophy. In this article, Gutas improves on his treatment of the philosopher in his previous book: Gutas, Avicenna and the Aristotelian Tradition, pp. 115 – 30 (see also pp. 43 – 9 for an English translation of the introduction of The Oriental Philosophy). Gutas, Avicenna’s Eastern (‘Oriental’) Philosophy, pp. 168 – 9. See Bertolacci, The Reception of Aristotle’s ‘Metaphysics’ in Avicenna’s ‘Kita¯b al-Sˇifa¯’’, pp. 305 and 605.

Possible Hebrew Quotations 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

179

in Cairo in 1910. The contents of this section appear to be partially different from those of other logical works by Avicenna.6 As for the physical section, it seems that the zoological part is lost; the other parts are found in some Arabic manuscripts. These have been published in 1993 in a Ph.D. thesis,7 and a critical edition of them should appear soon.8 By perusing these manuscripts, Gutas has concluded that, in contrast to the logical section of The Oriental Philosophy, most of the physical section of it is ‘copied verbatim from the Sˇifa’’. 9 This fact does not exclude a number of minor alterations with respect to the contents of the corresponding section of Avicenna’s Cure (al-Sˇifa¯’), some of which have been identified and studied by Gutas himself.10 As for the ethical section, apparently nothing of it has been found until now, although the possibility that some passages from it might be still extant cannot be totally excluded. As for the metaphysical section, as well, the question of the survival of at least some fragments of it appears to be open, since some apparent quotations are found in late medieval philosophic-theological works. The author of these works is Avner of Burgos, a Spanish Jewish philosopher and theologian, living in Castilia between 1270 and ca. 1345; he converted to Christianity (probably about 1320), assumed the name of ‘Alfonso de Valladolid’, and disputed about this fact with another Spanish Jew, Isaac Polgar.11 Avner wrote in Hebrew a number of works on religious and philosophical questions; some of them have been lost in their original text, but are still extant in medieval Castilian versions. The existence of Avner’s quotations of Avicenna’s Oriental Philosophy was first noted by Walter Mettmann, the critical editor of all the extant Castilian translations of his works, who, in a marginal note, pointed out this fact.12 However, Mettman did not examine these quotations in detail, nor did he study 6 See Gutas, Avicenna’s Eastern (‘Oriental’) Philosophy, p. 177: ‘For the Logic part of alHikma al-masˇriqiyya, Avicenna apparently wrote the text anew. Although it is quite ˙ parallel to his numerous other compositions on logic, there is no sustained literal correspondence with – i. e., copying from – earlier works’. 7 zcan, ˙Ibn Sina’nin el-hikmetu’l-mes¸rikiyye. I owe a copy of this work to the kindness of Jules Janssens. 8 Prof. Fehrullah Terkan (University of Ankara) is preparing a critical edition of the physical section of Avicenna’s Oriental Philosophy, based upon the extant six manuscripts of it (email of the author, July 8th, 2009). I am very grateful to Prof. Terkan for this kind information. It should be pointed out that the almost entire part of this section about optics has been translated into English in Hasse, Avicenna’s ‘De anima’, pp. 107 – 23. 9 Gutas, Avicenna’s Eastern (‘Oriental’) Philosophy, p. 177. 10 Gutas, Avicenna’s Eastern (‘Oriental’) Philosophy, pp. 174 – 7. 11 On Avner of Burgos, see the recent collection of bibliographical data found in Roth, Dictionary of Iberian Jewish and Converso Authors, pp. 121 – 4. 12 See Alfonso de Valladolid, Mostrador de Justicia, vol. I, p. 160, n. 251.

180 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Mauro Zonta

their close relationship, if any, to Avicenna’s philosophy and thought and to his possible sources. Gutas too knew the existence of them, but affirmed that they might not even be by Avicenna, since no trace of them was apparently found in any of Avicenna’s other works; finally, he left open the question as to whether The Oriental Philosophy was known in Andalusia, or in Maghreb at least.13 The four extant works by Avner of Burgos, in which Avicenna’s Oriental Philosophy is more or less explicitly quoted, possibly from the metaphysical section of it, are as follows: - Replies to the Critic (whose Hebrew original title was Tesˇuvot la-Meharef ); ˙ - Offering of Zeal (Oferta de Zelos, in Hebrew Minhat Qena’ot); ˙ - Book of the Law (Libro de la Ley, whose original Hebrew title might have been Sefer ha-Torah or Sefer ha-Misˇpat); ˙ - Master of Justice (Monstrador de Justicia, in Hebrew Moreh Sedeq). ˙ The first of these works is found both in its original Hebrew text, and in its medieval Castilian translation, which might have even been made by the author himself. The other three works are found in Castilian translations only. While the Castilian versions of all four works were published by Mettmann in his critical edition of them, the Hebrew text of the Replies to the Critic has been edited in a still unpublished Ph.D. thesis only.14 I have consulted the text and copied the relevant passages directly from the original unique manuscript of Parma, Biblioteca Palatina, parmense 2440.15 I will examine the passages (twelve in all) as found in their original Hebrew texts (if extant) or in their Castilian versions, together with my English translation of them. I will briefly discuss the contents of each passage, pointing out some possible similarities to corresponding passages of Avicenna’s extant works, and to the section of The Cure devoted to metaphysics in particular. Finally, I will point out some possible correspondences with ideas and passages found in Indian philosophical and religious texts, in particular the Veda, and with some elements of Indian thought. A number of them are found in a famous Arabic source contemporary to Avicenna: the Book of Enquiry about Indian Things (Kita¯b tahqı¯q ma¯ li-l˙ 13 See Gutas, Avicenna’s Eastern (‘Oriental’) Philosophy, pp. 171 – 2: ‘I have been informed [by Charles Manekin (University of Maryland)] that the 13th-14th century Rabbi Abner of Burgos, who converted to Christianity, mentions in some of his works that survive in Spanish Avicenna’s al-Hikma al-masˇriqiyya (filosofia oriental)’; however, according to ˙ Gutas, ‘the passages mentioned by Rabbi Abner as coming from Avicenna’s Eastern philosophy look suspect and cannot be readily identified in any of the extant portion of the work’. 14 Hecht, The Polemical Exchange Between Isaac Pollegar and Abner of Burgos (non vidi). 15 Two reproductions of this manuscript are found in the Institute for Microfilmed Hebrew Manuscripts of the National Library of Israel: the microfilm F 13444, and the CD Rom 149.

Possible Hebrew Quotations 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

181

Hind) by the Persian astronomer Abu¯ Rayha¯n al-Bı¯ru¯nı¯ (973 – 1048),16 with whom Avicenna was in correspondence.17

Quotations 1. Alfonso de Valladolid (Abner of Burgos), Tesˇuvot la-Meharef. Spanische ˙ Fassung, ed. W. Mettmann, Opladen: Westdeutscher Verlag, 1998: Passage 1.1., page 33, lines 25 – 7: Et como el ssabio Avicena le [i.e. Dios] asemej al sol quando fablava en aquellas luminarias divinales en la ‘Philosophia Oriental’. E l llmale sienpre ‘vierbo de Dios’. MS Parma, Biblioteca Palatina, parmense 2440, fol. 20v, lines 19 – 21: U-kemo ˇse-ha-hakam Ben Sina’ hemsˇilo ’el ha-sˇemesˇ, / ke-sˇe-hayah ˙ medabber be-’otan ha-me’orot ha-’elohiyyot ba-Filosofia’ ha-Mizrahit.18 / We˙ hu’ qore’ ’oto tamid ‘devar ’Elohim’. English translation: ‘Just like the sage Avicenna compared him (i. e. God) to the sun, when he speaks about those divine luminaries in The Oriental Philosophy; and he always calls him “God’s word”’. This passage is very similar to passage 3.2. A comparison not of God, but of the separate active intellect (‘aql bi-l-fi‘l) to the sun is found in Avicenna’s Cure, On the Soul, treatise 5, chapter 5, first lines, where Avicenna affirms that ‘its (i. e. the separate active intellect) relation to our souls is like the sun’s relation to our 16 The below quotations are taken from al-Bı¯ru¯nı¯’s India; see also the English translation of al-Bı¯ru¯nı¯’s work by Edward C. Sachau in Alberuni’s India. 17 The exchange of eighteen letters between Avicenna and al-Bı¯ru¯nı¯, apparently going back to ca. 1000, has been published in Nasr and Muhaqqiq, Abu¯ Rayha¯n Bı¯ru¯nı¯ wa-Ibn Sı¯na¯ ; ˙ ˙ Ibn Sina – al-Biruni see also their commented English translation in˙ Berjak and Iqbal, Correspondence. I owe a copy of the former to the kind courtesy of Amos Bertolacci. The letters do not include any clear reference to the subjects treated here by Avicenna. However, these subjects might have been discussed in a later correspondence, now apparently lost, which should go back to the period when al-Bı¯ru¯nı¯ was preparing his book on India (finished in 1030) and Avicenna was writing The Oriental Philosophy (ca. 1027 – 9, according to Gutas, Avicenna and the Aristotelian Tradition, p. 128). 18 See the Hebrew version of Ibn Tufayl’s work, in MS Parma, Biblioteca Palatina, parmense 2442, fol. 54v, lines 24 – 5,˙ where he ascribes to Avicenna the statement that: mi-heqer ’or ha-’emet … ke-’ilu hayu beraqim, ‘from the examination of the light of the ˙ (i. e. God) (there are things) … like lightnings’. See also the Arabic text in ’Abu¯ Truth Bakr ibn Tufayl, Hayy ibn Yaqza¯n, p. 17, line 22: min ’ittila¯‘i nu¯ri al-Haqqi … ka˙ ¯ qan. ˙ ˙ ˙˙ ˙ ’annaha¯ buru

182 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Mauro Zonta

visions’.19 The phrase ‘vierbo de Dios’, here ascribed to Avicenna himself,20 is a translation of the Hebrew davar ’elohim, which in its turn might be either a literal translation of the Arabic kala¯m ’Alla¯h, ‘God’s speech’,21 or even an imperfect translation of the Arabic ’amr ’ila¯hı¯, ‘divine thing’ (or ‘divine order’); in the latter case, this translation might result from Avner’s tentative adaptation of the concept to the Christian idea of Jesus as ‘God’s speech’ (John 1,1). However, the latter concept is found, e. g., in a passage of The Cure, Metaphysics, treatise 10, chapter 1 (about the typical Avicennian theme of al-mabda’ wa-lma‘a¯d, ‘the beginning and the return’ of the human soul from and to God), where a reference to the ‘first divine decree and command’ (al-’amr al-’ila¯hı¯ al’awwal) is found.22 This fact, together with the reference to the ‘divine luminaries’ (which is not found in On the Soul, but seems to concern a metaphysical question here), might suggest that the passage was taken from, or at least inspired by, the metaphysics of The Oriental Philosophy. Passage 1.2., page 38, lines 17 – 26: Et escrivi el Avicena en la ‘Philosophia Oriental’ que el omne conplido ha inteligenÅia separada; que si pudiesse ser que se partisse la inteligenÅia agente, diriemos que esta es parte della. E sinon porque fue nueva despues de que non era, diriemos que era aquella misma, e sinon porque se apropri en su cuerpo quando sse ffizo de nuevo, diriemos que non se fizo de nuevo. Fasta aqui fue su palabra.23 Et dio a entender en esto que la novedat que ha la inteligenÅia del omne separada es apropriarsse la inteligenÅia agente en su cuerpo. E como lo que dixo que la ssemejanÅa de la luz del sol proporÅionada al cuerpo que alunbra, que si sse toliesse el cuerpo, tol[i]erse-y´a su luz, de ssin que se toliese la luz. MS Parma, Biblioteca Palatina, parmense 2440, fol. 23v, lines 5 – 13: U-katav Ben Sina’ ba-Filosofia’ / ha-Mizrahit ˇse-ha-’adam ha-sˇalem yesˇ lo ˙ ´ekel ha-po‘el hayyinu ’omerim ´sekel nivdal, ˇse-’illu hayah / ’efsˇar ˇse-yithalleq ha-s ˙ 24 ˇse-hu’ heleq / mimmennu; we-luley ˇse -nithaddesˇ aharey ˇse-lo’ hayah hayyinu ˙ ˙ ˙ 19 See Avicenna, De Anima, p. 234, line 19: wa-nisbatuhu¯ ’ila¯ nufu¯sina¯ nisbatu al-sˇamsi ’ila¯ ’absa¯tina¯. 20 On˙ the contrary, on passage 3.2. the same phrase is explicitly ascribed not to Avicenna, but to Plato. See below, p. 189. 21 See Avicenna, Metaphysics, p. 359, line 6 of the English translation = line 2 of the Arabic text. 22 See Avicenna, Metaphysics, p. 363, lines 5 – 6 of the English translation = lines 3 of the Arabic text. 23 The sentence ‘fasta aqui fue su palabra’, which seems to be a literal translation of the typical Hebrew expression ‘ad ka’n devarayw, ‘up to this point, his words’ has no correspondence in the Hebrew text, as it is found in the Parma manuscript. Very probably, it was erroneously omitted by the copyist of the Hebrew manuscript. 24 In the manuscript: ke-sˇe, ‘when’.

Possible Hebrew Quotations 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

183

’omerim ˇse-hu’ / hu’ be-‘asmo, we-luley ˇse-hityahed be-gufo ke-sˇe-nithaddesˇ ˙ ˙ ˙ hayyinu ’omerim / ˇse-lo’ nithaddesˇ.25 Ki hodia‘ ba-zeh ˇse-ha-hiddusˇ ’asˇer la-s´ekel ˙ ˙ ha-nivdal / ˇsel ha-’adam hu’ hityahed ha-s´ekel ha-po‘el be-gufo, u-kemo ˇse˙ ’amar / ba-s´ekel ’or ha-sˇemesˇ ha-ne‘erak ’el ha-guf ˇse-zoreah ‘alayw ˇse-’illu ˙ hayah / sar ha-guf hayah sar ’oro mibli ˇse-yasur ha-’or. English translation: ‘Avicenna wrote in The Oriental Philosophy that the perfect man has a separate intellect. If it might be possible that the active intellect be divided into parts, we would say that it is a part of it; were it not that it is created ex novo, we would say that it is identical per se; were it not that it has not become proper (hityahed) to his body when it is created, we would say that it is not created.26 ˙ His words go up to this point. And he (i. e. Avicenna) means that the creation of man’s separate intellect is the fact that the active intellect becomes proper (hityahed) to his body,27 like he (i. e. Avicenna) says that the intellect is the light ˙ of the sun arranged for the body on which it shines, and if the body goes away, its light (too) goes away, but the light (in itself ) does not go away’. Something partially similar to the doctrine found in the first sentence (‘the perfect man has a separate intellect’) can be found e. g. in Avicenna’s Cure, On the Soul, treatise 5, chapter 6, as follows: ‘But if one escapes from the body and from the appendages of the body, then it is possible for him to come into perfect contact with the active intellect’28 – although this passage does not refer to the presence of the active intellect inside the human body. However, the general connection between this concept and what is found in the following statements (as well as in what appears to be Avner’s personal comment on them) is not found in this passage, nor in any other passage of The Cure. One might suggest 25 As pointed out by Szpiech, a similar passage is found in the Hebrew version of Ibn Tufayl’s work. See MS Parma, Biblioteca Palatina, parmense 2442, fol. 158v, lines 24 – 8: ˙ ar kak ra’ah le-‘asmo ‘asmut nifredet, lo’ ‘avad ˇse-yihyeh qesat ‘asmut ba‘al ha-sˇiv‘im ’elef Ah ˙ ˙ ˙ qesatah. … We-luley ˇse-zo’t ha-‘as˙ mut˙ hit haddesˇah ’ahar ˇse-lo’ panim, ’amarnu ˇse-hayyinu ˙ hayetah ’amarnu ˇse-hi’ hi’, we-luley hityahedah be-gufo ’es˙el hiddusˇo˙ ’amarnu ˙ˇse-hi’ lo’ ˙ tithaddesˇ. This version literally depends upon the Arabic ˙text˙in ’Abu¯ Bakr ibn Tufayl, ˙ ˙ Hayy ibn Yaqza¯n, p. 85, lines 8 – 9, 12 – 13: tumma ˇsa¯hada li-nafsihı¯ da¯tan mufa¯riqatan, ˙ g˘a¯za ’an ˙tataba‘‘ada da¯tu al-sab‘ı¯na alfi ¯wag˘hin, la-qulna¯ ’annaha¯¯ ba‘duha¯. … Walaw ˙ a¯ta¯ hadatat ba‘da ’an lam takun, la-qulna¯ ’annaha¯ hiya, ˙ lawla¯ ’anna ha¯dihi al-d wa-lawla¯ ¯ ¯ ¯ ˙ ’ihtisa¯suha¯ bi-badanihı¯ ‘inda hudu¯tihı¯, la-qulna¯ ’annaha¯ lam yahdut. ¯ in the author’s intentions, ˙three ¯ opinions which are ˙ ˙ above ˙ 26 The three points seem˙to be, opposite to his own. 27 This sentence seems to be not part of the quotation, but Avner’s personal interpretation of the above passages, ascribed to Avicenna. 28 See Avicenna, De anima, p. 248, lines 6 – 7: fa-’inna halasa ‘an al-badani wa-‘awa¯ridi al˘ ¯ li˙ tama¯ma al-ittisa¯li. I owe˙ this badani fa-h¯ına’idin yag˘u¯zu ’an yattasila bi-l-‘aqli al-fa‘‘a ˙ ˙ ˙ information to Dag Nikolaus Hasse.

184 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Mauro Zonta

that here, in the first lines at least, Avner of Burgos is literally quoting an original passage of the apparently lost metaphysical section of Avicenna’s Oriental Philosophy. As a matter of fact, a very similar passage is found in Avicenna’s Notes on Some Critical Points of Aristotle’s De anima (al-Ta‘liqa¯t ‘ala¯ hawa¯ˇs¯ı kita¯b al-nafs li-’Arista¯ta¯lı¯s): ‘The complete perfection in knowledge is ˙ ˙˙ only via the actual conjunction with the active intellect’.29 A similar doctrine is found in the medieval Indian philosophical school of Sa¯mkhya, which was well-known to al-Bı¯ru¯nı¯.30 Al-Bı¯ru¯nı¯ ascribed to this school ˙ the doctrine of purusa, in origin ‘man’, but according to Sa¯mkhya ‘pure ˙ ˙ consciousness, soul’.31 In Sa¯mkhya, purusa is the first of a list of twenty-five ˙ ˙ 29 See Badawı¯, ’Arist¸u¯ ‘inda al-‘Arab, p. 95, lines 14 – 15: al-istikma¯lu al-ta¯mmu bi-l-‘ilmi ’innama¯ yaku¯nu bi-l-’ittisa¯li bi-l-fi‘li bi-l-‘aqli al-fa‘‘a¯li. 30 Al-Bı¯ru¯nı¯ translated an ˙Indian philosophical work from Sanskrit into Arabic: the 5thcentury PataÇjali’s Yogasu¯tra, whose Arabic version by al-Bı¯ru¯nı¯ has been edited in Ritter, al-Bı¯ru¯nı¯’s bersetzung des Yoga-Su¯tra, and translated into English and commented in Pines and Gelblum, al-Bı¯ru¯nı¯’s Arabic Version of PataÇjali’s Yogasu¯tra. The Yogasu¯tra was a product of the Yoga Indian philosophical school, although its contents seem to reveal traces of the Sa¯mkhya school too (see Encyclopedia of Indian Philosophies. Volume IV, pp. 165 – 6). Al-Bı¯ru¯nı¯ apparently knew other Indian philosophical works, among which there might be the Sa¯mkhyaka¯rika¯bha¯sya by the Sa¯mkhya Indian philosopher Gaudapa¯da ˙ can be suggested ˙ ˙ ˙ lines (ca. 500 – 600). This fact by a passage of al-Bı¯ru¯nı¯’s India, p. 63, 16 – 17, where al-Bı¯ru¯nı¯ lists among the Indian books known to him kita¯bu ‘amilahu¯ ˙ awra al-za¯hidu wa-‘urifa bi-’ismihı¯ wa … Sa¯nka ‘amilahu¯ Kapila fı¯ al-’umu¯ri alG ’ila¯hiyyati, ‘the book which composed Gauda(pa¯da) the monk and which is known in its ˙ Kapila on divine subjects’ (see also the name, and the Sa¯mkh(y)a which composed English translation in Alberuni’s India, p. 132). Here, al-Bı¯ru¯nı¯ might have separated the unique work by Gaudapa¯da into two, giving an incorrect name to what Gaudapa¯da ˙ sna, commented on: the Sa˙¯ mkhyaka¯rika¯ by the 4th-century Indian philosopher ¯I´svarakr ˙ ˙˙ which he might have identified as a lost work by one of the founders of the Sa¯m˙khya ˙ see school, Kapila. On the possible origin of this reference, see Alberuni’s India, p. 267; also Encyclopedia of Indian Philosophies. Volume IV, p. 210. A tentative list of al-Bı¯ru¯nı¯’s Indian philosophical sources is also found and examined in Govind, al-Beruni’s Observations on Indian Philosophical Concepts, pp. 15 – 17; see also Karmakar, Hindu Philosophical Literature Known to al-Beruni, p. 246. According to Govind, al-Bı¯ru¯nı¯ knew the works of a number of Indian philosophical schools (not only the Sa¯mkhya, the ˙ the text Yoga, and the Vedantic one, but also the Ca¯rva¯ka and the Pu¯rva-Mı¯ma¯msa¯); but ˙ mentioned by al-Bı¯ru¯nı¯ would be the Gaudapa¯daka¯rika¯ by the Vedantic Indian ˙ philosopher Gaudapa¯da (ca. 600?). However, Govind too (pp. 17 – 18) thinks that al˙ mistakes in his references to Indian philosophers and philosophies, Bı¯ru¯nı¯ made some since the list of Indian philosophical works whose knowledge is usually ascribed to alBı¯ru¯nı¯ appears to be not always clear to him. 31 See al-Bı¯ru¯nı¯’s India, p. 19, lines 19 – 20: fa-’innahum yusammu¯na al-nafsa Purusˇa, wama‘na¯hu¯ al-rag˘ulu, ‘since they (i. e. the Hindus) call the soul purusˇa, and its meaning is “man”’; see also Alberuni’s India, p. 40. About the various meanings of the term, see also Encyclopedia of Indian Philosophies. Volume IV, pp. 74 – 5.

Possible Hebrew Quotations 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

185

principles, and the third of them is buddhi, ‘intellect’.32 Of course, this does not mean that the affirmation that ‘the perfect man has a separate intellect’ was surely influenced by Indian philosophy; it means that a possible influence of this philosophy, as found in other passages (see in particular 3.1.), cannot be denied. Passage 1.3., page 38, lines 30 – 31: Et assi escrivi el Avicena en la ‘Philosophia Oriental’ que la verdaderia de la substancia del omne conplido, aquella es la verdaderia de Dios, sin ningun mudamiento nin contrariedat. MS Parma, Biblioteca Palatina, parmense 2440, fol. 23v, lines 17 – 19: U-ken katav Ben Sina’ ba-Filosofia’ / ha-Mizrahit ˇse-’amittat ‘esem ha-’adam ˙ ha-sˇalem hi’ ’amittat ‘esem / ha-’eloah beli ´sum ˇsinnuy we-hilluf.˙33 ˙ English translation: ˙ ‘So Avicenna wrote in The Oriental Philosophy that the truth of the substance of the perfect man is, without any alteration and substitution, the truth of the substance of God’. The contents of this passage are found in some other quotations of Avicenna’s Oriental Philosophy in Avner’s works (see the passages 2.2., 3.4. and 3.6.). No similar statement is found in The Cure; however, a possible connection between the perfect man and God can be found in The Cure, Metaphysics, treatise 8, chapter 6, where Avicenna states that ‘the necessary existent (wa¯g˘ib al-wug˘u¯d) per se must be the furnisher of all existence, and every perfection of existence … and all that is a necessary existent is true’.34 Although there is no explicit identification between God and the perfect man here, the reference to ‘all existent’ and to ‘every perfection of existence’ might apparently include a reference to the existence of perfect man too. This might justify the hypothesis that the passage quoted by Avner of Burgos was really taken from Avicenna’s metaphysical section of The Oriental Philosophy.

32 See Encyclopedia of Indian Philosophies. Volume IV, p. 49. 33 A similar passage is found in the Hebrew version of Ibn Tufayl’s work. See MS Parma, Biblioteca Palatina, parmense 2442, fol. 151v, lines 5 – 6:˙ ’amittat ‘asmuto hi’ ‘asmuto ˙ 22: (sic). See also the Arabic text in ’Abu¯ Bakr ibn Tufayl, Hayy ibn Yaqza¯˙n, p. 81, line ˙ ˙ ˙ haqı¯qatu da¯tihı¯ hiya da¯tu al-Haqqi. ¯ ¯ hiyya¯t,˙ p. 356, lines 7 – 9: wa¯g˘ibu al-wug˘u¯di yag˘ibu ’an yaku¯na li34 ˙Avicenna,¯al-Sˇifa¯’, al-’Ila da¯tihi mufı¯dan li-kulli wug˘u¯din, wa-li-kulli kama¯li wug˘u¯din … fa-la¯ ’ahaqqu ’idan min ˙ Metaphysics, wa¯g˘ibi al-wug˘u¯di; the above English translation is found in Avicenna, p. 284, lines 15 – 19 = lines 9 – 11 of the Arabic text.

186 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Mauro Zonta

Passage 1.4., page 39, lines 17 – 21: Et como escrivi el Avicena otrossi que las obras que vienen de los que han las formas non son suyas dellos segund verdat, mas son de obrador que obra con ellos las obras que son anonbradas a ellos. Et como el dicho del que dizia: ‘Fuy oyr al qui oy conmigo, e ver al qui vio conmigo’. MS Parma, Biblioteca Palatina, parmense 2440, fol. 24r, lines 13 – 17: U-kemo ˇse-katav Ben Sina’: ‘Od / ˇse-ha-pe‘ullot ha-be’ot min ba‘aley hasurot / ’eynan lahem ba-’emet, ’aval hen le-po‘el ˇse-yif‘al bahem ha-pe‘ullot / ha˙meyuhasot ’aleyhem, u-ke-ma’amar ha-’omer: “Hayyiti ˇsema‘ le-’asˇer / ˇsema‘ bi, ˙ u-re’ot le-’asˇer ra’ah bi”’. English translation: ‘As Avicenna wrote: “Since the possessors of the forms do not truly possess the actions coming from them, but those actions are in possession of the actor which performs through them the actions proper to them, according to what is said: I was (the sense of ) hearing to Him who heard through me, and sight to Him to saw through me”’. This passage is found in a wider form in the passage 2.1. I have not been able to find any other passage by or ascribed to Avicenna about this point.35 However, a rather similar concept is found in the Sa¯mkhya philosophy, as it is reported by ˙ Hindus) this is also the form of the al-Bı¯ru¯nı¯: ‘And according to them (i. e. the liberated man, who is not different from Him (i. e. God) in those (things), apart from his beginning, since he does not exist in the previous eternity like (Him), because of the fact that, before it, he is in the “place of confusion”, knowing the object known, and his knowledge is like an imagination, acquired via an endeavour’.36 One can also compare a passage of the Yogasu¯tra of PataÇjali, which al-Bı¯ru¯nı¯ translated into Arabic: ‘Pure awareness is just seeing, itself; although pure, it usually appears to operate through the perceiving mind’.37 35 The final quotation is taken from al-Buha¯rı¯’s collection of ’aha¯dı¯t : see ’Abu¯ Bakr ibn ˙ ¯ Tufayl, Hayy ibn Yaqza¯n, p. 56, n. 1. ˘ ˙ al-Bı¯˙ru¯nı¯’s India, ˙p. 39, lines 20 – 21: wa-ha¯da¯ ’aydan sifatu al-mutahallasi ‘indahum, 36 See ˙ ¯˙ lam yakun ˘ fı¯ ˙al-’azali alfa-la¯ yanfasila ‘anhu¯ fı¯ha¯ ’illa¯ bi-l-mabda’i fa-’innahu ˙ mutaqaddimi ka-da¯lika, min ’ag˘li ’annahu¯ ka¯na qablahu¯ fı¯ mahalli al-irtiba¯ki ‘aliman bi-l-ma‘lu¯mi wa-‘ilmuhu¯ ka-l-haya¯li muktasabun bi-l-’ig˘tiha¯di;˙ see also the English translation in Alberuni’s India, ˘p. 81. 37 This passage is found in Yogasu¯tra, 2, 20: drasta¯ dr´si-ma¯trah ´suddho ’pi pratyaya¯nupa¯syah. ˙˙ taken ˙ from Hartranft, The Yoga-Su¯tra of The text and the above English translation are PataÇjali. As for the partially different Arabic version by al-Bı¯ru¯nı¯, it is as follows: al‘a¯limu bi-g˙ayri ma‘lu¯min yaku¯nu fı¯ da¯tihi ‘a¯liman bi-l-quwwati wa-la¯ yahrug˘u ’ila¯ al-fi‘li ˘ lines 10 – 11). ’illa¯ bi-l-ma‘lu¯mi (Ritter, al-Bı¯ru¯nı¯’s bersetzung des Yoga-Su¯tra, p. 181, See the English translation in Pines and Gelblum, al-Bı¯ru¯nı¯’s Arabic Version of PataÇjali’s Yogasu¯tra, p. 525 (and see also p. 541, nn. 97 – 8): ‘The knower, without the known, is in his essence a knower in potentia: he is actualized only through the known’.

Possible Hebrew Quotations 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

187

2. Alfonso de Valladolid (Abner of Burgos), Ofrenda de Zelos (Minhat Kena’ot) ˙ und Libro de la Ley, ed. W. Mettmann, Opladen: Westdeutscher Verlag, 1990: In the Ofrenda de Zelos: Passage 2.1., page 51, lines 7 – 12: E esto es como lo que escrivi el Abicepna en la ‘Philosofia Oriental’, que las obras que vienen de las cosas fformadas, quiere dezir que an fformas con que obren, non son por ellas ssegund verdat, nin sson dellas, mas sson de obrador que obra con ellas las obras que son anonbradas a ellas, e como el dezir del que dixo: ‘Ffuy oyr del que oy comigo, e veer del que vio comigo’. E dixo: ‘Non ech quando ech, mas Dios es el qui ech’.38 English translation: ‘And this is how Avicenna wrote in The Oriental Philosophy, that the actions coming from formed things, i. e. those (things) having forms by which they act, are not truly by them, nor they are possessed by them, but are in possession of the actor which makes through them the actions proper to them, according to what is said: “I have heard what He (i. e. God) hears together with me, and have seen what He sees together with me”; and it (also) says: “I have not pushed when I have pushed, but God is who has pushed”’. As said before, this passage looks like a wider version of passage 1.4. In the Libro de la Ley: Passage 2.2., page 111, lines 7 – 9: Y esto segund lo que escrivi el sabio AviÅienna en la ‘Filosofia Oriental’ que la verdaderia del omne conplido, esso es la verdaderia de Dios sin ningun departimiento. English translation:

38 See the Hebrew version of Ibn Tufayl’s work in MS Parma, Biblioteca Palatina, parmense ˙ 2442, fol. 115v, lines 19 – 23: We-hitba’er lo ˇse-ha-pe‘ullot ha-mesuddarot mehem ’eynam lahem, we-’omnam hem le-po‘el yif‘al bam ha-pe‘ullot ha-meyuhasot ’aleyhem, we-zeh ha‘inyan ’asˇer husˇqaf lo hu’ ‘inyan ’omero: ‘Hayyiti ˇsema‘ ’asˇer ˇsema‘˙ bo, u-re’ot ’asˇer yir’eh bo’, we-’amar be-’El hasˇak: ‘Lo’ hisˇlakti bi hisˇlakti, ’aval ha-’El hisˇlik’. See the Arabic text in ’Abu¯ Bakr ibn T˙ufayl, Hayy ibn Yaqza¯n, p. 55, line 24–p. 56, line 3: wa-tabayyana lahu¯ ˙ a¯dira˙ ‘anha¯ laysat˙fı¯ al-haqı¯qati laha¯, wa-’innama¯ hiya li-fa¯‘ilin yaf‘alu ’anna al-’af‘a¯la al-s ˙ ˙ ¯ da¯ al-ma‘na¯ ’alladı¯ la¯ha lahu¯, huwa qawlu biha¯ al-’af‘a¯la al-mansu ¯ bata ’ilayha¯. Wa-ha ˙ ¯ yabsaru bihı¯’. WaRasu¯li ’Alla¯hi: ‘Kuntu sam‘ahu¯ ’allad¯ı yasma‘u¯ bihı¯, wa-basarahu¯ ’alladı ¯ ˙ ˙ fı¯ muhkami al-tanzı¯li: ‘Fa-lam taqtulu¯hum, wa-la¯kinna ’Alla¯ha qatalahum, wa-ma¯ ˙ ramayta ’id ramayta, wa-la¯kinna ’Alla¯ha rama¯’. The latter quotation is taken from alQur’a¯n, 8, 17.

188 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Mauro Zonta

‘This is according to what the sage Avicenna wrote in The Oriental Philosophy, that the truth of the perfect man is the truth of God, without any difference’. This passage seems to reproduce almost literally what is found in passage 1.3. 3. Alfonso de Valladolid (Abner of Burgos), Mostrador de Justicia, 2 vols., ed. W. Mettmann, 1994 – 6: Passage 3.1., in volume I, page 160, lines 2 – 6: E esto mostr el AviÅena encubiertamientre en la ‘Filosofia Oriental’, en que dixo que la sustanÅia del entendimiento obrador39 a ssetenta mill caras, e que en cada cara a setenta mill bocas, e en cada boca setenta mill lenguas, con que alaban ssienpre al Criador, e que aquella muchedunbre que paresÅe non es muchedunbre segund verdat.40 English translation: ‘Avicenna explains this in a hidden way in The Oriental Philosophy, where he affirms that the substance of the active intellect has seventy thousand faces, and each face has seventy thousand mouths, and in each mouth there are seventy thousand tongues, by which they always praise the Creator; and that abundance which appears is not the(ir) real abundance’. This passage appears to include a statement (the fact that here Avicenna speaks ‘in a hidden way’) and some images which might be interpreted as having an almost mystical character, so being apparently at variance with Gutas’s interpretation of Avicenna’s work.41 However, it might have been found either in the metaphysical section or, even better, in the ethical section of The Oriental Philosophy. In fact, one should remember that the main theme of the ethical 39 In this case, ‘entendimiento obrador’ appears to be a different Spanish version of the Hebrew term ´sekel po‘el, ‘active intellect’; in other cases, the same term has been differently rendered as inteligenÅia agente (see e. g. here above, passage 1.2.). 40 See the Hebrew version of Ibn Tufayl’s work, in MS Parma, Biblioteca Palatina, parmense 2442, fol. 158v, lines 17˙– 20: U-le-zo’t ha-‘asmut ˇsiv‘im ’elef panim, be-kol ˙ˇon mesˇabbehet ‘asmut ha-’emet panim ˇsiv‘im ’elef peh, be-kol peh ˇsiv‘im ’elef lesˇon, we-kol les ˙ ˇsov ˙ha-ribbuy weu-meqaddesˇah u-mehallelah, lo’ tanuah, we-ra’ah le-zo’t ha-‘asmut ’asˇer yah ˙ ˙ ˙ Yaqza¯n, p. 85, ’eyno ribbuy. See also the Arabic text in ’Abu¯ Bakr ibn Tufayl, Hayy ibn ˙ lines 2 – 4: Wa-li-ha¯dihi al-da¯ti sab‘u¯na ’alfi wag˘hin, fı¯ kulli wag˙˘hin sab‘u¯na ˙’alfi famin, ¯ ¯ na ’alfi ¯ lisa¯nin, tusabbihu biha¯ da¯ta al-Wa¯hidi al-Haqqi, wawa-fı¯ kulli famin sab‘u ¯ al-da¯ti ’allatı ˙ fı¯ha¯ al˙¯ li-ha¯dihi ˙ ¯ tu¯hamu tuqaddisuha¯ wa-yumag˘g˘iduha¯, la¯ tafturu. Wa-ra’ya ¯ ¯ katratu, wa-laysat kat¯ıratan. ¯ Gutas, Avicenna’s¯ Eastern (‘Oriental’) Philosophy, pp. 160 – 66. However, about the 41 See possible influence of Indian philosophies on medieval Islamic mysticism see Massignon, Essai sur les origines du lexique technique de la mystique musulmane, pp. 81 – 98.

Possible Hebrew Quotations 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

189

section, according to Gutas, should have been a treatment of ‘the life to come’ – a theme which might be found in this passage. Here, an apparent trace of the influence of Indian philosophical and theological tradition is found. In two fundamental Vedic texts, the Atharvaveda and the Rgveda, there are passages that include a rather similar idea: ‘A thousand ˙ Purusha, a thousand eyes, a thousand feet …. This Purusha is all that heads hath yet hath been and all that is to be’.42 An even more similar passage can be read in the Bha¯gavata Pura¯na, a Hindu religious text probably written only one or two ˙ centuries before Avicenna: ‘The Lord [Maha¯-Visnu], although lying in the Causal Ocean, came out of it, and dividing ˙ ˙Himself as Hiranyagarbha ˙ 43 with [Brahma¯], He entered into each universe and assumed the vira¯t-ru¯pa, ˙ 44 thousand of legs, arms, mouths, heads, etc.’. Of course, the author of this passage ascribed to Avicenna tried to adapt this concept to his religious believing, by identifying the many things that the Indian gods are with the many things that praise God. Passage 3.2., in volume I, page 160, lines 28 – 9: E assi Platon le llam ‘Verbo de Dios’, assi como el AviÅena le asemej al sol en la ‘Filosofia oriental’.45 English translation: ‘So Plato called him (i. e. God) “God’s word”, just like Avicenna compared him to the sun in The Oriental Philosophy’. This passage is very similar to the passage 1.1., quoted above. It should be noted that the sun was regarded to be one of the most important gods of Hinduism, especially under the name of A¯ditya. The sun is regarded by the Hindus as the main divine ‘light’ (jyotis), being son of Aditi, ‘infinity’, the father of the seven principal deities of Vedic Hinduism. This information might have come to Avicenna via al-Bı¯ru¯nı¯.46 Moreover, it should be pointed out that a similar idea 42 See Rgveda, X, 90, 1a and 2a (sahasras´¯ırsa¯ purusah sahasra¯ksah sahasrapa¯t … Purusa ˙ sarvam yad bhu¯tam yac ca bhavyam), ˙ as translated ˙ evedam into˙ English in The Hymns ˙of the Rgveda, vol. II, p. 517. See also the partially corresponding passage in Atharvaveda, XIX,˙ 6, 1. 43 I.e. ‘excellent form’. 44 See Bha¯gavata Pura¯na, II, 5, 35 (sa eva purusas tasma¯d andam nirbhidya nirgatah ˙ ah sahasra¯nana-s´¯ırsava¯n), ˙ as translated ˙ into English by Swami sahasrorv-an´ghri-ba¯hv-aks ˙ ˙ Prabhupa¯da on www.srimadbhagavatam.org. 45 See the Hebrew version of Ibn Tufayl’s work, in MS Parma, Biblioteca Palatina, parmense 2442, fol. 158v, lines 21 – ˙2: u-ke-’ilu zo’t ha-‘asmut surat ha-sˇemesˇ. See also the Arabic text in ’Abu¯ Bakr ibn Tufayl, Hayy ibn Yaqza¯n, p.˙ 85,˙line 5: wa-ka-anna ha¯dihi ˙ ˙ ¯ ˙ al-da¯ta su¯rata al-sˇamsi. ˙ 46 About al-Bı¯ru¯nı¯’s knowledge of this concept, see e. g. al-Bı¯ru¯nı¯’s India, p. 56, lines 1 – 2: wa-min al-’asna¯mi al-masˇhu¯rati sanamu Mu¯lta¯n bi-’ismi al-sˇamsi wa-li-da¯lika summiya ˙ ˙

190 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Mauro Zonta

is found in the Vedantic Indian philosophy.47 Obviously, if the metaphysics of The Oriental Philosophy, as it was read by Avner, really included such reference, its author should have adapted the Indian polytheistic concept of the sun as a god to the Islamic concept of the unique God. Passage 3.3., in volume I, pages 234, last line-235, line 2: E la rrazon daquellos espejos e de aquellas fazes que nonbraron los sabios del Talmud, essa es misma la razon de los espejos e de las fazes que nonbr el Avicena en la ‘Filosofia oriental’.48 English translation: ‘And the reason of those mirrors and those faces mentioned by the wise men of the Talmud is the same reason of the mirrors and the faces mentioned by Avicenna in The Oriental Philosophy’. The contents of this passage appear to be rather similar to what can be found in the passage 3.1. The reference to the Talmud by Avner might in reality hint to a passage found in Genesis Rabbah, chapter 4, where the role played by the ‘big and little mirrors’ (in Hebrew, mar’ot and bevu’ot gedolot and qetannot) in the world is mentioned.49 As for the reference to Avicenna, one ˙ can compare Avicenna’s idea of the two faces of the human soul, as well as his mention of the thesis that the soul is like a mirror (although this thesis seems to be refuted by Avicenna himself ).50 The reference to ‘mirrors’ is found also in the Indian philosophy, as pointed out by al-Bı¯ru¯nı¯.51

47 48 49 50 51

’A¯ditya; see also the English translation by in Alberuni’s India, p. 116: ‘A famous idol of their (i. e. of Hindus) was that of Multn, dedicated to the sun, and therefore called ffditya’. See Encyclopedia of Indian Philosophies, Volume III, p. 24, where it is said that, according to the Vedantic philosophy, ‘the “way of the gods” (devaya¯na) … leads to the sun’. See ’Abu¯ Bakr ibn Tufayl, Hayy ibn Yaqza¯n, p. 86, line 20, where a reference to al-mara¯ya¯ ˙ ˙ the images’, ˙ is found. wa-l-suwar, ‘the mirrors and ˙ See Jastrow, Dictionary of Talmud Babli, vol. II, p. 835a. See Avicenna, De anima, p. 47, lines 17 – 18 (about the existence of two sides of human soul, the inferior and the superior ones) and p. 245, line 16 (where the statement that fataku¯nu al-nafsu ka-mir’a¯tu, ‘the soul is like a mirror’, is found). See al-Bı¯ru¯nı¯’s India, p. 22, lines 18 – 19 (’intiba¯‘u su¯rati al-sˇamsi wa-hiya wa¯hidatun fı¯ ˙ ¯ ’aw˙ miya¯hin masbu¯batin fı¯˙ ’awa¯nin ‘iddati mara¯ya¯ mansu¯batin ‘ala¯ muha¯da¯tiha ˙ mawdu¯‘atin ‘ala¯ muwa¯za¯tiha¯), as translated in Alberuni’s India, p.˙ 46: ‘The image of ˙ though he is only one, is represented in many mirrors which are placed opposite the sun, to him, as also in the water of vessels placed opposite’. This philosophical interpretation of the mirrors might come from the Veda, where the term a¯dars´a, ‘mirror’, as ‘ideal perfection’, is found. As a matter of fact, it should be noted that a very similar concept is discussed in the Upades´asa¯hasrı¯ by the well-known Vedantic Indian philosopher S´amkara ˙ (ca. 750). See Encyclopedia of Indian Philosophies. Volume III, pp. 242 – 3: ‘The reflection of a face is different from the face since it conforms to the mirror, whereas the face is different from the reflection because it does not conform to the mirror …. The reflection

Possible Hebrew Quotations 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

191

Passage 3.4., in volume II, page 16, lines 10 – 13: E segund esta lengua misma dixo AviÅena, como prov por sus palabras, que las intelligenÅias separadas no sson esso mismo que Dios, nin otre cosa; e otrossi que la substancialidat del omne conplido non es essa misma que la intelligenÅia separada, nin otra cosa. English translation: ‘And in the same way Avicenna said, as it is proved by his words, that the separated intellects are nothing but God, and nothing else; and also that the substance of the perfect man is nothing but the separated intellect, and nothing else’. What can be read here is partially similar to what is found in passage 1.3., and identical to what is found at the beginning of passage 1.2. Passage 3.5., in volume II, page 54, lines 15 – 19: E sta ffue la entenÅion de Aben Rrost ssin dubda en aquella palabra en que alabava aquel ayuntamiento divinal, como lo ovo Moysen e los otros omnes que llaman Åoffim, e como lo alabava el AviÅena en el ‘Libro de la Filosofia oriental’, donde aprendi Aben Rrost aquella unidad de Dios, que nonbr, ssegund pareÅe, en la manera de su fablar. English translation: ‘This was surely the intention of Averroes when he praised that divine conjunction, like that Moses and the other men called sofim had, and like that Avicenna praised in the Book of the Oriental Philosophy,˙where Averroes learned that unity of God which he called, as it appears, according to his own words’. Unfortunately, this passage seems to be too vague to be useful for our purpose here, and it is probably impossible to identify a similar passage in another work by Avicenna. However, it should be pointed out that an explicit reference to a very similar concept (that the Necessary Existence, i. e. God, is not separated from the celestial bodies), ascribed to Avicenna’s The Oriental Philosophy, is found in Averroes’s The Incoherence of the Incoherence (Taha¯fut al-Taha¯fut). 52 Has Avner taken this reference from Averroes, or directly from the work ascribed to Avicenna?

of the face in the mirror is a property of neither the face nor the mirror …’ (Upades´asa¯hasrı¯, II, 18, 32 and 37). See also a passage of the Hasta¯malakas´loka¯h by Hasta¯mala (ca. 800?), as translated into English by E. Harzer: yatha¯ darpana¯tha¯va ˙ also a¯tha¯maha¯nau mukham vidyate kalpana¯hı¯namekam, ‘when the mirror vanished then the reflection perishes; only the face remains, free from confusion, alone’ (Encyclopedia of Indian Philosophies. Volume III, p. 602). 52 See Averroes, Tahafot at-Tahafot, p. 421, lines 4 – 9.

192 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Mauro Zonta

Passage 3.6., in volume II, page 55, lines 9 – 11: La verdaderia de la sustancia del omne conplido es la verdaderia de Dios, ssin ningun demudamiento, como es provado en la ‘Filosofia oriental’. English translation: ‘The truth of the substance of the perfect man is the truth of God, without any difference, as it is proved in The Oriental Philosophy’. This passage is identical to passage 1.3.

Some Tentative Hypotheses In the footnotes to this article, I have quoted all the passages in Ibn Tufayl’s ˙ Living Son of the Watchful which are more or less similar to the passages explicitly ascribed by Avner to Avicenna’s Oriental Philosophy, basing myself on Szpiech’s article and a direct examination of the Ibn Tufayl’s work, both in the ˙ Arabic original and in the anonymous Medieval Hebrew version. As can be seen from a close linguistic comparison between the Arabic and the Hebrew, the latter is a very literal translation of the former, so that the evident analogies between those passages and the quotations ascribed to Avicenna by Avner cannot, in my view, rule out the possibility that both of them were directly based upon an Arabic text. Now, the question is: was this common Arabic source Ibn Tufayl’s work? Or was it the Arabic text of Avicenna’s Oriental ˙ Philosophy, which, if we assume that Ibn Tufayl’s and Avner’s statements are ˙ 53 true, would have been in its turn a common source of both of them? The second hypothesis might be confirmed by the fact that, after having quoted one of the above passages ascribed to Avicenna (n. 3.5.), Avner himself points out that a similar passage is found in Ibn Tufayl.54 In doing so, he apparently reveals ˙ that he had indeed read something of The Oriental Philosophy as ascribed to Avicenna and compared it to what he found in Ibn Tufayl. Such comparison, if ˙ 53 According to Gutas, Ibn Tufayl did not really know the contents of The Oriental Philosophy (see Gutas, Ibn˙ Tufayl on Ibn Sı¯na¯’s Eastern Philosophy), although he ˙ explicitly mentions it in two passages of his own work. However, it should be pointed out that part of Ibn Tufayl’s work was influenced by a work usually ascribed to Avicenna, which had the same ˙title (Living Son of the Watchful) and was well known in Medieval Jewish Spain, so that it was even rendered into Hebrew in a paraphrastic version by Abraham Ibn Ezra (see Abraham Ibn Ezra, Igeret Hay ben Mekitz, ed. I. Levin); can we really exclude that the same thing happened in the case of The Oriental Philosophy? 54 This reference is found in Alfonso de Valladolid, Mostrador de Justicia, vol. II, p. 54, line 30; see Szpiech, In Search of Ibn Sı¯na¯’s Oriental Philosophy, pp. 204 – 5. According to Szpiech, this reference would prove that the passage 3.5. was evidently taken from Ibn Tufayl’s introduction to his own work. ˙

Possible Hebrew Quotations 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

193

proved true, might even explain the evident parallelisms between Ibn Tufayl’s ˙ and Avner’s passages. Therefore, in my opinion, the question of the source of Avner’s quotations remains partially open, as it is possible that the text of The Oriental Philosophy, including at least the section on physics (as suggested by the existence of a Judeo-Arabic manuscript of it)55 and that on metaphysics (as suggested by Averroes’s reference mentioned here above, n. 3.5.), might have circulated among some Jewish scholars in Spain or in Maghreb during the Middle Ages. Some of the above quotations allegedly taken from Avicenna’s Oriental Philosophy found in Avner of Burgos’s works are about a similar subject (see e. g. nn. 1.2., 3.4. and 3.6.; 3.1. and 3.3.); some other are almost identical (see e. g. 1.1. and 3.2.; 1.3., 2.2. and 3.6.; 1.4. and 2.1.). Some of them are vague (see e. g. nn. 3.3. and 3.5.), while some others appear to be precise: Avner affirms that one of them at least (n. 1.2) is very explicit (as shown by the closing formula found in the Spanish version), so that it seems to be a literal translation of a passage of an original Arabic text ascribed to Avicenna. Although none of these quotations literally corresponds to any extant passage by Avicenna, there are similarities between some of them and some passages or terms of The Cure, in particular to statements found in On the Soul, treatise 5, and in the Metaphysics. This fact might confirm the possible Avicennian origin of these passages. As a matter of fact, although it is certain that the physical section of The Oriental Philosophy is almost identical to the corresponding section of The Cure, this does not mean that the same is true for the metaphysical section. Of course, the fact that Avner was writing works with theological and apologetical purposes might justify his employment of the metaphysical section of The Oriental Philosophy, where the theme of God’s nature was surely discussed, as it results from what Avicenna affirms in the introduction to his work.56 Finally, it should be pointed out that there are some apparent, although not literal, correspondences between the passages ascribed to Avicenna by Avner and some concepts of Hinduism and classical Indian philosophies, as found in the Veda and in the works by the Vedantic, Sa¯mkhya and Yoga school. The general ˙ ¯runı¯, as it results from his major contents of these texts were known by al-Bı ¯ book on India. Al-Bı¯ru¯nı¯, who exchanged a number of letters with Avicenna, might have transmitted to the latter some information about contemporary Indian thought, so being a possible cultural mediator between Indian philosophy and Avicennian one. These references to Indian philosophies, if they were confirmed, might even explain what could have been the original 55 This is the MS Oxford, Bodleian Library, Poc. 181 (Neubauer 1334), folios 61v–152v ; see Neubauer, Catalogue of the Hebrew Manuscripts in the Bodleian Library, c. 475. 56 See Gutas, Avicenna’s Eastern (‘Oriental’) Philosophy, pp. 167 – 8.

194 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Mauro Zonta

meaning of the title of this work, ascribed to Avicenna: The Oriental Philosophy, i. e. a philosophy more or less inspired by some ideas which came from an eastern place with respect to the Arabic world.

Bibliography al-Bı¯ru¯nı¯, Alberuni’s India. An account on the religion, philosophy, literature, geography, chronology, astronomy, customs, laws and astrology of India about AD 1030, 2 vols, transl. E.C. Sachau, London: Trubner and Co., 1888. –– , al-Bı¯ru¯nı¯’s India. An account on the religion, philosophy, literature, geography, chronology, astronomy, customs, laws and astrology of India about A. D. 1030, ed. E. C. Sachau, London: Trubner and Co., 1887 (repr. Leipzig: Harrassowitz, 1925). Alfonso de Valladolid (Abner of Burgos), Mostrador de Justicia, 2 vols, ed. W. Mettmann, Opladen: Westdeutscher Verlag, 1994 – 6. –– , Ofrenda de Zelos (Minhat Kena’ot) und Libro de la Ley, ed. W. Mettmann, Opladen: Westdeutscher Verlag, ˙1990. –– , Tesˇuvot la-Meharef. Spanische Fassung, ed. W. Mettmann, Opladen: Westdeutscher Verlag, 1998. ˙ Averroes, Tahafot at-Tahafot (L’incohrence de l’incohrence), ed. M. Bouyges, 3rd edn, Beirut: Dar el-Machreq, 1992. Avicenna, al-Sˇifa¯’, al-Ila¯hiyya¯t, ed. I. Madku¯r, Cairo: al-Matba‘a al-’Amı¯riyya, 1380/ ˙ 1960. –– , The Metaphysics of ‘The Healing’, transl. and ed. M. Marmura, Provo, Utah: Brigham Young University Press, 2005. –– , De anima (Arabic Text), ed. F. Rahman, London/New York/Toronto: Oxford University Press, 1959. ‘A. Badawı¯, ’Aristu¯ ‘inda al-‘Arab, al-Kuwayt: Waka¯la¯t al-matbu‘a¯t, 1978. R. Berjak and M.˙ Iqbal, Ibn Sina–al-Biruni Correspondence,˙Islam and Science, 1, 2003, pp. 91 – 8, 253 – 60; 2, 2004, pp. 57 – 62, 181 – 8; 3, 2005, pp. 57 – 62, 167 – 70; 4, 2006, pp. 197 – 212; 5, 2007, pp. 53 – 60. A. Bertolacci, The Reception of Aristotle’s ‘Metaphysics’ in Avicenna’s ‘Kita¯b al-Sˇifa¯’’. A Milestone of Western Metaphysical Thought, Leiden/Boston: Brill, 2006. V. Govind, al-Beruni’s Observations on Indian Philosophical Concepts with References to Their Christian, Greek and Islamic Parallels, Bharata Manisha, 4/3 – 4, 1978 – 9, pp. 13 – 25. R.T.H. Griffith, transl. and ed., The Hymns of the Rgveda, 2 vols, Varanasi: ˙ Chaukhamba Amarabharati Prakashan, 1971. D. Gutas, Avicenna and the Aristotelian Tradition, Leiden: Brill, 1988. –– , Avicenna’s Eastern (‘Oriental’) Philosophy. Nature, Contents, Transmission, Arabic Sciences and Philosophy, 10, 2000, pp. 159 – 80. –– , Ibn Tufayl on Ibn Sı¯na¯’s Eastern Philosophy, Oriens, 34, 1994, pp. 222 – 41. ˙ The Yoga-Su¯tra of PataÇjali. Sanskrit-English Translation and Glossary, on C. Hartranft, hrih.net/patanjali/archive/ysp-skrit-eng-chip-hartranft.pdf. D.N. Hasse, Avicenna’s ‘De anima’ in the Latin West. The Formation of a Peripatetic Philosophy of the Soul, 1160 – 1300, London/Turin: The Warburg Institute/Nino Aragno Editore, 2000.

Possible Hebrew Quotations 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

195

J.L. Hecht, The Polemical Exchange Between Isaac Pollegar and Abner of Burgos/Alfonso de Valladolid according to Parma MS 2440 ‘Iggeret Teshuvat Apikoros’ and ‘Teshuvot laMeharef ’, Ph.D. Thesis discussed at the New York University, New York, 1993. Ibn Ezra, Abraham, Igeret Hay ben Mekitz. A Critical Edition Supplemented with a Hebrew Translation of the Arabic Original Hay Ibn Yaqizian by Abu Abi Alhusain Ibn Abdalla Ibn Sina, ed. I. Levin, Tel Aviv: Katz Institute for the Study of Hebrew Literature, Tel Aviv University, 1983. Ibn Tufayl, ’Abu¯ Bakr, Hayy Ibn Yaqza¯n, ed. A. Nader, Bayru¯t: Da¯r al-masˇriq, 1993. ˙ ˙ Yerushalmi, Midrashic Literature and Targumim, M. Jastrow, Dictionary of˙ Talmud Babli, 2 vols, New York: Pardes Publishing House, 1950. R.G. Karmakar, Hindu Philosophical Literature Known to al-Beruni, Annals of the Bhandarkan Oriental Institute (Poona), 38, 1957, pp. 245 – 8. G.J. Larson and R. S. Bhattacharya, eds, Encyclopedia of Indian Philosophies. Volume IV. Sa¯mkhya. A Dualist Tradition in Indian Philosophy, Delhi: Motilal Banarsidass, 1987. L. Massignon, Essai sur les origines du lexique technique de la mystique musulmane, Paris: Vrin, 1954. S.H. Nasr and M. Muhaqqiq, Abu¯ Rayha¯n al-Bı¯ru¯nı¯ wa-Ibn Sı¯na¯, al-’as’ala wa-l-’ag˘wiba, ˙ ˙ 1352/1973. Tehran, C.A. Nallino, Filosofia ‘orientale’ od ‘illuminativa’ d’Avicenna?, Rivista degli studi orientali, 10, 1923 – 5, pp. 433 – 67. A. Neubauer, Catalogue of the Hebrew Manuscripts in the Bodleian Library, Oxford: Clarendon Press, 1886. A. zcan, ˙Ibn Sina’nin el-hikmetu’l-mes¸rikiyye adli eseri ve tabiat felsefesi, Ph.D. Thesis discussed at the Marmara University, Istanbul, 1993. S.L.A. Pandey, Comparative Study of Indian and Persian Philosophy, Islamic Culture, 33, 1959, pp. 81 – 7. S. Pines and T. Gelblum, al-Bı¯ru¯nı¯’s Arabic Version of PataÇjali’s Yogasu¯tra, Bulletin of the School of Oriental and African Studies, 29, 1966, pp. 302 – 25; 40, 1977, pp. 522 – 49; 46, 1983, pp. 258 – 304; 52, 1989, pp. 265 – 305. K.H. Potter, ed., Encyclopedia of Indian Philosophies. Volume III. Advaita Veda¯nta Up to S´amkara and His Pupils, Delhi: Motilal Banarsidass, 1981. H. Ritter, al-Bı¯ru¯nı¯’s bersetzung des Yoga-Su¯tra des PataÇjali, Oriens, 9, 1956, pp. 165 – 200. N. Roth, Dictionary of Iberian Jewish and Converso Authors, Madrid/Salamanca: Aben Ezra Ediciones, 2007. R. Szpiech, In Search of Ibn Sı¯na¯’s Oriental Philosophy in Medieval Castile, Arabic Sciences and Philosophy, 20, 2010, pp. 185 – 206.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

On the Latin Reception of Avicenna’s Metaphysics before Albertus Magnus: An Attempt at Periodization Amos Bertolacci Introduction

˘

The Latin Middle Ages are a relatively well-known area of the reception of Avicenna’s philosophy.1 For at least a hundred years, the precise mode of this reception has attracted scholarly attention and raised a lively debate in which different labels involving the name of Avicenna have been proposed to characterize philosophical authors and currents variously indebted to Avicenna’s thought. Thus, expressions such as ‘Avicennizing Augustinism’, ‘Latin Avicennism’, ‘Avicennizing Aristotelianism’, etc., are quite common.2 This proliferation of labels – in some cases very different from one another – can be taken as a symptom of a still immature stage of research; more positively, however, it also shows the multiplicity of modes and the different areas of the transmission of Avicenna in Latin. Although Avicenna’s philosophical writings did not enter the official curricula of medieval universities, and were therefore less frequently copied than Aristotle’s works, and never commented upon as such (with the exception of some parts of the section of the Sˇifa¯ on meteorology),3 they were extensively used by philosophers and theologians from the late twelfth century onward. Thus, the temporal scope of their influence surpassed the limits of the

1 2

3

I wish to thank warmly Dag Nikolaus Hasse for his insightful remarks on a first draft of the present article. These formulae were coined, respectively, by Gilson, Les sources greco-arabes, De Vaux, Notes et textes, and Van Steenberghen, La philosophie au XIIIe sicle, pp. 451 – 8. The expression ‘Avicennizing Boethianism’ is used to designate Gundissalinus’ epistemology in Fidora, Die Wissenschaftstheorie des Dominicus Gundissalinus, pp. 89 – 95. The Latin translation of three excerpts of the fifth section on natural philosophy of the Sˇifa¯ (taken from chapters I, 1 and I, 5), under the cumulative title of De mineralibus, was appended to the Latin translation of Aristotle’s Meteorologica. For this reason, this was by far the most often copied philosophical text by Avicenna in Latin translation (Kishlat, Studien, p. 53, counts 134 mss.; Schmitt, Knox, Pseudo-Aristoteles Latinus, p. 44, mention 148 codices). ˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

198

˘

˘

Middle Ages, and reached modern authors such as Descartes, Spinoza and Leibniz.4 Thus far, studies have focused mainly on the Latin reception of Avicenna’s psychology in the Kita¯b al-nafs of the Sˇifa¯ , whose translation into Latin (De anima) has been critically edited as first in the series Avicenna Latinus.5 The reception of some other parts of the Sˇifa¯ available to Latin medieval readers is comparable, in terms of diffusion and impact, to that of the De anima,6 but an overall study of their influence is still a desideratum. Avicenna’s metaphysics, as expressed in the Ila¯hiyya¯t of the Sˇifa¯ , is a case in point: a comprehensive history of the influence of its Latin translation (Philosophia prima) in the Middle Ages has yet to be written.7 Previous scholarship on the Latin reception of the Philosophia prima has provided insightful accounts of the influence of this work on single authors of the second half of the thirteenth and of the fourteenth century, such as Albertus Magnus (d. 1280), Thomas Aquinas (d. 1274), Henry of Ghent (d. 1293), John Duns Scotus (d. 1308), and others.8 The picture that emerges from these studies, however, is incomplete, if compared with the diffusion of Avicenna’s metaphysics both before and afterwards. The present contribution tries to fill the lacuna a parte ante by providing a tripartite periodization of the circulation of the Philosophia prima in Latin philosophy before the middle of the thirteenth century (§ 1), a detailed analysis of the first of these three periods (§ 2), and an account of the evidence attesting the first diffusion of Avicenna’s metaphysics in the University of Paris, shortly before its employment by William of Auvergne (§ 3). ˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Amos Bertolacci

4

5 6

7 8

On the reception of Avicenna’s philosophy after the Middle Ages, see, among other studies, Davidson, Proofs for Eternity, pp. 388 – 405 (‘Proofs of the existence of God as a necessarily existent being in modern European philosophy’); Gaskill, Was Leibniz an Avicennian?; Jolivet, L’pistmologie de Descartes; Hasnawi, La conscience de soi; Rashed, Thodice et approximation; Hasse, Arabic Philosophy and Averroism; Yaldir, Ibn Sı¯na¯ (Avicenna) and Ren Descartes (further bibliographical information on Avicenna and Descartes in Hasse, Avicenna’s ‘De anima’, p. 80, n. 5). The use of Avicenna’s De anima by Latin thinkers has been thoroughly investigated by Hasse, Avicenna’s ‘De anima’. Whereas the De anima is preserved in 50 known manuscripts, the De animalibus is attested by 33 codices, the Philosophia prima by 25, the Liber primus naturalium (chapters I–III, 1) by 22, the Logica by 13, and the De diluviis by 11 (see d’Alverny, Notes; Bertolacci, A Community of Translators, and the bibliography quoted therein). On the manuscript dissemination of the De mineralibus, see above, n. 3. The overviews of the Latin impact of the Ila¯hiyya¯t in Anawati, La Mtaphysique d’Avicenne, and Verbeke, Avicenna’s Metaphysics, are selective and cursory. See in this volume the contributions of Galluzzo, Hasse, Pickav, Pini and Richardson, and the further bibliography indicated therein.

On the Latin Reception of Avicenna’s Metaphysics before Albertus Magnus 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

199

§ 1 The Influence on Latin Authors until Albertus Magnus: Preliminary Remarks § 1.1 The Issue of Continuity Mainstream scholarship on the reception of the Philosophia prima has been governed by two main assumptions. The first is that the earliest significant recipient of this work, around the third or fourth decade of the thirteenth century, is William of Auvergne (d. 1249), who is regarded as a forerunner of the authors of the second half of the thirteenth century who fully display the influence of this work; on this view, Avicenna’s De anima – whose translation is coeval with that of the Philosophia prima – would have influenced Latin culture long before the Philosophia prima. This assumption, which can be traced back to 1926, the year of Roland-Gosselin’s edition of Thomas Aquinas’ De ente et essentia and Gilson’s first fundamental study on the ‘Avicennizing Augustinism’,9 posits a decided discontinuity between the translation of the Philosophia prima around the middle of the twelfth century and its full reception in the second half of the thirteenth. This hiatus has been variously explained. According to some, it was due to the initial resistance of traditional Latin metaphysics to the new Avicennian metaphysics.10 According to others, it was a consequence of the inner logic of Avicenna’s philosophy, in which the theory of knowledge naturally precedes metaphysics.11 According to a further explanation, it reflected a general shift in the cultural climate at the middle of the thirteenth century from more concrete, physiological issues, to more abstract, metaphysical concerns.12 The 9 In the doctrinal-historical studies on the principle of individuation and the distinction of essence and existence that complement his edition of the De ente et essentia of Thomas Aquinas (Introduction, Notes et tudes historiques), Roland-Gosselin takes William of Auvergne as the starting point of the Latin reception of Avicenna’s metaphysics on these two issues. In Pourquoi saint Thomas, conversely, Gilson regards psychology as the doctrinal core of the Latin reception of Avicenna’s philosophy from its very beginning. See the frequent references to Roland-Gosselin’s aforementioned study in Goichon, La philosophie d’Avicenne. 10 Jolivet, The Arabic Inheritance, pp. 130 – 31, remarks that Avicenna’s central distinction of essence and existence entered the philosophy and theology of Latin thinkers only in the second half of the 13th century, because until then its adoption was prevented by the key-notion of traditional Latin metaphysics (ultimately deriving from Boethius and instantiated by Gilbert of Poitiers), namely the idea of the complementarity of quo est (subsistence) and quod est (that which subsists). 11 Goichon, La philosophie d’Avicenne, p. 103: ‘Chronologiquement la thorie de la connaissance [d’Avicenne] fut la premire  influencer la pense occidentale’; ibid., p. 117: ‘Dans l’ensemble, l’tude d’Avicenne philosophe de l’Þtre est postrieure  l’tude d’Avicenne thoricien de la connaissance’. 12 Hasse, Avicenna’s ‘De anima’, contends that, contrary to Avicenna’s De anima, the Philosophia prima had greater impact in the second half of the 13th century than in the

200 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Amos Bertolacci

second assumption of previous scholarship on the reception of the Philosophia prima is that the impact of Avicenna’s philosophy in general, and of his metaphysics in particular, began to decrease when the Latin translations of Averroes’ commentaries on Aristotle started to gain success.13 These two assumptions look incompatible, since the first posits the startingpoint of the Latin reception of Avicenna’s metaphysics more or less when the second places the beginning of its decline. In fact, both are in need of some revision. As to the first, whatever the reasons adduced in its support, the thesis of a temporal gap between the translation of the Philosophia prima, on the one hand, and its reception, on the other, needs to be complemented by a more documented investigation: the available data show that the translation of this work and its subsequent employment by philosophers and theologians are two steps of a continuous process.14 The circulation of the manuscripts of the translation confirms this impression.15 The second assumption, likewise, does not match the available evidence: the diffusion of Averroes’ Aristotelian commentaries appears to have prompted not a progressive eclipse of Avicenna’s thought in Latin philosophy, but a better grasp of Avicenna’s philosophy, an outspoken acknowledgment of its value, and a strenuous defense of Avicenna’s positions against Averroes’ frequent and harsh criticisms. This attitude emerged first (p. 225), because at that time the interest of scholars shifted from physiology to more theoretical problems (p. 77). 13 Goichon, La philosophie d’Avicenne, p. 92: ‘[Le courant avicennien] s’affaiblit lorsque grandit l’influence d’Averros’; De Libera, D’Avicenne  Averros, p. 179: ‘pass l’ge d’or de la fine du XIIe sicle et de l’‘augustinisme avicennisant’ de la premire moiti du XIIIe, Ibn Sı¯na¯ s’efface devant Averros’. 14 The thesis of an earlier diffusion of Avicenna’s De anima with respect to the Philosophia prima seems to project on history the precedence of the study of the Latin impact of Avicenna’s psychology over that of his metaphysics in modern scholarship. The first fundamental study on the Latin reception of Avicenna’s De anima is Gilson, Pourquoi saint Thomas; the diffusion of the Philosophia prima has started to be studied later. De Libera, Penser au Moyen ffge, p. 112, rightly remarks that ‘Avicenne a t lu et exploit ds la fin du XIIe [sicle]’. 15 The most ancient extant manuscript of the translation dates from the first half of the 13th century (ca. 1240; see Van Riet, Traduction latine et principes d’dition, pp. 125*–6*); not being an archetype of the tradition, this codex attests the existence of previous manuscript circulation. The translation of a chapter of the work (III, 5) has circulated ˙ aza¯lı¯’s Summa, in a manuscript of the earlier, attached to the Latin translation of al-G th beginning of the 13 century copied in Spain, as well as in four other codices. D’Alverny, Les traductions d’Avicenne, pp. 154 – 5 (followed by Van Riet, Traduction latine et principes d’dition, p. 125*, and n. 12), surmises that the five manuscripts that include ˙ aza¯lı¯’s Summa) might reproduce a collection of Ila¯hiyya¯t III, 5 (placed at the end of al-G works constituted ‘ la source mÞme des traductions’, on account of the presence of the same collection in a number of distinct codices (a fact that indicates an ancient common ancestor) and of the heterogeneous character of the collection (a fact that suggests a dependence on the quaterni of the scholares coming to Toledo from all over Europe).

201

On the Latin Reception of Avicenna’s Metaphysics before Albertus Magnus

in the first half of the thirteenth century but continued in different ways also later. The replacement of Avicenna by Averroes as ‘Commentator’ of the Metaphysics and the other Aristotelian writings was gradual16 and did not imply a total dismissal of Avicenna’s philosophy, but only a change in the view adopted toward the latter. To summarize: in this as in other cases, historia – like natura – non facit saltus: an uninterrupted line of interpreters can be traced, which starts before William of Auvergne, and continues after the diffusion of Averroes’ commentaries. This line begins with the probable translator of the Philosophia prima (Gundissalinus) in the second half of the twelfth century, involves significant authors of the very beginning of the next century, such as John Blund and Michael Scot, passes through a series of fundamental figures of the first half of the following century, such as Robert Grosseteste in Oxford and William of Auvergne and Roger Bacon in Paris, and continues with Albertus Magnus and the other main authors of the second half of the thirteenth century. §1.2 A Three-fold Periodization

˘

The Latin reception of Avicenna’s metaphysics presents two main features. First, the Philosophia prima, i. e. the Latin translation of the Ila¯hiyya¯t of the Sˇifa¯ , is the only work of Avicenna by means of which Avicennian metaphysics was transmitted into Latin. Second, the fate of Avicenna’s metaphysics in Latin is closely related to the more or less parallel reception of Aristotle’s Metaphysics. These two features are mutually linked: since the Sˇifa¯ , by Avicenna’s own admission, is the summa of his in which the endorsement of Peripatetic philosophy is most evident, and the Ila¯hiyya¯t is a reworking of Aristotle’s Metaphysics, it is not surprising that the areas of diffusion of Avicenna’s and Aristotle’s work came to overlap. Moreover, these two traits are peculiar: they sharply distinguish, for example, the Latin side of the reception of Avicenna’s metaphysics from its Arabic counterpart, in which the success of Avicenna’s stance is not exclusively linked with the Ila¯hiyya¯t of the Sˇifa¯ , but is primarily connected with other works, and Avicenna’s metaphysics soon replaces, rather than interacting with, Aristotle’s work. ˘

˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

16 William of Auvergne, for example, still regards Avicenna as an expositor of Aristotle (De universo II, 8, in Opera omnia, vol. I, p. 690BH: ‘… et Avicenna post eum [sc. Aristotelem] … Similiter et alii expositores eiusdem Aristotelis’), and refers often in effect to Avicenna when quoting by name Aristotle (as noticed, among others, by Hasse, Avicenna’s ‘De anima’, p. 44 and n. 184; Teske, William of Auvergne’s Debt to Avicenna, pp. 154 – 5).

202 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Amos Bertolacci

The early Latin reception of the Philosophia prima is reconstructed here on the basis of the relationship that this work, in its different recipients and uses, holds with Aristotle’s Metaphysics. On the basis of this criterion, three main modalities of the reception of the Philosophia prima before Albertus, roughly corresponding to three chronological phases, can be distinguished. The first phase goes from the second half of the twelfth century, when the Philosophia prima was translated, until the beginning of the following century, when the first attestations of its use in European universities occur. This first phase is geographically centered, either directly or indirectly, in the Spanish city of Toledo.17 The second phase is documented since the beginning of the thirteenth century, whereas the third started around 1240: both phases were institutionally linked, in different ways, with the Universities of Paris and Oxford, although they followed distinct paths until Albertus Magnus. In the first phase (Gundissalinus; De causis primis et secundis; the anonymous treatise published by M.-T. d’Alverny in 1940 – 1942; Michael Scot’s writing on the classification of the sciences18), the Philosophia prima is both quoted and silently reproduced within independent treatises, of which it represents the main text, or one of the main texts, on metaphysics. Recourse to Aristotle’s Metaphysics, on the contrary, is absent or very scanty, since the Latin translations of this work have still a very limited diffusion. Besides Avicenna, the ˙ aza¯lı¯’s other metaphysical sources are works by Arabs and Jews (al-G metaphysics, the Liber de causis and Ibn Gabirol’s Fons vitae), although the Latin metaphysical tradition (Augustine, Boethius, Eriugena) is also influential. In the absence of the metaphysical text par excellence (Aristotle’s Metaphysics), the Philosophia prima, on account of its comprehensiveness and articulacy, performs the role of ‘vicarious’ canonical text. Averroes’ Long Commentary on the ‘Metaphysics’ is not yet available. The second phase (John Blund; Robert Grosseteste; William of Auvergne; Roland of Cremona; Roger Bacon) is marked by the joint consideration of Aristotle’s Metaphysics and Avicenna’s Philosophia prima by philosophers and 17 It is almost certain that the works of Gundissalinus and Michael Scot belonging to this phase were accomplished in Toledo. On the other hand, the anonymous Liber de causis primis et secundis (a work formerly ascribed to Gundissalinus) might have been written either in Toledo or in England (see below, n. 37), whereas the place of composition of the anonymous treatise published by D’Alverny might be either Toledo or Bologna (see below, n. 44). Cultural exchanges between Toledo and the rest of Europe were frequent at the time, as the cases of Gerard of Cremona and Daniel of Morley, among others, witness. 18 The thesis of Vicaire, Le Porrtains et l’avicennisme, according to which Avicenna would have influenced the school of Gilbert of Poitiers, is dismissed by Jolivet, The Arabic Inheritance, p. 131 and n. 58, pointing out that ‘when one finds in a twelfth-century writer an idea or a formula that recall … Avicenna, one must not immediately assume that he has been influenced by Avicenna’.

On the Latin Reception of Avicenna’s Metaphysics before Albertus Magnus 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

203

theologians in universities. Aristotle’s Metaphysics is now regarded as the main text on the subject, but Avicenna’s Philosophia prima represents the privileged way of access to Aristotle’s work and its main tool of interpretation. Traces of this tendency can be found in Robert Grosseteste in Oxford; its full development occurs, however, in philosophical and theological works produced in Paris. Here, the Philosophia prima is frequently mentioned together with Aristotle’s Metaphysics by masters of arts such as John Blund, and professors of theology such as William of Auvergne, Roland of Cremona and Roger Bacon: all these authors read the Metaphysics through the lenses of the Philosophia prima. The university of Paris documents a progressive acceptance of Avicenna’s work: initially used with no restriction in the arts faculty, as John Blund witnesses, and possibly involved in the Parisian condemnations of 1210 and 1215, the Philosophia prima was critically scrutinized, but also widely endorsed, by prime exponents of the faculty of theology such as William of Auvergne and Roland of Cremona, and enthusiastically received, with very few provisos, by Roger Bacon. Averroes’ Long Commentary on the ‘Metaphysics’, on the other hand, once it becomes available, is substantially ignored or even criticized. The third phase (Roger Bacon’s commentaries on the Metaphysics; Oxford commentators of the Metaphysics)19 attests the recourse to the Philosophia prima within the exegesis of the Metaphysics. Averroes’ Long Commentary on the ‘Metaphysics’ replaces Avicenna’s Philosophia prima in the role of authoritative interpretation of the Metaphysics. Yet, both in Oxford and in Paris, commentators of the Metaphysics continue to refer to the Philosophia prima, even though their references to Avicenna are much less frequent and systematic than those to Averroes’ Long Commentary. For the sake of brevity, these three phases can be labeled, respectively, ‘Philosophia prima without Metaphysics’, ‘Philosophia prima and Metaphysics’, ‘Philosophia prima in the exegesis of the Metaphysics’. They correspond grosso modo to three literary genres (independent treatises influenced by the format of the translation literature; philosophical and theological works produced in universities; commentaries) and to three modalities of the reception of the Philosophia prima (doctrinal endorsement; instrumental use for philosophical and theological purposes; occasional recourse for the explanation of Aristotle). From a sociological point of view, they are linked with different institutional contexts (non-universitarian centers of instruction; faculties of arts and faculties of theology within universities; faculties of arts only). Seen diachronically, they reflect an increasing assimilation of this work: the introduction of the Philosophia prima within the doctrinal debate in the first phase is followed by a period of critical evaluation, which allows the use of the main doctrinal points 19 The earliest extant Latin commentaries on the Metaphysics date from about 1240.

204 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Amos Bertolacci

of this work either in philosophical and theological writings in the second phase, or in the exegesis of the Metaphysics in the third phase. Obviously, the proposed periodization is not perfectly rigorous. The chosen arrangement, however, seems to provide a sufficiently coherent and systematic way of understanding the wide and complicated historical event under consideration.

§ 2 Philosophia prima without Metaphysics (Gundissalinus; De causis primis et secundis; Anonymous d’Alverny; Michael Scot) § 2.1 The Early Diffusion of the Philosophia prima and of the Latin Translations of Aristotle’s Metaphysics According to a widespread contention, the Philosophia prima was known in the Latin world before Aristotle’s Metaphysics. 20 This contention is substantially correct, although it is true with respect to the diffusion, rather than the composition, of the translations of the works under consideration. The Philosophia prima was translated into Latin between 1150 and 1175 in Toledo.21 Two Latin versions of the Metaphysics were produced before or at the same time of the Philosophia prima: the earliest Latin version of the Metaphysics, the so-called Translatio Iacobi sive Vetustissima by James of Venice (active between 1125 and 1150), and the translation called Anonyma sive Media, accomplished by an unknown author of the twelfth century.22 Thus, with regard to their composition, the translation of the Philosophia prima is not chronologically prior to that of the Metaphysics. 20 De Vaux, Notes et textes, p. 10, states that the works of Avicenna translated into Latin were ‘un ensemble comme on n’en possdait point d’autre alors, pas mÞme d’Aristote, dont les œuvres physiques et mtaphysiques n’arrivrant que plus tard et par tapes’; Goichon, La philosophie d’Avicenne, p. 90: ‘La Mtaphysique d’Avicenne a t connue un demi-sicle avant celle d’Aristote … La philosophie d’Avicenne … tait le premier ensemble de doctrine vraiment constitu qui parvint  l’Occident’; De Libera, Penser au Moyen ffge, p. 112: ‘le texte d’Avicenne est la premire grand œuvre philosophique qui soit parvenu en Occident’. 21 See Bertolacci, A Community of Translators. 22 Vuillemin-Diem, Praefatio, in Aristoteles Latinus XXV 1 – 1a, p. xxvi. The translation called Vetus, accomplished before 1230 (when it starts to be quoted), is just a revision of the Vetustissima in the form in which this latter is extant (see Vuillemin-Diem, ibid., pp. xxix–xxxii; Vuillemin-Diem, Praefatio, in Aristoteles Latinus XXV 3.1, pp. 4 – 5). Burnett, A Note on the Origins, advances a new hypothesis on the origin of the Media: according to him, this translation would have been composed in Antioch, in the second quarter of the 12th century.

On the Latin Reception of Avicenna’s Metaphysics before Albertus Magnus

If we consider, instead, the diffusion of the translations, the situation is different. At an unknown stage of its early transmission, the Vetustissima, originally more encompassing or even complete, underwent the loss of its second part (only the portion corresponding to A–C, 4, 1007a31 is fully extant, whereas excerpts of the following books may have survived as glosses in some manuscripts of the Media).23 On the other hand, the more comprehensive Media (books A–I, K–M) apparently had a limited circulation before the middle of the thirteenth century, when it started gaining diffusion.24 Signs of acquaintance with the Metaphysics can be detected in various authors and works of the second half of the twelfth century, but it is difficult to establish whether these quotations (which often do not concern specific passages of the Metaphysics or, if they do, do not refer explicitly to this work, or which concern doctrines that occur also elsewhere in the Aristotelian corpus) are first-hand or second-hand.25 Even before the translations, Latin scholars could draw information on the Metaphysics from the writings of Boethius and the quotations of these latter in subsequent authors (Abelard, Liber sex principiorum).26 The lack of a complete Greek-Latin translation of the Metaphysics in the first decades of the thirteenth century (due to the incompleteness of the Vetustissima and the late circulation of the Media) is confirmed by the immediate and wide success of the Arabic-Latin translation of the Metaphysics known as Nova, namely the collection of lemmata of Aristotle’s text taken from Michael Scot’s Latin translation of Averroes’ Long Commentary.27 23 See Vuillemin-Diem, Praefatio, in Aristoteles Latinus XXV 1 – 1a, pp. xxiv–xxv, Vuillemin-Diem, Praefatio, in Aristoteles Latinus XXV 3.1, p. 3. Since the two manuscripts preserving the Vetustissima in its incontaminated extant form (Avranches, Bibl. Munic., 232; Oxford, Bibl. Bodl., Seld. sup. 24) are of the 12th century, the loss of the second part of this translation probably occurred in this same century (‘schon frhzeitig verlorenen vollstndigeren Text’, Vuillemin-Diem, Praefatio, in Aristoteles Latinus XXV 3.1, p. 3). Thomas Aquinas might have quoted as alia littera (additional translation) in his commentary on the Metaphysics some fragments of the lost parts of the Vetustissima (see Reilly, The Alia Littera). 24 See Vuillemin-Diem, Praefatio, in Aristoteles Latinus XXV 2, pp. xxv-xxxiv; VuilleminDiem, Praefatio, in Aristoteles Latinus XXV 3.1, pp. 5 – 7. 25 See Vuillemin-Diem, Praefatio, in Aristoteles Latinus XXV 1 – 1a, pp. xv-xvi; VuilleminDiem, Praefatio, in Aristoteles Latinus XXV 3.1, pp. 31 – 2. Also the quotation of Aristotle in Gundissalinus’ De divisione philosophiae, despite its resemblance with a passage of the Metaphysics, is taken more probably from the Physics (see below, § 2.2). 26 See Vuillemin-Diem, Praefatio, in Aristoteles Latinus XXV 3.1, pp. 30 – 31; Speer, The Hidden Heritage. Burnett, The Blend of Latin and Arabic Sources, pp. 42 – 3, shows that themes of the Metaphysics are echoed in Abu¯ Ma ˇsar’s Great Introduction to Astrology (mid 9th century), translated into Latin by Hermann of Carinthia in 1141. 27 See Vuillemin-Diem, Praefatio, in Aristoteles Latinus XXV 3.1, pp. 7 – 8. This translation even contaminated the archetype of all the extant codices of the Metaphysica media (see Vuillemin-Diem, Praefatio, in Aristoteles Latinus XXV 2, pp. xxx–xxxi, xlii). ˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

205

206

All this implies that for a few decades – from the loss of the second part of the Vetustissima, sometime in the second half of the twelfth century, until 1220 – 1224, the probable date of Michael Scot’s translation of Averroes’ Long Commentary – the Philosophia prima might have been the only comprehensive account of Aristotelian metaphysics available to Latin philosophers. Later on, the diffusion of the Philosophia prima intersected with the spread of Aristotle’s Metaphysics, known first through the Translatio nova and Averroes’ Long Commentary, then through the Translatio media. § 2.2 Gundissalinus The influence of Avicenna on works of Gundissalinus (d. after 1190) such as the De anima, and on areas of his thought such as epistemology, has already been noticed.28 The Philosophia prima exerted a similar influence on his metaphysics. Gundissalinus is, so to say, ‘originally’ linked with the Latin transmission of Avicenna’s metaphysics: if we accept his traditional identification with Dominicus Gundisalvi, he was responsible, alone or in cooperation with another scholar, for the translation of this work into Latin.29 Thus, it is not surprising to find that at least two of his original works depend visibly on the Philosophia prima. 30 In the first of these, the De divisione philosophiae, the account of metaphysics – both in its themes and its structure – is based on continuous and extensive implicit quotations of Philosophia prima I, 1 – 3, thus reflecting all the main aspects of Avicenna’s preliminary characterization of the science of metaphysics.31 Since, in this context the silent citations of the

˘

28 Hugonnard-Roche, La classification des sciences; Hasse, Avicenna’s ‘De anima’, pp. 13 – 18; Fidora, Die Wissenschaftstheorie des Dominicus Gundissalinus. 29 The distinction of Gundissalinus (or Gundisalvus), author of original works, from ˙ aza¯lı¯ Dominicus Gundisalvi, the Latin translator of al-Kindı¯, al-Fa¯ra¯bı¯, Avicenna, al-G and Ibn Gabirol, proposed by Rucquoi, Gundisalvus ou Dominicus Gundisalvi?, is convincingly rejected by Fidora, Die Wissenschaftstheorie des Dominicus Gundissalinus pp. 14 – 18, and Hasse, The Social Conditions, p. 73 and n. 30. 30 Among the other works by Gundissalinus, the De scientiis (a treatise on the classification of the sciences probably antedating the De divisione philosophiae) is a paraphrase/ adaptation (not a bare translation, as sometimes it is portrayed) of al-Fa¯ra¯bı¯’s Ihsa¯ al˙˙ De ulu¯m (see Hugonnard-Roche, La classification des sciences, p. 41 and nn. 6 – 8). The unitate et uno relies mainly, on the one hand, on Boethius and Augustine, and, on the other hand, on Ibn Gabirol’s Fons vitae (see Jolivet, The Arabic Inheritance, p. 135). 31 Gundissalinus, De divisione philosophiae, apparatus fontium ad p. 35, 15-p. 42, 17. Jolivet, The Arabic Inheritance, p. 136, aptly contends that in this work ‘the influence of al-Fa¯ra¯bı¯ persists, but that of Avicenna is much more prominent’. According to A. Fidora, the influence of the Philosophia prima in Gundissalinus’ account of metaphysics in the De divisione philosophiae is limited to the discussion of the subject˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Amos Bertolacci

On the Latin Reception of Avicenna’s Metaphysics before Albertus Magnus

Philosophia prima are occasionally accompanied by equally silent citations of al˙ aza¯lı¯’s metaphysics,32 we may wonder whether Gundissalinus is not indirectly G relying on Avicenna’s authority also when, in other parts of the work, he ˙ aza¯lı¯ (as main source) together with Avicenna (as implicitly quotes al-G complementary evidence).33 Remarkably, in the account of metaphysics in the De divisione philosophiae Aristotle’s Metaphysics is never quoted.34 matter of this discipline (see Fidora, Die Wissenschaftstheorie des Dominicus Gundissalinus, p. 27, n. 21; p. 129; pp. 141 – 2, n. 51), whereas Gundissalinus’ Latin translation of al-Fa¯ra¯bı¯’s Ihsa¯ al- ulu¯m (De scientiis) would be the main source of the overall account ˙˙ Verhltnis, p. 72 and n. 16). Although al-Fa¯ra¯bı¯’s Ihsa¯ al- ulu¯m lies (see Fidora, Zum ˙˙ certainly in the background of Ila¯hiyya¯t I, 1 – 3 (see Bertolacci, The Reception, p. 464 and nn. 114 – 5), the latter rather than the former seems to be the main and direct source of the account of metaphysics in the De divisione philosophiae. For the influence of the Philosophia prima on other parts of the De divisione philosophie, see Fidora, Die Wissenschaftstheorie des Dominicus Gundissalinus, p. 155, n. 39. 32 See Gundissalinus, De divisione philosophiae, apparatus fontium ad pp. 36 – 7. 33 Ibid., apparatus fontium ad pp. 9 – 19 (Prologus), 20, 3 – 10 (Sciencia naturalis). 34 Aristotle is quoted in the prologue of the work, with no explicit mention of the work’s title, with regard to the tripartition of theoretical philosophy in physics, mathematics and metaphyics: ‘Unde Aristoteles: ideo scienciarum sunt species tres, quoniam una speculatur quod movetur et corrumpitur ut naturalis, et secunda quod movetur et non corrumpitur ut disciplinalis; tercia considerat quod nec movetur nec corrumpitur ut divina’ (Gundissalinus, De divisione philosophiae, p. 15, 12 – 15). Neither the apparatus fontium, nor the commentary of the edition (pp. 188 – 90), provides information on the exact provenience of the quotation. This renowned point of Aristotle’s epistemology, however, is reported by Gundissalinus differently than in the Metaphysics: the idea of corruptibility is totally absent in Metaph. E, 1, 1026a13 – 16, where the objects of the three theoretical sciences are distinguished according to their possession or lack of separation, on the one hand, and motion, on the other. Corruptibility and motion determine, at different levels, the tripartite classification of substances in Metaph. K, 1, 1069a30–b2, but this classification conveys only an epistemological bipartition (between two branches of physics and metaphysics, to the exclusion of mathematics) rather than a tripartition. Gundissalinus’ quotation resembles rather, in a reverse order, Aristotle’s tripartition of theoretical sciences in Phys. B, 7, 198a29 – 30, where metaphysics is portrayed as the science of immovable things, mathematics as the science of movable but incorruptible things, and physics as the science of corruptible things (I wish to thank Resianne Fontaine for having brought this point to my attention; cf. Gundissalinus, De divisione philosophie. ber die Einteilung der Philosophie, p. 68, n. 19). This quotation is markedly different from the report of Metaph. E, 1, 1026a13 – 16 in Boethius’ De trinitate II, p. 8, 5 – 18, or from the tripartition of the theoretical sciences in Avicenna’s ˙ aza¯lı¯’s Maqa¯sid al-fala¯sifa (Lat. transl. in Algazel’s Philosophia prima and in al-G ˙ Metaphysics, p. 2, 31–p. 3, 32). No specific tripartition of the theoretical sciences occurs in al-Fa¯ra¯bı¯’s Ihsa¯ al- ulu¯m or in the treatise De ortu scientiarum associated with the name ˙ Latin tradition. The term disciplinalis used in the quotation to indicate of al-Fa¯ra¯bı¯ in ˙the mathematics echoes the terminology of the Latin translations from Arabic, where this adjective renders two Arabic terms (riya¯d¯ı, ta lı¯mı¯) expressing mathematics (see, for ˙ example, Avicenna Latinus, Liber de philosophia prima sive Scientia divina, I–X, Lexiques, p. 204b). The quotation might therefore be indirect, depending on the Latin translation ˘

˘

˘

˘

˘

˘

˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

207

208 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Amos Bertolacci

In the second relevant work by Gundissalinus, the De processione mundi (an original cosmogonical treatise written after 1160), implicit quotations of Philosophia prima I, 6 – 7 can be found in the initial discussion of necessarium esse and possible esse.35 In this work, neither the Metaphysics nor any other Aristotelian writing is referred to, and Avicenna, together with Ibn Gabirol’s Fons vitae, is the main philosophical authority.36 § 2.3 De causis primis et secundis The anonymous De causis primis et secundis et de fluxu qui consequitur eas, a treatise on the procession of the world from the first causes through successive stages of emanation, dates to the end of the twelfth, beginning of the thirteenth century.37 Its massive dependence on Avicenna is witnessed by its presence (ascribed to Avicenna and under the title De intelligentiis) in the 1508 Venetian edition of Avicenna’s works. Its editor, R. De Vaux, regarded it as the first and clearest expression of ‘Latin Avicennism’.38 M.-T. d’Alverny and J. Jolivet stressed, more recently, its nature of synthesis between Latin authors (Augustine, Boethius and Eriugena, the latter conveying doctrines of Pseudo-Dionysius) and

35

36

37

38

of an Arabic text. Since the description of mathematics in the quotation applies properly to astronomy, the source of the quotation might be a Latin translation of an Arabic text on astronomy. De processione mundi, pp. 227 – 30 (§§ 7 – 13 of the list of sources, with reference to pp. 126 – 38 of the edited text; in § 7, the reference to Philosophia prima I, 7, 26 – 37, is in fact to I, 6, p. 44, 24 – 37). See Jolivet, The Arabic Inheritance, pp. 138 – 40; Soto Bruna, Estudio fil sofico, pp. 34 – 42. A detailed survey of the sources of the De processione mundi is available in The Procession of the World; Burnett, The Blend of Latin and Arabic Sources, pp. 52 – 60 and nn. 33 – 5, has amended the edited text by taking into account a source of the De processione mundi previously disregarded (Hermann of Carinthia’s De essentiis). See Jolivet, The Arabic Inheritance, p. 139 (‘But throughout this treatise the influence of the philosophers plays a major role … he exploits to the full two rich mines of speculative writing: Avicenna’s Metaphysics and the Fons vitae of Ibn Gabirol’); Soto Bruna, Estudio fil sofico, pp. 81 – 95. On the date, sources and doctrine of this treatise, see De Vaux, Notes et textes, pp. 63 – 80. D’Alverny, Deux traductions latines, pp. 129 – 30, supposes that ‘Master Maurice’ (d. 1238), archdeacon of Toledo and bishop of Burgos, might have been the author of this treatise, on account of his interest in Islamic theology and the philosophy of PseudoDionysius; later on (d’Alverny, Une rencontre symbolique, p. 177), she states that ‘les philosophes et thologiens anglais ont quelque responsabilit dans la diffusion du Pseudo-Avicenne, et peut-Þtre dans sa redaction’, since most of the manuscripts preserving the De causis primis et secundis are of English origin. See also Jolivet, The Arabic Inheritance, pp. 145 – 6, and Davidson, Alfarabi, Avicenna and Averroes on Intellect, pp. 210 – 11. De Vaux, Notes et textes, p. 12 (see above, n. 2).

On the Latin Reception of Avicenna’s Metaphysics before Albertus Magnus

Arabic sources, prominent among which is Avicenna.39 The De causis primis et secundis reproduces in different extents and contexts, for the most part implicitly, a large amount of the Philosophia prima: more precisely, chapters I, 1, II, 4, and III, 1 of its ontological part, and chapters VIII, 7 and IX, 2 – 5 of its theological part.40 Significantly, in its few explicit quotations the Philosophia prima is called Methaphisica or Liber de methaphisica. 41 This expression is applied also to Aristotle’s Metaphysics, which, however, is quoted indirectly through al-Fa¯ra¯bı¯.42 The use of the same title for both suggests that the author of the treatise regarded the two metaphysical works of Avicenna and Aristotle as one and the same writing. Remarkably, the theological part of the Philosophia prima is quoted jointly with the Liber de causis. 43 § 2.4 Anonymous d’Alverny The anonymous treatise on human nature and man’s afterlife that M.-T. d’Alverny discovered in MS Paris, BNF, lat. 3236 A, fols 85v–87, and published in 1940 – 42, dates to end of the twelfth century. Its place of composition might be either Spain or Northern Italy.44 D’Alverny describes it as ‘un des tmoins les plus curieux de la conjonction du no-platonisme arabe avec la culture

39 D’Alverny, Une rencontre symbolique; Jolivet, The Arabic Inheritance, p. 145. 40 The apparatus fontium of the edition is quite accurate, except for a few imprecisions (the reference to Philosophia prima II, 3 at p. 105, n. 1, is too vague to be considered a quotation; the same can be said of the reference to IX, 2 at p. 116, n. 2; the two quotations at p. 114, nn. 3 – 4, refer to c. IX, 5, rather than IX, 2, of the Philosophia prima). 41 Ps.-Avicenna Latinus, Liber de causis primis et secundis, p. 102, 19; p. 107, 8. 42 ‘Prima igitur creaturarum est intellectualis, et est intelligentia de qua est sermo apud philosophum in libro de metaphisica’ (Ps.-Avicenna Latinus, Liber de causis primis et secundis, p. 98, 4 – 7). De Vaux, Notes et textes, p. 98, n. 1, rightly points out that this quotation mirrors a passage of al-Fa¯ra¯bı¯’s De intellectu et intellecto (‘et hoc est intelligencia quam ponit Aristoteles … in libro de metaphisica’, Liber Alpharabii de intellectu et intellecto, p. 126, 393 – 4, cf. p. 115, 9), corresponding to Risa¯la fı¯ l- aql, p. 36, 1 (cf. p. 4, 3). The philosophus in question, therefore, appears to be Aristotle, not Avicenna, as De Vaux surmises (p. 71; p. 98, n. 1). 43 Ps.-Avicenna Latinus, Liber de causis primis et secundis, pp. 108 – 23 (chapters 6 – 8), especially p. 110, n. 1; p. 116, n. 1. 44 Whereas the wide array of Arabic sources and the inter-confessional approach point to a Toledan (or Catalan) milieu (see D’Alverny, Les prgrinations de l’me, pp. 266 – 7), the frequent medical references, as well as some codicological features of the manuscript in which the work is preserved, indicate Bologna as a possible place of composition (see d’Alverny, Les traductions d’Avicenne (Moyen Age et Renaissance), p. 79; d’Alverny, Avicennisme en Italie, pp. 121 – 2). ˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

209

210 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Amos Bertolacci

chrtienne occidentale’.45 The Arabic Neoplatonism reflected in this treatise is ˙ aza¯lı¯’s represented mainly by Avicenna’s Philosophia prima, together with al-G metaphysics, the Liber de causis and Ibn Gabirol’s Fons vitae. Among Latin authors, d’Alverny stresses its dependence on Gundissalinus.46 In this writing, the Philosophia prima is always quoted implicitly, sometimes ad litteram, elsewhere ad sensum. 47 To the first category belongs the quotation of the first sentence of Philosophia prima IX, 7 at the beginning of the treatise.48 The second category includes the similarities with Avicenna’s hierarchy of celestial intelligences in Philosophia prima IX, 4, his description of God’s attributes in VIII, 4 – 7, and his view of the misery of afterlife in IX, 7.49 Aristotle’s Metaphysics, on the other hand, is never quoted.50 This treatise is not the only example of the tendency to conjoin the Philosophia prima with Latin metaphysics. D’Alverny points at the existence of other similar, still unedited and uninvestigated, witnesses of the synthesis of Islamic (mainly Avicennian) and Christian Neoplatonism.51 § 2.5 Michael Scot The dependence of Michael Scot (d. 1235 ca.) on Avicenna’s works, with particular regard to the De anima, has been already documented.52 The introduction to philosophy that is fragmentarily preserved in Vincent of Beauvais’ (d. 1264 ca.) Speculum doctrinale witnesses a wide recourse to the Philosophia prima. 53 Michael wrote this introduction probably before his 45 D’Alverny, Les prgrinations de l’me, p. 240. 46 D’Alverny, Les prgrinations de l’me, pp. 242 – 3. 47 Avicenna might be one of the ‘philosophiae peritissimi’ mentioned in the first pages (Anonymous, Homo, cum in honore esset …, p. 282, 11). 48 Anonymous, Homo, cum in honore esset …, p. 282, 8 – 10. 49 Anonymous, Homo, cum in honore esset …, pp. 291 – 3, 297, 299; cf. d’Alverny, Les prgrinations de l’me, pp. 245, 247. 50 Aristotle is mentioned only once as author of the Liber de causis (Anonymous, Homo, cum in honore esset …, p. 281, 11 – 12). 51 D’Alverny, Une rencontre symbolique, pp. 178 – 9, mentions an unedited treatise (MS Munich, Bayerische Staatsbibliothek, Lat. 380, fols 21v-24r, described in d’Alverny, Alain de Lille, p. 220; Oxford, Bodleian Library, Digby 28, fols 74r-81v) on the First Cause, the Trinity and the emanation of creatures, in which the account of the Trinity ‘utilize largement des terms tirs de la Mtaphysique d’Avicenne’. 52 See Hasse, Avicenna’s ‘De anima’, pp. 23 – 30. 53 Michael Scot, Introduction to Philosophy, fragments 1 – 6. Thus, the dependence of this work of Michael on Arabic sources, pointed out by Baur (Gundissalinus, De divisione philosophiae, pp. 366 – 7; cf. Van Steenberghen, La philosophie au XIIIe sicle, p. 113), can be regarded as a reliance, among others, on Avicenna.

On the Latin Reception of Avicenna’s Metaphysics before Albertus Magnus 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

211

translations of Averroes’ long commentaries.54 The Philosophia prima is quoted not only indirectly, through the citations that Michael finds in his main source, Gundissalinus’ De divisione philosophie,55 but also directly. The four-fold division of metaphysics in fragment 5, for example, is taken directly from Philosophia prima I, 2.56 In this case, Michael mentions Avicenna explicitly: however, he refers to the metaphysics of both Avicenna and Aristotle (‘Hae quattuor partes continentur in metaphysica Aristotelis et Avicennae’), as if the metaphysical views of these two authors were one and the same.57 He also adds, ˙ aza¯lı¯ and Ibn Gabirol, together with, but in accessorily, the names of al-G distinction from, Aristotle and Avicenna (‘… et in Algazel et in Avicebronte’).58 This quotation is significant in many respects. First, it is the first known attestation of a joint mention of Aristotle and Avicenna about a metaphysical doctrine: since, in fact, Michael deals with a doctrine of Avicenna, he shows that he takes Avicenna’s metaphysics as representative also of Aristotle’s. Second, this is one of the first cases in which other important exponents of Arabic thought, ˙ aza¯lı¯, are mentioned together with Avicenna on a metaphysical such as al-G ˙ aza¯lı¯’s summary of Avicenna’s philosophy topic. Aristotle’s Metaphysics and al-G start to be known, but the Philosophia prima remains for Michael, as for previous authors, the main text on metaphysics.

54 The absence of any reference to Averroes in these fragments, and their dependence on Gundissalinus, suggest that this work was composed in Toledo in the early period of Michael’s career, i. e. before 1220 (cf. Burnett, Michael Scot, p. 105). 55 L. Baur documents the dependence of Michael’s works on the De divisione philosophie in Gundissalinus, De divisione philosophiae, pp. 365 – 6; see also Fidora, Die Wissenschaftstheorie des Dominicus Gundissalinus, p. 13. 56 Gundissalinus, De divisione philosophiae, p. 400, fr. 5; cf. Philosophia prima I, 2, p. 14, 68–p. 15, 74 [p. 14, 14 – 18] (the tripartition of metaphysics in Avicenna’s original text is rendered as a quadripartition in the Latin translation). Gundissalinus reports Avicenna’s view on the articulation of metaphysics differently (see Gundissalinus, De divisione philosophiae, p. 37, 10 – 17), in accordance with another passage of the Philosophia prima (I, 2, p. 16, 2–p. 17, 11 [p. 15, 17–p. 16, 5]). Square brackets refer to the Arabic text. 57 Whereas the explicit quotation of Aristotle in fragment 6 is taken from Gundissalinus, the ones in fragments 1 – 2 do not derive from the corresponding passages of the Divisione philosophiae: since in these passages Gundissalinus implicitly quotes doctrines by Avicenna, Michael might have added the name of Aristotle, on account of the identity of views that he ascribes to Aristotle and Avicenna in the aforementioned passage of fragment 5. ˙ aza¯lı¯’s Summa. 58 No division of metaphysics occurs in al-G

212 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Amos Bertolacci

§ 3 The Entrance of the Philosophia prima at the University of Paris (John Blund; Prohibitions of 1210 and 1215) The first evidence at our disposal regarding the second phase of the periodization proposed above comes from masters of arts and professors of theology at the University of Paris during the first two decades of the thirteenth century, who attest the introduction of Avicenna’s Philosophia prima in the university environment and the recourse to this latter as a complement of Aristotle’s Metaphysics. Whereas the exponents of the faculty of arts considered here (John Blund) show a positive attitude towards Avicenna’s work, the faculty of theology has probably expressed a veto against the Philosophia prima, including it in the first Aristotelian condemnations of 1210 and 1215. § 3.1 John Blund The Tractatus de anima of John Blund (1175 ca.–1248) was written in Paris (or in Oxford shortly after the author’s stay in Paris) in the first years of the thirteenth century. It is an important witness of the early diffusion of the Philosophia prima, since it provides the first signs of its consideration as a complement of Aristotle’s Metaphysics. Some of the frequent explicit mentions of Avicenna in this work are quotations of the Philosophia prima, of which both the ontological (I, 1, I, 5, II, 1) and the theological part (IX, 2) are cited.59 Aristotle’s Metaphysics, on the other hand, is never referred to directly. In two cases Blund mentions not only Avicenna, but also the title of his work. These mentions are revealing, for in both cases the doctrine quoted is the same – i. e. the classification of substances, including the soul, at the end of Philosophia prima II, 1 – but the cited work of Avicenna, namely the Philosophia prima, is differently described: in one case, Blund calls it Metaphysica, as it was called in the De causis primis et secundis;60 in the other case, by contrast, he portrays it as a ‘commentary’ (commentum) on the Metaphysics. 61 This latter is one of the first 59 See, besides the quotations reported in the following footnotes, also those occurring in John Blund, Tractatus de anima, § 13 (Philosophia prima IX, 2, p. 387, 12 – 13 [p. 455, 10 – 14]), § 18 (Philosophia prima I, 2, p. 10, 6 – 8 [p. 9, 59 – 62]), § 85 (Philosophia prima I, 5, p. 29, 5 – 6 [p. 31, 2 – 3]). 60 John Blund, Tractatus de anima, § 316, p. 85, 26-p. 86, 1: ‘ut dicit Avicenna in Metaphysica’. See above, n. 41. 61 John Blund, Tractatus de anima, § 32, p. 9, 12 – 13: ‘Ab Avicenna habemus in commento Metaphysice …’. In another passage, Blund ascribes a commentum prime philosophie to ˙ aza¯lı¯ (see John Blund, Tractatus de anima, § 101, p. 27, 28 – 9: both Avicenna and al-G ‘sicut testatur tam Avicenna quam Algazel in commento prime philosophie’). Although it ˙ aza¯lı¯’s Summa with respect to reveals awareness of the doctrinal similarity of al-G

On the Latin Reception of Avicenna’s Metaphysics before Albertus Magnus 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

213

known attestations in Latin philosophy of the consideration of Avicenna’s Philosophia prima as a commentary on the Metaphysics. In the Tractatus, Blund establishes an identical relationship of commentary-text commented upon between Avicenna’s and Aristotle’s De anima; in this context, Avicenna is also named Commentator. 62 The term commentum in Blund’s Tractatus does not mean a work providing the literal exegesis of another, but a writing recasting in a different form, and therefore ‘explaining’, the doctrine of another.63 Thus, whereas the first quotation considered here is reminiscent of the phase in which Aristotle’s Metaphysics was substantially absent from the philosophical scenario and replaced with Avicenna’s metaphysics, the second quotation indicates that the Metaphysics starts circulating in the University of Paris at the beginning of the thirteenth century and that scholars grasp the doctrinal affinity between it and the Philosophia prima, regarding this latter as a sort of ‘companion’ to Aristotle’s work. Significantly, in a few cases of joint quotation of Aristotle and Avicenna, the doctrine at stake is phrased according to Avicenna’s text, rather than Aristotle’s.64 However, apart from this and other scattered statements, the actual use of the Philosophia prima in Paris as a tool for the interpretation of the Metaphysics at the turn between the twelfth and the thirteenth century remains undocumented. § 3.2 The Prohibitions of 1210 – 1215 Furthermore, the two interdictions of the teaching of Aristotle at the University of Paris in 1210 and 1215, if attentively examined, attest that in the second decade of the thirteenth century Aristotle’s Metaphysics is viewed as distinct from Avicenna’s Philosophia prima, and that this latter is considered as a commentary on the former. In this regard, the interdictions mark the transition from the first to the second phase of the Latin reception of Avicenna’s work. The first prohibition – issued by the council of the ecclesiastic province of Sens, held in Paris in 1210 under the lead of Peter of Corbeil – bans lectures on ˙ aza¯lı¯ also in § 13), the Avicenna’s works (cf. the joint quotation of Avicenna and al-G ˙ aza¯lı¯’s psychology ascription is puzzling, since the reference is to Avicenna’s and al-G (doctrine of vision), rather than metaphysics. 62 See the loci mentioned by Hasse, Avicenna’s ‘De anima’, p. 20, n. 41. 63 See the remarks in Hasse, Avicenna’s ‘De anima’, p. 20. Blund’s consideration of Avicenna’s De anima as a commentary on Aristotle’s text may have influenced his view of the relation of Avicenna’s Philosophia prima vis--vis Aristotle’s Metaphysics. 64 See, for instance, John Blund, Tractatus de anima, § 18: ‘Praeterea. Ab Aristotele et ab Avicenna habemus quod subiectum physici est corpus mobile in quantum ipsum est subiectum motus et quietis’; Philosophia prima I, 2, p. 10, 6 – 8 [p. 9, 59 – 62]: ‘Dico autem quod suum subiectum scientiae naturalis est corpus … inquantum est subiectum motui et quieti’.

214 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Amos Bertolacci

‘Aristotle’s physical books’ (libri Aristotelis de naturali philosophia) and the ‘commentaries’ (commenta) thereupon.65 Five years later, the renewed prohibition – introduced in the university statutes of 1215 by the delegate of the Pope, Robert of CourÅon – clarifies the range of the books of Aristotle under interdiction (mentioning not only the physical but also the metaphysical books), and describes differently the literary genre of their forbidden explanation, speaking of ‘summaries’ (summae de eisdem).66 By comparing the first with the second prohibition, two noteworthy differences emerge. One is that Aristotle’s Metaphysics is mentioned only in the second prohibition, not in the first. The other is that the writings associated with Aristotle’s works are named commenta in the first prohibition, summae in the second. Since the second prohibition is formally the renewal of the first, the precisions that it introduces can be either clarifications of, or additions to, the previous interdiction. Thus, with regard to the first difference we have noticed, the second prohibition appears to clarify the scope of the books of Aristotle being interdicted, imprecisely reckoned in the first ban: it makes clear that the ‘physical books’ of Aristotle are to be taken in a wide meaning and include also the metaphysical writings of the Philosopher.67 This being the case, the commenta on the physical books of Aristotle mentioned in the first prohibition include in all likelihood also ‘commentaries’ on the Metaphysics. On the other hand, with regard to the second difference, it is less clear how the summae in the second prohibition relate to the commenta in the first. One possibility is that both expressions refer to the same writings, differently described: we know, for example, that the title of summa (as well as those of compendium or tractatus) was used since the twelfth century by masters of logic to designate explanations

65 Chartularium Universitatis Parisiensis, vol. I, p. 70, § 11: ‘… nec libri Aristotelis de naturali philosophia nec commenta legantur Parisius publice vel secreto …’. 66 Chartularium Universitatis Parisiensis, vol. I, pp. 78 – 9, § 20: ‘Non legantur libri Aristotelis de methafisica et de naturali philosophia nec summe de eisdem …’. 67 Although it cannot be excluded that the Metaphysics gained popularity in the University of Paris only after 1210, the historical evidence at our disposal suggests that it was already included in the first prohibition. Grabmann, I divieti ecclesiastici, p. 44, aptly notices that the first prohibition rests on a division of philosophy in rationalis, naturalis and moralis, in which the philosophia naturalis encompasses also metaphysics. At pp. 11 – 12, Grabmann reports a text of William Brito, mentioning, with regard to the first prohibition, the interdiction of libelli quidam ab Aristotele, ut dicebatur, compositi, qui docebant metaphysicam. This text is recorded among the witnesses of the first Latin diffusion of the Metaphysics by Vuillemin-Diem, Praefatio, in Aristoteles Latinus XXV 1 – 1a, p. xvi (Sharpe, A Handlist, p. 756, § 1998, places Brito’s activity towards the end of the 13th century, rather than at its beginning, as Vuillemin-Diem does). The many respects in which Brito’s report differs from the actual text of the proscription are evidenced by Grabmann, ibid.

On the Latin Reception of Avicenna’s Metaphysics before Albertus Magnus

˘

of Aristotle’s works.68 Another possibility, by contrast, is that the summae in the second prohibition are not the same as the commenta in the first: since the second prohibition informs us of the existence of summae not only of Aristotle’s books, but also of the writings of masters such as David of Dinant, Amalricus of Bene and ‘Mauricius Hispanus’ (whoever this latter may be),69 one might take the summae in question to be abstracts and summaries of the works of Aristotle and the other condemned authors, made by professors of the faculty of arts in order to escape the limitations imposed by the proscription of 1210.70 If this is the case, however, it seems difficult to identify the summae with the commenta. Coming to Avicenna, scholars generally agree that both the commenta and the summae mentioned in the prohibitions can be identified with the Latin translations of the Sˇifa¯ .71 This point of view is, with some provisos, acceptable. The term commenta in the first prohibition very likely designates the Latin translations of Avicenna, for three reasons. First of all, the consideration of the Sˇifa¯ as intimately connected with Aristotle’s corpus, by being its derivation, complement or explanation, is shared by the first Latin translators of Avicenna72 ˘

68 See the evidence discussed in Hasse, Der mutmaßliche Einfluss. 69 ‘Non legantur libri Aristotelis de methafisica et de naturali philosophia nec summe de eisdem aut de doctrina magistri David de Dinant, aut Almarici heretici aut Mauricii hyspani’ (Chartularium Universitatis Parisiensis, vol. I, pp. 78 – 9, § 20; emphasis added). 70 This hypothesis is advanced by Mandonnet, Siger de Brabant. The counter-argument of Grabmann, I divieti ecclesiastici, p. 30, namely the fact that no coeval abbreviation of Aristotle’s texts is extant, does not seem conclusive. Daniel of Morley (d. 1210 ca.), in his famous polemical report on the status of teaching in the university of Paris towards the end of the 12th century, mentions the habit of studying jurisprudence by means of summaries (sub compendio) rather than on the original texts (Daniel of Morley, Philosophia, p. 212). 71 De Vaux, Notes et textes, pp. 45 – 52; Grabmann, I divieti ecclesiastici, pp. 49 – 51; d’Alverny, Les prgrinations de l’me, p. 242; Van Steenberghen, La philosophie au XIIIe sicle, p. 85; Van Riet, La traduction latine, p. 54*; Elamrani-Jamal, La rception, pp. 34 – 5; Bianchi, Censure et libert intellectuelle, p. 94 and n. 20. The initiator of this interpretative tendency, R. De Vaux, points at the compatibility between the literary genre of the Sˇifa¯ and the terms commenta/summae, in order to include Avicenna’s works in the condemnations and corroborate in this way the hypothesis of the existence of a ‘heretical’ Latin Avicennism, condemned by Parisian theologians. Grabmann, I divieti ecclesiastici, pp. 49 – 50 (cf. pp. 12 – 13), adds further evidence to De Vaux’s arguments, namely the reference to Spain and Toledo (place of composition of the Latin translations of Avicenna) that can be found in the report of the Parisian condemnations in the Speculum ecclesiae of Gerald of Wales (1146 – 1226); on this author, see Sharpe, A Handlist, pp. 134 – 7 (§ 350). 72 Several significant examples can be adduced. (1) In the Prologue of the Latin translation of Avicenna’s Liber de anima, Avendeuth portrays this work as a book that, in the most complete form (plenissime), gathers and replaces what Aristotle says in his De anima and De sensu et sensato (Avendeuth mentions also the pseudo-Aristotelian De intellectu et intellecto): ‘Habetis ergo librum … ex arabico translatum: in quo quidquid Aristoteles ˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

215

216

and is widespread at the time.73 Secondly, an author active in Paris a few years before the condemnations like John Blund expressly portrays Avicenna’s De anima and Philosophia prima as commenta of, respectively, Aristotle’s De anima and the Metaphysics, as we have seen. Thirdly, the commentaries par excellence on Aristotle’s works, namely Averroes’ long commentaries, cannot be the commenta referred to in the 1210 prohibition, since they were translated into Latin only later (around 1220 – 1235).74 Thus, the Philosophia prima is quite probably alluded to in the first prohibition as commentum on the Metaphysics. The summae in the second prohibition, on the other hand, do (or do not) designate the Sˇifa¯ , depending on whether they are (or they are not) the same as the ˘

˘

˘

˘

dixit in libro suo de anima et de sensu et sensato et de intellectu et intellecto, ab auctore libri sciatis esse collectum; unde, postquam, volente Deo, hunc habetis, in hoc illos tres plenissime vos habere non dubitetis’ (Avicenna Latinus, Liber de anima, p. 4, 21 – 5). On the Latin translation of the De intellectu of Alexander of Aphrodisias, see Burnett, Arabic into Latin: the Reception of Arabic Philosophy into Western Europe, p. 392 and n. 11. (2) Likewise, Alfred of Sareshel prolonged Aristotle’s meteorology with the mineralogy contained in chapters I, 1 and I, 5 of Avicenna’s corresponding section, thus revealing an acute perception of the Aristotelian inspiration of the Sˇifa¯ , especially if he regarded these chapters as written by Avicenna rather than by Aristotle himself (according to Otte, Alfred of Sareshel’s Commentary, Alfred did not know Avicenna’s authorship of these two chapters, a suggestion discarded by Mandosio, Di Martino, La Mtorologie d’Avicenne, pp. 413 – 15). Significantly, Alfred commented not only on Aristotle’s Meteorology, but also on Avicenna’s aforementioned chapters, and described Avicenna as imitator of Aristotle and as the second most important philosophical authority after the Stagirite (‘imitator Aristotelis precipuus, immo ipso Aristotele excepto, philosophorum maximus’, Alfred of Sareshel, Commentary, p. 50, 18 – 19). (3) Michael Scot’s translation of Avicenna’s reworking of Aristotle’s Historia animalium, De partibus animalium and De generatione animalium followed later his translation, from Arabic, of these three Aristotelian works (as a single unit, with the title De animalibus). It is reasonable to suppose that, after having translated Aristotle’s works on zoology, Michael wanted to provide, with the translation of the part of the Sˇifa¯ on this topic, its interpretative tool. (4) By inserting the Latin translation of fragments of the rhetoric of the Sˇifa¯ into his Arabic-Latin translation of Aristotle’s Rhetoric, also Hermann the German aimed at clarifying the obscurities of the Arabic text of Aristotle’s work (see d’Alverny, Notes, pp. 339, 347). On account of all this evidence, it is safe to conclude that the first translators of the Sˇifa¯ ‘intended to provide Western scholars with a commentary on Aristotle’s works’ (D’Alverny, Translations and Translators, p. 451). 73 See De Vaux, Notes et textes, p. 10: ‘Et cet ensemble [des traductions d’Avicenne] passait pour Þtre un commentaire autoris – le meilleur, mieux: le seul – de toute la philosophie aristotelicienne’; Goichon, La philosophie d’Avicenne, p. 90: ‘Celui-ci [i.e. le Sˇifa¯ ] passait pour un commentaire’; Van Steenberghen, La philosophie au XIIIe sicle, p. 173: ‘les paraphrases d’Avicenne ont servi d’instrument de travail aux premiers exgtes d’Aristote, jusq’au jour o elles ont t dtrnes par les commentaries littraux d’Averros’; Soto Bruna, Estudio fil sofico, p. 38: ‘Este ultimo [i.e. el Sˇifa¯ ] fue considerado en su momento como un comentario de la propia Metafsica de Arist teles’. 74 Denifle contends the contrary in Chartularium Universitatis Parisiensis, vol. I, p. 71, n. 15. ˘

˘

˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Amos Bertolacci

On the Latin Reception of Avicenna’s Metaphysics before Albertus Magnus

˘

˘

commenta in the first prohibition. Also apart from stylistic considerations (the term summa fits the literary format of the Sˇifa¯ quite well, and it is even ˙ aza¯lı¯’s Maqa¯sid, i. e. Summa contained in the title of the Latin translation of al-G ˙ of summae in theoricae philosophiae), it is not impossible to take the occurrence 75 ˇ the second prohibition as referring to the Sifa¯ . Thus, it seems safe to conclude that the Philosophia prima entered the curriculum of the faculty of arts of the University of Paris some time before 1210, playing there the role of an authoritative text to be read together with the Metaphysics in order to convey its interpretation. The quotations of the Philosophia prima in John Blund fit into this scenario.76 Together with the Metaphysics and the other writings of Aristotle, Avicenna’s work must have aroused the suspicion and alarm of the members of the faculty of theology, who promoted the condemnations of 1210 and 1215 in front of the ecclesiastic authorities. This is confirmed a posteriori. When the prohibitions lose their validity, and Aristotle’s writings were ‘rehabilitated’ in Paris in 1231, the Philosophia prima regained its role of interpretive tool of the Metaphysics. The Parisian ‘Guide of the Student’ of 1230 – 1240, for example, reveals a certain silent influence of Avicenna in metaphysics.77 The authors active in Paris in the fourth decade of the thirteenth century will rely massively on the Philosophia prima, providing the first known attestations of the use of this work no longer without, but together with the Metaphysics.

˘

75 In this prohibition, the term summae might have two different interrelated connotations: one indicating the explanatory summaries of Aristotle’s books on metaphysics and natural philosophy, in accordance with the attested practice of naming summae the exegetical works produced within the faculty of arts (see above, n. 68); another designating more specifically the abridgements of the teaching of the masters involved in the condemnation. In its first meaning, the term summae would encompass the Sˇifa¯ . 76 Other works written in the faculty of arts of the University of Paris at the beginning of the 13th century – like the surviving fragments of the Quaternuli of David of Dinant, condemned in 1210 – show acquaintance with the Metaphysics, but no significant recourse to the Philosophia prima (see Anzulewicz, Person und Werk des Davids von Dinant; Anzulewicz, David von Dinant, pp. 81, 90). Vuillemin-Diem, Zum Aristoteles Latinus, p. 30, remarks, however, that the extant fragments of David’s Quaternuli ‘reprsentieren zweifellos nur einen sehr kleinen Teil aus der verlorenen wissenschaftlichen und literarischen Produktion Davids’. 77 See De Libera, Structure du corps scolaire, p. 75: ‘Cette conception du systme de la mtaphysique comme science [dans la Guide] n’est pas trangre  la presentation de la prima philosophia dans la Mtaphysique du Shifa¯ d’Avicenne’. ˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

217

218 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Amos Bertolacci

Conclusion The stage of the Latin reception of the Philosophia prima that antedates William of Auvergne is quite rich and interesting. Several authors and works are involved, and virtually the entire Philosophia prima is taken into account. Historically, this initial phase connects the period of the translation of the Philosophia prima in the second half of the twelfth century with its employment by theologians in Oxford and Paris from the third decade of the following century onward. Doctrinally, the reception of the Philosophia prima in this early phase is worth considering. On the one hand, it is still, in a way, immature, since it mainly consists in the repetition, often silent, of Avicenna’s views on scattered topics, rather than in their critical evaluation and theoretical refinement. On the other hand, however, authors focus on some crucial points of Avicenna’s metaphysics: this is the case of chapter I, 6 of the Philosophia prima (the distinction of necessary and contingent, and of necessary per se and necessary in virtue of something else) in Gundissalinus’ De processione mundi; and of chapter I, 5 (the idea of ‘existent’ as first intelligible) in John Blund’s Tractatus de anima. Thus, distinctions and doctrines that are central in Avicenna’s metaphysics, and provide evidence of the endorsement of Avicenna’s thought in later authors, are already at stake in this early stage. More than its features, the very existence of this stage is significant. It attests that the transmission of Avicenna’s metaphysics into Latin represent a historical and doctrinal continuum. Future research will have to investigate whether the absence of Aristotle’s Metaphysics at this early stage is just accidental, or rather is causally linked with the diffusion of the Philosophia prima. The Metaphysics is occasionally mentioned by Latin authors at this stage, but knowledge of it remains little more than virtual. The fact that two Latin translations of the Metaphysics were made in the twelfth century, but underwent a partial loss or remained unexploited until later, calls for an explanation. One might think that interest in metaphysical issues at this stage was, in general, not too strong. But one might also surmise that the success of the Philosophia prima somehow prevented the diffusion of the Metaphysics, by providing a metaphysical system that was regarded by Latin scholars as more coherent and complete than Aristotle’s. These are questions that wait to be answered. The fact remains that Avicenna’s Philosophia prima seems to have spread in Latin philosophy before and, initially, without the Metaphysics.

On the Latin Reception of Avicenna’s Metaphysics before Albertus Magnus

Bibliography Primary Sources al-Fa¯ra¯bı¯, Liber Alpharabii de intellectu et intellecto, in Gilson, Les sources greco-arabes de l’augustinisme avicennisant, Archives d’Histoire doctrinale et littraire du Moyen Age, 4, 1929, pp. 5 – 149 (pp. 115 – 26). –– , Risa¯la fı¯ l- aql, ed. M. Bouyges, Beirut: Imprimerie Catholique, 1938 (reprinted Beirut: Da¯r al-Masˇriq, 1983). Alfred of Sareshel, Commentary on the ‘Meteora’ of Aristotle, ed. J.K. Otte, Leiden/New York/Copenhagen/Cologne: Brill, 1988. ˙ aza¯lı¯, Algazel’s Metaphysics. A Mediaeval Translation, ed. J.T. Muckle, Toronto: al-G St. Michael’s College, 1933. Anonymous, Homo, cum in honore esset, … (MS Paris, Bibliothque Nationale, Lat. 3236 A, fols 85v-87), in M.-T. d’Alverny, Les prgrinations de l’me dans l’autre monde d’aprs un anonyme de la fin du XIIe sicle, Archives d’Histoire doctrinale et littraire du Moyen Age, 15 – 17, 1940 – 42, pp. 280 – 99. Avicenna Latinus, Liber de anima seu Sextus de Naturalibus, I-II-III, ed. S. Van Riet, Louvain/Leiden: Peeters/Brill, 1972. –– , Liber de philosophia prima sive Scientia divina, I-IV, ed. S. Van Riet, Louvain/ Leiden: Peeters/Brill, 1977. –– , Liber de philosophia prima sive Scientia divina, V-X, ed. S. Van Riet, Louvain/ Leiden: Peeters/Brill, 1980. –– , Liber de philosophia prima sive Scientia divina, I-X, ed. S. Van Riet, Louvain/Leiden: Peeters/Brill, 1983. Ps.-Avicenna Latinus, Le Liber de causis primis et secundis et de fluxu qui consequitur eas, in R. De Vaux, ed., Notes et textes sur l’avicennisme latin aux confins des XIIe-XIIIe sicles, Paris: Vrin, 1934, pp. 83 – 140. Boethius, De trinitate, in The Theological Tractates, transls and eds H. F. Stewart, E. K. Rand, London/Cambridge, Massachusetts: Heinemann/Harvard University Press, 1962, pp. 2 – 31. Chartularium Universitatis Parisiensis, ed. H. Denifle, A. Chatelain, 4 vols, Paris 1889 – 97; reprinted Bruxelles: Culture et Civilisation, 1964. Daniel of Morley, Philosophia, ed. G. Maurach, in Mittellateinisches Jahrbuch, 14, 1979, pp. 204 – 55. Gundissalinus (Dominicus Gundisalvi), De divisione philosophiae, ed. L. Baur, Mnster: Aschendorff, 1903. –– , De divisione philosophiae. ber die Einteilung der Philosophie. Lateinisch-Deutsch, transl and ed. A. Fidora und D. Werner, Freiburg/Basel/Vienna: Herder 2007. –– , De processione mundi. Estudio y edicin crtica del tratado de Domingo Gundisalvo, eds M. J. Soto Bruna, C. Alonso Del Real, Pamplona: Ediciones Universidad de Navarra, 1999. –– , The Procession of the World (De processione mundi), transl. and ed. J. A. Laumakis, Milwaukee: Marquette University Press, 2002. John Blund, Tractatus de anima, ed. D. A. Callus, R. W. Hunt, London: Oxford University Press, 1970. Michael Scot, Introduction to Philosophy (Fragments), in De divisione philosophiae, ed. L. Baur, Mnster: Aschendorff, 1903, pp. 364 – 7, 398 – 400. ˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

219

220

William of Auvergne, Opera omnia, 2 vols, ed. F. Hotot, Supplementum, ed. B. LeFeron, Paris/Orlans 1674; reprinted Frankfurt a. M.: Minerva, 1963.

Secondary Sources M.-T. d’Alverny, Alain de Lille. Textes indits. Avec une introduction sur sa vie et ses œuvres, Paris: Vrin, 1965. –– , Avicennisme en Italie, in Accademia Nazionale dei Lincei, Fondazione A. Volta, eds, Oriente e Occidente nel Medioevo: Filosofia e Scienze (Atti del Congresso Internazionale, 9 – 15 aprile 1969; n8 13), Rome: Accademia Nazionale dei Lincei, 1971, pp. 117 – 39; reprinted in Avicenne en Occident, Paris: Vrin, 1993, XVI. –– , Deux traductions latines du Coran au Moyen Age, Archives d’Histoire doctrinale et littraire du Moyen Age, 22 – 23, 1947 – 48, pp. 69 – 131; reprinted in La connaissance de l’Islam dans l’Occident mdival, ed. C. Burnett, Aldershot: Ashgate, 1994. –– , Les prgrinations de l’me dans l’autre monde d’aprs un anonyme de la fin du XIIe sicle, Archives d’Histoire doctrinale et littraire du Moyen Age, 15 – 17, 1940 – 42, pp. 239 – 99. –– , Les traductions d’Avicenne (Moyen Age et Renaissance), in Problemi attuali di Scienza e di Cultura, Quaderno n8 40 (Avicenna nella storia della cultura medievale), Rome: Accademia Nazionale dei Lincei, 1957, pp. 73 – 87; reprinted in Avicenne en Occident, Paris: Vrin, 1993, V. –– , Les traductions d’Avicenne: quelques rsultats d’une enquÞte, in R. Marchal, ed., Actes du Ve Congrs international des Arabisants (Bruxelles, 1970), Bruxelles: Centre pour l’tude des Problmes du Monde Musulman Contemporain, 1971, pp. 151 – 8; reprintred in Avicenne en Occident, Paris: Vrin, 1993, VII. –– , Notes sur les traductions mdivales d’Avicenne, Archives d’Histoire doctrinale et littraire du Moyen Age, 19, 1952, pp. 337 – 58 reprinted in Avicenne en Occident, Paris: Vrin, 1993, IV. –– , Une rencontre symbolique de Jean Scot Erigne et d’Avicenne. Notes sur le De Causis Primis et Secundis et Fluxu qui consequitur eas, in S. O’Meara, C. Bieler, eds, The Mind of Eriugena, Dublin: Irish University Press, 1973, pp. 170 – 81. –– , Translations and Translators, in R.L. Benson, G. Constable, eds, Renaissance and Renewal in the Twelfth Century, Cambridge, Massachusetts: Harvard University Press, 1982, pp. 421 – 62. G.C. Anawati, La Mtaphysique d’Avicenne dans l’Occident latin, in Avicenna, La Mtaphysique du Shifa¯ . Livres I  V, transl. and ed. G.C. Anawati, Paris: Vrin, 1978, pp. 56 – 79. H. Anzulewicz, David von Dinant und die Anfnge der aristotelischen Naturphilosophie im Lateinischen Westen, in L. Honnefelder, R. Wood, M. Dreyer, M.-A. Aris, eds, Albertus Magnus und die Anfnge der Aristoteles-Rezeption im lateinischen Mittelalter. Von Richard Rufus bis zu Franciscus de Mayronis, Mnster: Aschendorff, 2005, pp. 71 – 112. –– , Person und Werk des Davids von Dinant im literarische Zeugnis Alberts des Grossen, Mediaevalia Philosophica Polonorum, 34, 2001, pp. 15 – 58. A. Bertolacci, A Community of Translators: The Latin Medieval Versions of Avicenna’s Book of the Cure, in J.N. Crossley, C.J. Mews, eds, Communities of Learning: ˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Amos Bertolacci

On the Latin Reception of Avicenna’s Metaphysics before Albertus Magnus

Networks and the Shaping of Intellectual Identity in Europe 1100 – 1450, Turnhout: Brepols, 2011, pp. 37–55. –– , The Reception of Aristotle’s ‘Metaphysics’ in Avicenna’s ‘Kita¯b al-Sˇifa¯ ’: A Milestone of Western Metaphysical Thought, Leiden: Brill, 2006. L. Bianchi, Censure et libert intellectuelle l’universit de Paris, Paris: Les Belles Lettres, 1999. C. Burnett, A Note on the Origins of the Physica Vaticans and Metaphysica Media, in R. Beyers, J. Brams, D. Sacr, K. Verrycken, eds, Tradition et traduction. Les textes philosophiques et scientifiques grecs au Moyen ffge latin. Hommage Fernand Bossier, Leuven: Leuven University Press, 1999, pp. 59 – 68. –– , Arabic into Latin: the Reception of Arabic Philosophy into Western Europe, in P. Adamson, R. Taylor, eds, The Cambridge Companion to Arabic Philosophy, Cambridge: Cambridge University Press, 2005, pp. 370 – 404. –– , Michael Scot and the Transmission of Scientific Culture from Toledo to Bologna via the Court of Frederick II Hohenstaufen, Micrologus, 2, 1994, pp. 101 – 26. –– , The Blend of Latin and Arabic Sources in the Metaphysics of Adelard of Bath, Hermann of Carinthia, and Gundisalvus, in M. Lutz-Bachmann, A. Fidora, A. Niederberger, eds, Metaphysics in the Twelfth Century. On the Relationship among Philosophy, Science and Theology, Turnhout: Brepols, 2004, pp. 41 – 65. H.A. Davidson, Alfarabi, Avicenna and Averroes on Intellect. Their Cosmologies, Theories of the Active Intellect, and Theories of Human Intellect, New York/Oxford: Oxford University Press, 1992. A. De Libera, Penser au Moyen ffge, Paris: ditions du Seuil, 1991. –– , D’Avicenne  Averros, et retour. Sur les sources arabes de la thorie scolastique de l’un transcendental, Arabic Sciences and Philosophy, 4, 1994, pp. 141 – 79. –– , Proofs for Eternity, Creation and the Existence of God in Medieval Islamic and Jewish Philosophy, New York/Oxford: Oxford University Press, 1987. –– , Structure du corps scolaire de la mtaphysique dans la premire moiti du XIIIe sicle, in C. Lafleur, J. Carrier, eds, L’enseignement de la philosophie au XIIIe sicle. Autour du ‘Guide de l’tudiant’ du ms. Ripoll 109, Turnhout: Brepols, 1997, pp. 61 – 88. R. De Vaux, Notes et textes sur l’avicennisme latin aux confins des XIIe-XIIIe sicles, Paris: Vrin, 1934. A. Elamrani-Jamal, La rception de la philosophie arabe  l’univesit de Paris au XIIIe sicle, in C.E. Butterworth, B.A. Kessel, eds, The Introduction of Arabic Philosophy into Europe, Leiden/New York/Kçln: Brill, 1994, pp. 30 – 39. A. Fidora, Die Wissenschaftstheorie des Dominicus Gundissalinus. Voraussetzungen und Konsequenzen des zweitens Anfangs der aristotelischen Philosophie im 12. Jahrhundert, Berlin: Akademie Verlag, 2003. –– , Zum Verhltnis von Metaphysik und Theologie bei Dominicus Gundissalinus, in M. Lutz-Bachmann, A. Fidora, A. Niederberger, eds, Metaphysics in the Twelfth Century. On the Relationship among Philosophy, Science and Theology, Turnhout: Brepols, 2004, pp. 67 – 84. T.E. Gaskill, Was Leibniz an Avicennian? Leibniz and Avicenna on the Principle of Sufficient Reason, Proceedings of the Patristic, Medieval and Renaissance Conference, 16 – 17, 1992 – 3, pp. 105 – 13. . Gilson, Les sources greco-arabes de l’augustinisme avicennisant, Archives d’Histoire doctrinale et littraire du Moyen Age, 4, 1929, pp. 5 – 149. –– , Pourquoi saint Thomas a critiqu saint Augustin, Archives d’Histoire doctrinale et littraire du Moyen Age, 1, 1926 – 7, pp. 5 – 127. ˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

221

222

A.-M. Goichon, La philosophie d’Avicenne et son influence en Europe mdivale, 2nd edn, Paris: Maisonneuve 1951. M. Grabmann, I divieti ecclesiastici di Aristotele sotto Innocenzo III e Gregorio IX, Miscellanea Historiae Pontificiae, Vol. V, Collect. 7: I Papi del Duecento e l’Aristotelismo, fasc. I, Roma: Saler, 1941. A. Hasnawi, La conscience de soi chez Avicenne et Descartes, in J. Biard, R. Rashed, eds, Descartes et le Moyen ffge, Paris: Vrin, 1997, pp. 283 – 91. D.N. Hasse, Avicenna’s ‘De anima’ in the Latin West. The Formation of a Peripatetic Philosophy of the Soul, 1160 – 1300, London/Turin: The Warburg Institute/Nino Aragno Editore, 2000. –– , Arabic Philosophy and Averroism, in J. Hankins, ed., Cambridge Companion to Renaissance Philosophy, Cambridge: Cambridge University Press, 2007, pp. 113 – 36. –– , Der mutmaßliche arabische Einfluss auf die literarische Form der Universittsliteratur des 13. Jahrhunderts, in L. Honnefelder, ed., Albertus Magnus und der Ursprung der Universittsidee, Berlin: Berlin University Press, 2011, pp. 241–58 and pp. 487–91. –– , The Social Conditions of the Arabic-(Hebrew-)Latin Translation Movement in Medieval Spain and in the Renaissance, in A. Speer, ed., Wissen ber Grenzen. Arabisches Wissen und lateinisches Mittelalter, Berlin/New York: De Gruyter, 2006, pp. 68 – 86. H. Hugonnard-Roche, La classification des sciences de Gundissalinus et l’influence d’Avicenne, in J. Jolivet and R. Rashed, eds, tudes sur Avicenne, Paris: Les Belles Lettres, 1984, pp. 41 – 75. J. Jolivet, L’pistmologie de Descartes dans le Regulae et celle d’Avicenne, in J. Biard and R. Rashed, eds, Descartes et le Moyen ffge, Paris: Vrin, 1997, pp. 187 – 97. –– , The Arabic Inheritance, in P. Dronke, ed., A History of Twelfth-Century Western Philosophy, Cambridge: Cambridge University Press, 1987, pp. 113 – 48. H. Kishlat, Studien zur Verbreitung von bersetzungen arabischer philosophischer Werke in Westeuropa 1150 – 1400. Das Zeugnis der Bibliotheken, Mnster: Aschendorff, 2000. P. Mandonnet, Siger de Brabant et l’averro sme latin au XIIIe sicle, Fribourg (Suisse), 1899; 2nd edn: Louvain: Institut suprieur de philosophie, 1908 – 11 (2 vols). J.-M. Mandosio, C. Di Martino, La Mtorologie d’Avicenne (Kita¯b al-Sˇifa¯ V) et sa diffusion dans le monde latin, in A. Speer, ed., Wissen ber Grenzen. Arabisches Wissen und lateinisches Mittelalter, Berlin/New York: De Gruyter, 2006, pp. 406 – 24. J.K. Otte, Alfred of Sareshel’s Commentary on Avicenna’s De congelatione et conglutinatione lapidum, in G. Freibergs, ed., Aspectus et affectus. Essays and Editions in Grosseteste and Medieval Intellectual Life in Honor of R. C. Dales, New York: AMS Press, 1993, pp. 105 – 11. M. Rashed, Thodice et approximation: Avicenne, Arabic Sciences and Philosophy, 10, 2000, pp. 223 – 57. J.P. Reilly, The Alia Littera in Thomas Aquinas’ Sententia Libri Metaphysicae, Medieval Studies, 50, 1988, pp. 559 – 83. M.-D. Roland-Gosselin, Introduction, Notes et tudes historiques, in Le ‘De ente et essentia’ de S. Thomas d’Aquin. Texte tabli d’aprs les manuscripts parisiens, Paris: Vrin, 1926; 2nd edn: 1948. A. Rucquoi, Gundisalvus ou Dominicus Gundisalvi?, Bulletin de Philosophie Mdivale, 41, 1999, pp. 85 – 106. ˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Amos Bertolacci

On the Latin Reception of Avicenna’s Metaphysics before Albertus Magnus 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

223

C.B. Schmitt, D. Knox, Pseudo-Aristoteles Latinus. A Guide to Latin Works Falsely Attributed to Aristotle before 1500, London: The Warburg Institute, 1985. R. Sharpe, A Handlist of the Latin Writers of Great Britain and Ireland Before 1540, Turnhout: Brepols, 2001. M.J. Soto Bruna, Estudio fil sofico. Introducci n y comentario al De Processione Mundi, in Gundissalinus, De Processione Mundi. Estudio y edicin crtica del tratado de Domingo Gundisalvo, eds M.J. Soto Bruna, C. Alonso Del Real, Pamplona: Ediciones Universidad de Navarra, 1999, pp. 17 – 107. A. Speer, The Hidden Heritage: Boethian Metaphysics and Its Medieval Tradition, Quaestio, 5, 2005, pp. 163 – 81. R. Teske, William of Auvergne’s Debt to Avicenna, in J. Janssens and D. De Smet, eds, Avicenna and His Heritage. Proceedings of the International Colloquium ‘Avicenna and his Heritage’, Leuven-Louvain-la-Neuve, 8 – 11 September 1999, Leuven: Leuven University Press, 2002, pp. 153 – 70. S. Van Riet, La traduction latine, Annexes, in Avicenna Latinus, Liber primus naturalium. Tractatus primus de causis et principiis naturalium, ed. S. Van Riet, Peeters/Brill, Louvain/Leiden 1992, pp. 53*-105*. –– , Traduction latine et principes d’dition, in Avicenna Latinus, Liber de philosophia prima sive Scientia divina, I–IV, ed. S. Van Riet, Louvain/Leiden: Peeters/Brill, 1977, pp. 123*–66*. F. Van Steenberghen, La philosophie au XIIIe sicle, Louvain/Paris: Publications Universitaires-Nauwelaerts, 1966; 2nd edn: Louvain/Paris: Peeters, 1991. G. Verbeke, Avicenna’s Metaphysics and the West, in M. Wahba, ed., Islam and Civilization, Cairo: Ain Shams University Press, 1982, pp. 53 – 64. M.H. Vicaire, Le Porrtains et l’avicennisme avant 1215, Revue des sciences philosophiques et thologiques, 26, 1937, pp. 449 – 82. G. Vuillemin-Diem, Praefatio, in Aristoteles Latinus XXV 1 – 1a, Metaphysica lib. I-IV.4. Translatio Iacobi sive ‘Vetustissima’ cum Scholiis et Translatio Composita sive ‘Vetus’, ed. G. Vuillemin-Diem, Bruxelles/Paris: Descle de Brouwer, 1970, pp. xi–lix. –– , Praefatio, in Aristoteles Latinus XXV 2, Metaphysica lib. I–X, XII–XIV. Translatio Anonyma sive ‘Media’, ed. G. Vuillemin-Diem, Leiden: Brill, 1976, pp. ix–lxix. –– , Praefatio, in Aristoteles Latinus XXV 3.1, Metaphysica lib. I–XIV. Recensio et Translatio Guillelmi de Moerbeka, ed. G. Vuillemin-Diem, Leiden/New York/ Cologne: Brill, 1995. –– , Zum Aristoteles Latinus in den Fragmenten der Quaternuli des David von Dinant, Archives d’Histoire doctrinale et littraire du Moyen Age, 70, 2003, pp. 27 – 136. H. Yaldir, Ibn Sı¯na¯ (Avicenna) and Ren Descartes on the Faculty of Imagination, British Journal for the History of Philosophy, 17.2, 2009, pp. 247 – 78.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Avicenna’s ‘Giver of Forms’ in Latin Philosophy, Especially in the Works of Albertus Magnus Dag Nikolaus Hasse

˘

The giver of forms (wa-hib as-suwar, dator formarum) is a piece of Avicennian ˙˙ philosophy that went against the grain of most scholastic philosophers.1 It is part of Avicenna’s emanation theory, which the Latins knew from books 8 and 9 of the Ila-hiyya-t (Divine Things) of Kita-b asˇ-Sˇifa- (Book of the Cure): the emanation of intelligences and accompanying celestial spheres from the first cause, the necessary being (wa-g˘ib al-wu¯g˘u¯d, necesse esse), which is an eternal efficient cause. In a number of passages, Avicenna calls one of the celestial intelligences the ‘giver of forms’.2 He apparantly refers to the lowest intelligence, from which emanate the substantial forms of the sublunar world.3 This intelligence is called ‘the active intellect’ in other passages.4 The forms emanate from the lowest intelligence when the elemental mixture reaches a certain disposition towards a form. In most of his writings, Avicenna uses the concept of a giver of forms not in an epistemological but in an ontological sense: the wa-hib as-suwar is not the ˙˙ giver of intelligible forms, but of the forms that combine with prepared matter.5 1

˘

2

I am grateful for the advice of Amos Bertolacci, Jon Bornholdt, Katrin Fischer, Jçrn Mller, Adam Takahashi and for suggestions from the audiences in Menaggio, Jena and Berlin (Leibniz-Kreis), where the paper was presented. Research on this paper was funded by the Volkswagen Foundation. Avicenna mentions the ‘giver of forms’ in five passages outside the Ta liqa-t: (1) Avicenna, Metaphysics, c. IX,5, p. 335, line 18 (‘the principles giving forms’); (2) ibid., c. IX,5, p. 337, line 26 (‘When it becomes prepared, it attains the form from the giver of forms’). These two passages appear in the same wording in Avicenna’s Nag˘a-t (The Salvation). (3) Avicenna, al-Kawn wa-l-fasa-d (On Generation and Corruption), c. 13, p. 187, line 3 (‘the giver of forms’); (4) ibid., c. 14, p. 190, line 14 (‘the giver of forms’); this passage is cited below, see n. 44. (5) Avicenna, Fı¯ l-af a-l wa-l-infi a-la-t (On Actions and Passions), p. 256, line 10 (‘the giver of forms’). See also n. 7 below for one occurrence in the Da-nesˇna-me. Avicenna, Metaphysics, c. IX,5, p. 335: ‘It follows necessarily, then, that the separate intellects – rather, the last of them, which is close to us, is the one from which there emanates, in participation with the celestial movements, something having the configuration of the forms of the lower world … ’. Avicenna, Metaphysics, c. IX,4, p. 331: ‘This is the state of affairs in each successive intellect and each successive sphere, until it terminates with the active intellect that governs our selves’. As I have argued in my Avicenna’s ‘De anima’, pp. 187 – 9.

3

4 5

˘

˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

226

˘

An exception is Avicenna’s late treatise Ta liqa-t (Notes), where the expression appears more than twenty times in various contexts, some of them epistemological. In the Ta liqa-t, the wa-hib as-suwar supplies substantial forms ˙˙ in the first place, but also provides first principles of knowledge, the forms of the things known (suwar al-ma lu¯mat), an excellent moral disposition and the ˙ 6 In the inflationary usage of the expression ‘giver of actualisation of light. ˙ aza-lı¯’s Maqa-sid alforms’, the Ta liqa-t resemble a text by a later author: al-G ˙ the falasifa (Intentions of the Philosophers) of the late eleventh century AD. Here expression is used, for instance, in the context of the theory of odours and visual forms.7 It is likely, therefore, that the epistemological interpretation of the expression was developed by Avicenna toward the end of his life and adopted by ˙ aza-lı¯. When the scholastics refer to the dator some of his readers, such as al-G formarum, they do this in the context of theories of substantial forms and not of intelligible forms (with very few exceptions).8 In modern literature, however, Avicenna’s concept is often misrepresented as epistemological.9 Around 1160 in Toledo, Dominicus Gundisalvi translated the metaphysics part of asˇ-Sˇifa- into Latin under the title Liber de philosophia prima sive scientia divina. Among its first Latin readers in the twelfth and early thirteenth century, there are some who adopt central doctrines of Avicenna’s emanation system: Gundisalvi himself in his treatise De processione mundi and the anonymous author of The Book of First and Second Causes (Liber de causis primis et secundis). 10 But the great majority of the later scholastic tradition considers Avicenna’s emanation theory to be in conflict with the idea that the world is created. This creation is not a necessary process, it is argued, but depends upon ˘

˘

˘

˘

˘

˘ ˘

6 As shown by Janssens, The Notions, pp. 551 – 62, esp. pp. 554 – 7. ˙ aza-lı¯, Maqa-sid, p. 350, line 17; p. 352, line 14; p. 359, line 5; p. 369, line 12. One 7 al-G ˙ might suspect that the Maqa-sid reflect Avicenna’s original usage of the term, since ˙ Avicenna’s Danesˇna-me-ye Ala-¯ı (Philosophy for Ala- -al-Dawla) is the ultimate source of the Maqasid (see Janssens, Le Da¯nesh-Na¯meh, pp. 163 – 77). But, in fact, only one of the ˙ aza-lı¯’s text has a parallel in eight ˙occurrences of the term ‘giver of forms’ in al-G Avicenna’s Da-nesˇna-me (see Janssens, The Notions, p. 552; the Persian expression is: su¯ra ˙ dinanda). 8 Hasse, Avicenna’s ‘De anima’, p. 189, n. 620. Possible exceptions are the following: Thomas Aquinas, De veritate, q. 11 a. 1, p. 349 (‘formas omnes sensibiles esse ab agente extrinseco quod est substantia vel forma separata, quam appellant datorem formarum vel intelligentiam agentem’) and Anonymous (Van Steenberghen), Quaestiones de anima, 2.19, p. 228, line 47 (‘… et datricem intelligibilium et naturalium quam dixit [sc. Avicenna] motricem decimi orbis’). 9 Examples are: Weisheipl, Aristotle’s Concept, p. 150: ‘to be receptive of new concepts from the dator formarum, the “agent intellect”’; Dales, The Problem of the Rational Soul, p. 8: ‘intelligible objects provided by the Giver of Forms’. 10 Anonymous (de Vaux), Liber de causis primis et secundis. ˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Dag Nikolaus Hasse

Avicenna’s ‘Giver of Forms’ in Latin Philosophy 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

227

the will of God, and it is not dependent upon intermediaries such as angels and intelligences, and hence not upon a giver of forms.11 It is remarkable that, in spite of this, the dator formarum is often mentioned in Latin sources, well into the seventeenth century. The theory, which is usually attributed to Plato or Avicenna, was obviously thought to be important – so important that it could not be passed over in silence. I suspect that the reception was not so entirely negative as it appears. I have therefore been searching for authors and passages with a positive reaction to Avicenna’s theory of the giver of forms – indications that the theory was thought to be a strong theory, even if it was refuted. I start with a brief overview of the Latin fortuna of the concept and then discuss the rare positive reactions to it, four briefly – those of William of Auvergne, John Buridan, Marsilio Ficino and Tiberio Russiliano – and one at length: that of Albertus Magnus.

I The Latin fortuna of the Giver of Forms The Avicennian theory of the giver of forms never firmly set foot on Latin soil. This contrasts with Avicenna’s theory that the active intellect is a separate substance, which was adopted by a good number of authors, especially in the thirteenth century. Some of them identified this separate substance with God, thus forming what tienne Gilson has called the position of ‘Augustinisme avicennisant’. These authors combine Avicenna’s teaching of ‘abstractions emanating from the active intellect’ (De anima V,5) with Augustine’s theory of illuminatio. Early exponents of this current are Jean de la Rochelle, the Summa fratris Alexandri and Vincent of Beauvais; later in the thirteenth century, the active intellect was identified with God by Roger Bacon, John Pecham, Roger Marston, Vital du Four, and also Henry of Ghent (though only in parts of his work).12 As far as I can see, the epistemological current of ‘Augustinisme avicennisant’ did not have a parallel in ontology. The two Avicennian concepts, that is, the active intellect as the source of intelligible forms, and the giver of forms as the source of substantial forms, saw a very different Latin reception. It is remarkable that the dator formarum concept was unsuccessful even within the Franciscan tradition that favoured Avicennian epistemology. 11 See, for example, Thomas Aquinas, Summa contra gentiles, c. II.26, vol. 13, p. 332: ‘Per haec autem exluditur quorundam philosophorum positio dicentium quod ex hoc quod deus seipsum intelligit, fluit ab ipso de necessitate talis rerum dispositio: quasi non suo arbitrio limitet singula et universa disponat, sicut fides catholica profitetur.’ Ibid., c. II.42, vol. 13, p. 365: ‘Excluditur autem ex praedictis opinio Avicennae, qui dicit quod deus, intelligens se, produxit unam intelligentiam primam … Et sic inde procedens diversitatem rerum causari instituit per causas secundas’. 12 Hasse, Avicenna’s ‘De anima’, pp. 203 – 23.

228 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Dag Nikolaus Hasse

The scholastics preferred other explanations for the origin of forms. An early example is provided by a passage from Bonaventure, in the second book of his commentary on the Sentences, dating from about 1248.13 There are four opinions on the coming-to-be of forms (eductio formae in esse), Bonaventure says – and many similar divisions of opinions can be found in later scholastic literature,14 all the way until Francisco Suarez’ Metaphysical Disputations: 15 First, the theory of latitatio or latitudo formarum of Anaxagoras (as presented by Aristotle in Physics, 187a26–b7): the forms are latent in matter and are only made manifest by an agent. Second, the theory of more modern philosophers (philosophorum magis modernorum – here Avicenna is implied) that all forms derive from a creator. The efficient cause of everything is God; the particular causes only prepare matter for the reception of a form.16 Third, the position of Aristotle and of the doctores in philosophia et theologia that the forms are in the potentiality of matter and are made actual by the particular agent. There are two variants of this position, according to Bonaventure: either you say that the form derives from an agent which multiplies its own form, or – and this is opinion four – you say that the form is already in matter before it is actualized. Bonaventure favours this last position, the pre-existence of forms. One advantage of this position, in the eyes of Bonaventure, is that it accords with Augustine’s well-known theory of ‘seminal reasons’ which exist in matter (rationes seminales). The scholastic discussion of substantial generation is mainly about the last two alternatives: do the forms preexist in matter somehow, as Bonaventure, Albertus Magnus and others say, or: is the role of matter purely passive, as Thomas Aquinas insists?17 Whether there is a small or a large difference between 13 Bonaventure, In quatuor libros Sententiarum, lib. II dist. VII p. II a. 2 q. 1, pp. 197 – 8. 14 See the references in the editors’ scholion: Bonaventure, ibid., p. 200. 15 Suarez, Disputationes metaphysicae, disp. XV sect. II, pp. 505 – 12, esp. p. 508 (on Plato and Avicenna). 16 Bonaventure, ibid., p. 198: ‘Alia fuit positio philosophorum magis modernorum, quod omnes formae sunt a creatore. Et haec positio potest dupliciter intelligi: uno modo quod deus sit principaliter agens et producens in omnis rei eductione, et sic habet veritatem; vel ita quod deus sit tota causa efficiens, et agens particulare non faciat nisi materiam adaptare, ut sicut producit animam rationalem, ita et alias formas; et iste intellectus videtur fuisse illorum philosophorum. Et iste intellectus est impossibilis, quia agens particulare aut inducit aliquid aut nihil. Et si nihil, ergo nihil agit. Si aliquid inducit, ergo videtur quod aliquam efficiat dispositionem; sed qua ratione potest in unam et in aliam? Quare ista positio non est rationabilis.’ 17 See, e. g., Thomas Aquinas, De veritate, q. 11 a. 1, p. 350: ‘Et ideo secundum doctrinam Aristotilis via media inter has duas tenenda est in omnibus praedictis: formae enim naturales praeexistunt quidem in materia, non in actu, ut alii dicebant, sed in potentia solum de qua in actum reducuntur per agens extrinsecum proximum, non solum per agens primum, ut alia opinio ponebat’.

Avicenna’s ‘Giver of Forms’ in Latin Philosophy 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

229

the position of Albertus and Thomas is a matter of dispute in modern scholarship, but that need not concern us here.18 Both sides of the medieval discussion jointly maintained that all substantial forms are educed from matter, with the exception of human souls, which are created by God. This is the general picture of the reception of Avicenna’s theory. Let me point to some distinctive features of it. A noteworthy feature is the association of the giver of forms with the term colcodea (or colcodrea or colchodea).19 This is a Latin term derived from Arabic astrological literature. In many Arabic sources, the length of life is calculated by al-kadhuda-h, a planet with specific attributes ¯ on the birth chart. The source of the˘ association of the two concepts was probably Pietro d’Abano, the early fourteenth-century author of the Conciliator, as Bruno Nardi has shown.20 Pietro explains the astrological term alcocoden with a reference to the giver of forms: quando [scil. Saturn and Mars] fuerint alcocoden, idest datores formarum vitae. 21 In Pietro’s eyes, the alcocoden can be called a ‘giver of forms’ because it is a source of life and a secondary cause (the first and universal cause being the heaven); both are true also of the giver of forms. The term colcodea was presumably derived from alcocoden, but not (according to the present state of our knowledge) by Pietro d’Abano, since a phrase with the term cholcodea in Pietro’s Conciliator is missing in the earliest two Renaissance editions and in at least three manuscripts and thus seems to be a Renaissance addition.22 It is usually assumed that Agostino Nifo coined the term for his commentary on Averroes’ Destructio destructionum of 1495. But in fact Niccolo Tignosi already uses colcodrea several times in 1474 when discussing Avicenna’s giver of forms theory in his De anima commentary, as in the sentence: ‘Avicenna has posited the Colcodrea, that is, an intelligence which is the giver of forms to the things below’ (ipsum posuisse colcodream, id est, intelligentiam datricem formarum istis inferioribus).23 Later, the term also appears in texts by Agostino Nifo,24 Pietro Pomponazzi,25 Marcantonio Zimara,26 18 On this difference see Suarez, Disputationes metaphysicae, disp. XV sect. II, p. 506; Nardi, La dottrina, pp. 69 – 101, esp. pp. 92 – 8; Weisheipl, The Axiom, pp. 455 – 6. 19 On this term see Nallino, La ‘Colcodea’ d’Avicenna, pp. 84 – 91, Wolfson, Colcodea, pp. 573 – 6, and Porro, Colcodea, pp. 2009 – 2010. 20 Nardi, Origine, p. 234, n. 1. 21 Pietro d’Abano, Conciliator, diff. 10, propter tertium, fol. 16vb. 22 This was again shown by Nardi. See Pietro d’Abano, Conciliator, diff. 71, propter tertium, fol. 108rb (‘quam cholcodeam vocabat’) and Nardi, Origine, p. 235, n. 1. 23 Tignosi, In De anima commentarii, p. 384: ‘Ad Avicennam dicatur ipsum posuisse colcodream, id est, intelligentiam datricem formarum istis inferioribus et ipsam irradiare supra intellectum possibilem’. 24 Nifo, In librum Destructio destructionum Averroys, fols 97vb, 98ra (‘dator formarum latine, arabice vero colcodea’), 101ra ; Nifo, In librum de anima, c. II.34. 25 Pomponazzi, In XII Metaphysicae, quoted from Nardi, Origine, p. 234n: ‘… et tandem venit ad Colchodeam omnium creatricem formarum istorum inferiorum; quae

230 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Dag Nikolaus Hasse

Tiberio Russiliano27, Julius Caesar Scaliger,28 Daniel Sennert29 and, in the early seventeenth century, Tommaso Campanella.30 It was employed because it was thought to be the original Arabic term. As Agostino Nifo writes: colchodea quam latine dator formarum exponitur: ‘Colcodea, which in Latin is rendered as dator formarum.’31 The term also appears in Hebrew sources of the sixteenth and seventeenth centuries that are influenced by Nifo.32 Another remarkable feature is that the Latin reception of Avicenna’s concept was much influenced by Averroes. Averroes, in his Long Commentary on the

26 27

28 29

30

31 32

Colchodea est intellectus agens. … Avicenna autem, quia tenet quod formae substantiales non possunt agere immediate et quia accidentia non possunt agere substantis, ideo oportet ponere datorem formarum. Non enim ipse videbat agens immediatum formarum, quia non accidens neque substantia; ergo est Colchodea’; and Pomponazzi, Utrum deus concurrat …, quoted from Nardi, Origine, p. 234: ‘… aut in opinionem Avicennae, qui tenuit quod immediate [sc. anima creetur] a Colcodea …’, and p. 237: ‘… non quia creetur anima nostra a deo, ut solvit Scotus, aut hai Colcodheia, ut voluit Avicenna’. Zimara, Contradictionum solutiones, fol. 421vb : ‘… quia secundum Avicennam aliae formae, quae de novo inducuntur in materia, non sunt eductae de potentia materiae, sed sunt ab extrinseco motore, quem datorem formarum appellat seu colcodeam’. Russiliano, Apologeticus, disp. 5, p. 177: ‘… cum illa [sc. anima rationalis] secundum Avicennae mentem fuerit infusa ex colcodea omnium formarum generatrice; modo colcodea, dum sufficientem dispositionem in materia habeat, semper formam inducit, vel illa sit ex seminis habita praeparatione vel ex putrefactione’. I discuss Russiliano’s reception of the giver of forms theory in section II of this article. Scaliger, De subtilitate, c. 97, p. 333: ‘Colcodea, nescio quae, ut aiunt, ab Avicenna ficta est, quae formarum conda, et proma, imo vero fabra esset, tuus iste liber, qui etiam dictamo putrido vitam molitur, etiam sesquicolcodea dici mereatur’. Sennert, Hypomnemata physcia, lib. IV c. 2, p. 150: ‘Avicennas animas viventium non a parentibus, sed a quadam formarum datrice seu, ut Scaliger, exerc. 97. loquitur, formarum proma conda intelligentia, quam Colcodeam nominat, provenire statuit … Procul dubio autem istam sententiam ex Platone et Platonicis hausit Avicennas.’ See Hirai, Atomes vivants, pp. 479 – 80. Campanella, De homine, c. I.1, p. 14: ‘Avicenna autem Colchodeae hoc munus permandat utenti elementorum materia et qualitatibus: propterea putat omnia animalia et homines posse oriri sponte, sicubi tellus sit apta ad Colchodeae sigillum suscipiendum, quod philosophi multi, licet animae mundi vel casui hoc opus adscribant, olim et nunc docent’; ibid. c. V.5, p. 70: ‘Verum cum oblivio contingat et scientia deleatur, putavit Avicenna quod, licet ab ideis sit scientia, non tamen, inquit, ab innatis (sic enim nulla fieret oblivio), sed a defluentibus a Colchodea, quae sit ultimus intellectus aut anima mundi secundum alios, in nostram animam, quae a sensibus movetur ut ad illas respiciat ideas’; ibid., c. V.5, p. 78: ‘… per quas excitatur ad species ex Colchodea effluentes considerandas’. Nifo, In librum Destructio destructionum Averroys, fol. 97vb. Cf. n. 24 above. See Wolfson, Colcodea, pp. 573 – 6. On the Hebrew reception of the giver of forms theory, see Goldstein, Dator formarum, pp. 107 – 21.

Avicenna’s ‘Giver of Forms’ in Latin Philosophy 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

231

Metaphysics, VII.31 and XII.18,33 criticizes Avicenna for holding that all substantial forms derive from the active intellect, ‘which he calls “giver of forms”’.34 Averroes refutes the theory and adds that al-Fa¯ra¯bı¯ and Avicenna are in fundamental agreement with Plato on this issue. As a result, Latin scholastics often attack Plato and not Avicenna for holding the dator formarum theory.35 Averroes’ association of the giver of forms with Plato influenced the understanding of Avicenna’s theory in the West.36 Latin knowledge of Plato was, for the most part, confined to the Timaeus. And hence, in the scholastic view, the Platonic standpoint was that the forms are given by the second gods of the Timaeus, as Albertus Magnus puts it: ‘everything is generated by the second gods (a diis secundis), who were given the seed of generation by the god of gods (deus deorum)’37 – or, in the Renaissance interpretation, by the world soul, anima mundi. 38 Hence it came that Plato’s anima mundi and Avicenna’s colcodea were thought to mean the same and that Plato and Avicenna were considered to be the major exponents of a giver of forms theory. That the Platonic association tainted the understanding of Avicenna’s Metaphysics is evident in that the scholastics often use the term creare for the activity of Plato’s and Avicenna’s dator formarum. An example is provided by the passage by Bonaventure which I cited above: the second of the four possible positions on the generation of forms was the theory that all forms derive from a 33 Averroes, Tafsı¯r, c. VII.31, pp. 881 – 6, c. XII.18, pp. 1496 – 8; Averroes, Commentarium in libros Metaphysicorum, fols 181ra-vb, fols 304ra-va. The latter passage is translated in: Genequand, Ibn Rushd’s Metaphysics, pp. 107 – 9. 34 Averroes, Tafsı¯r, c. VII.31, p. 882, line 19; Averroes, Commentarium in libros Metaphysicorum, fol. 181ra : ‘Et ideo quia Avicenna oboedit istis propositionibus, credidit omnes formas esse ab intelligentia agente, quam vocat datorem formarum’. 35 Hasse, Plato Arabico-Latinus, pp. 42 – 5. For the early reception of Averroes’ commentary in general see Bertolacci, The Reception, pp. 457 – 80. 36 Averroes’ understanding of Plato was influenced by Themistius (d. 388 AD), who had argued in his paraphrase of Aristotle’s Metaphysics, book XII, that the forms of the living beings are implanted in matter by the gods and that this is Plato’s theory; see Hasse, Spontaneous Generation, pp. 154 and 158 – 9. 37 Albertus, Metaphysica, lib. II tr. 1, c. 8, p. 468b: ‘Platonis igitur sententia est omnia fieri a diis secundis, quibus deus deorum dedit sementem generationis. Dii autem secundi sunt stellae et orbes caelestium moventes materiam ad omnium generabilium productionem. Et illa sementis dicitur forma quaedam formans materiam ad conveniens sibi in nomine. Hanc enim et huiusmodi formam dicit communicari materae per datorem formarum et ipsam materiam aptari formae recipiendae per qualitates activas et passivas’. 38 Nifo, Expositiones in libros Metaphysices, lib. 7 disp. 12, p. 201r : ‘Virtus autem generandi est in anima mundi apud Platonem, quae ab Avicenna dicitur cholchodea.’; Campanella, De homine, c. V.5, p. 70: ‘Colchodea, quae est ultimus intellectus aut anima mundi secundum alios’. Cf. the pseudo-Paracelsian Apocalpysis Hermetis of ca. 1560, as quoted by Jantz, Goethe’s Faust, p. 176: ‘Dieser Geist [sc. the quintessence] wirdt von Avicenna genandt die Seel der Welt’.

232

˘

creator.39 But ‘creation’ does not adequately describe Avicenna’s theory of emanation. He had redefined the term ‘creation’ (ibda- ) as the permanent causation of the existence of a thing, in Ila-hiyya-t, chapters VI.240 and VIII.341. There is existence after non-existence, but the posteriority is essential, not temporal. The giver of forms does not create forms, but continuously reacts with the emanation of forms if the material disposition in the sublunar world requires it. When the elemental qualities change and exceed certain limits, argues Avicenna, matter becomes disposed towards a new form, which flows upon matter from the giver of forms: ‘The augmentation and reduction [of the elemental qualities] has two well-defined limits; when they are exceeded, the entire disposition of the matter towards its form is extinguished, and it becomes completely disposed towards a different form. It is characteristic of matter that when it is completely disposed towards a form, that this form flows upon the matter from the giver of forms to matter, and that it receives this form.’42 The Latin term creare fails to capture the necessity and automatism of the process. The Platonic colouring of Avicenna’s theory is obvious in Thomas Aquinas’ presentation of it: Plato and Avicenna, says Thomas in De potentia, posit an agens supernaturale, ‘a supernatural agent’, which is able to produce ex nihilo. 43 This, however, is in disaccord with Avicenna, who clearly holds that the active

˘

39 See n. 13 above. Another example is Albertus, Super Dionysium De divinis nominibus, c. 4, p. 194b: ‘Quidam enim, ut Plato et Avicenna et plures alii, formas dicunt advernire ab extrinseco, ponentes eas dari a datore, et sic ponebant esse per creationem, non quod non fiant in aliquo subiecto, sed quia non fiunt ex aliquo suae essentiae praeexistente.’ 40 Avicenna, Metaphysics, c. VI.2, p. 203: ‘This, then, is the meaning that, for the philosophers, is termed “creation” (ibda- ). It is the giving of existence to a thing after absolute nonexistence. For it belongs to the effect in itself to be nonexistent and [then] to be, by its cause, existing. That which belongs in the thing intrinsically is more prior in essence for the mind ([though] non in time) than that which belongs to it from another. Hence, every effect constitutes an existence after non-existence, in terms of essential posteriority’. Cf. ibid., p. 204: ‘It is good [however] to call everything not coming into existence from a previous matter not “generated” (mutakawwin), but “created” (mubda ).’ For context, see Marmura, Efficient causality, p. 184. 41 Avicenna, Metaphysics, c. VIII.3, p. 272: ‘This is the meaning of a thing’s being created (mubda ) – that is, attaining existence from another. … Thus, origination from absolute nonexistence, which is creation, becomes false and meaningless [sc. if this posteriority were temporal]. Rather, the posteriority here is essential posteriority’. 42 Avicenna, al-Kawn wa-l-fasa-d, c. 14, p. 190; Avicenna, Liber tertius naturalium de generatione et corruptione, c. 14, p. 139. 43 Thomas Aquinas, De potentia, q. 3 a. 8, p. 61: ‘Et quia operatio naturae non potest esse ex nihilo, et per consequens oportet quod sit ex praesuppositione, non operabatur secundum eos natura nisi ex parte materiae disponendo ipsam ad formam. Formam vero, quam oportet fieri et non praesupponi, oportet esse ex agente qui non praesupponit aliquid, sed potest ex nihilo facere; et hoc es agens supernaturale, quod Plato posuit datorem formarum. Et hoc Avicenna dixit esse intelligentiam ultimam inter substantias separatas’. ˘

˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Dag Nikolaus Hasse

Avicenna’s ‘Giver of Forms’ in Latin Philosophy 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

233

intellect is part of nature. Like all existents of the sublunar world, it is a possible existent per se and a necessary existent only through something else. Avicenna in this sense downplays very much the difference between the supralunar and the sublunar world – a position which Averroes criticizes with vehemence in the Taha-fut at-Taha-fut, because he prefers to describe the supralunar world, in contrast to the sublunar, as ‘necessary through its substance’.44 The term supernaturale shows that the gist of Avicenna’s theory was lost to Thomas Aquinas, partly because it was understood through the eyes of Averroes.

II Positive Reactions: William of Auvergne, John Buridan, Marsilio Ficino and Tiberio Russiliano If Avicenna’s theory never set firm foot in the West, what was the context in which it nevertheless was found attractive? The first context is theories which attribute a greater power of daily creation to God. As was said above, the principal scholastic line was to reserve the generation of souls to God, whereas all other forms are educed from matter. William of Auvergne and John Buridan diverge from the mainstream position in that they extend God’s role to the forms of all animate beings. William of Auvergne, who is writing in the 1230s, is among the first readers of Avicenna’s Metaphysics, after the translator, Dominicus Gundisalvi.45 William adopts from Avicenna the description of God as the necesse esse per se 46 and as that whose existence is its being.47 But William at the same time criticizes the Arabic followers of Aristotle, as he calls them, for denying the freedom of the creator48 and for describing his creative activity as eternal.49 He also rejects the

44 Averroes, The Incoherence of the Incoherence, vol. 1, p. 238. 45 See also the article by Amos Bertolacci in this volume. On Avicenna’s influence on William’s metaphysics, see the articles collected in Teske, Studies. 46 William of Auvergne, De universo, IIa IIae c. 10, p. 853b: ‘Proprium nomen vero seu propria nominatio est quam impossibile est naturaliter praedicari de multitudine, quare necesse esse per se est propria nominatio ipsius’; id., De trinitate, c. 3, p. 25: ‘Iam igitur incipit nobis elucere ens essentiale esse necesse, aeternum et incorruptibile, non causatum.’ 47 Willliam of Auvergne, De trinitate, c. 1, p. 17: ‘… ens, cuius essentia est ei esse et cuius essentiam praedicamus cum dicimus “est”’. See Teske, Individuation, p. 77. 48 William of Auvergne, De universo, Ia Iae c. 27, pp. 623b-4a. The critique is levelled against ‘sequaces Aristotelis et qui famosiores fuerunt de gente Arabum in disciplinis Aristotelis’ (ibid., p. 618b). 49 Willliam of Auvergne, De trinitate, c. 10, p. 66: ‘… opinati sunt inter philosophantes praecipui, scilicet peripatetici, eiusdem operationes aeternas esse’.

234 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Dag Nikolaus Hasse

idea that the tenth intelligence is the source of the causation of many things and of the human souls in particular.50 Despite this criticism, Avicenna’s influence is still felt. Everything is educed into being and falls back into non-being through God, William says, or through an intermediate cause dependent upon God. There is no being in the world which is not from God and not sustained through him51 – which reminds one of the Avicennian ontological theory of causation. God fills the world in the way the light of the sun illuminates the universe. At first sight, William simply appears to continue a Christian tradition holding with the apostle Paul that ‘everything is from him, through him and in him’ (omnia ex ipso, per ipsum et in ipso, Rom. 11:36).52 But, in fact, William attributes to God what Avicenna had claimed for the active intellect: according to William, God reacts upon the preparedness of matter by giving forms fitting to that part of matter. A telling case are animals that are generated spontaneously, that is, without there being any parents, such as worms in decay: These animals are created a virtute omnipotentissima creatoris: [The fire which is said to lead to generation] prepares matter by removing from it the dispositions which offer resistance to the generated form and which deter it from the matter in which they are. But the most generous and virtuous goodness of the creator is ready to immediately give the form (dat formam) which is adequate to the part of matter. And this appears clearly in the generation of animals. Wherever matter is prepared to receive life or soul, the creator immediately infuses it into the matter. There is no room for any idiocy whatsoever to hallucinate or feign that there is some power in cheese or wood or in very solid rock which could infuse or bring life or soul into the aforementioned animals [sc. animals generated without parents].53 50 See Teske, Individuation, pp. 84 – 5. 51 William of Auvergne, De universo, Ia Iae c. 27, p. 624a: ‘Non intellexerunt … fortitudinem virtutis eius [sc. creatoris] qua attingit a summo universi usque deorsum … omnia continens, tenens et retinens, prout vult et quamdiu vult, alioquin reciderent in non esse, unde educta sunt ab ipso et per ipsum’; id., De trinitate, c. 5, p. 35: ‘Omne igitur ens debet suum esse et omne ens debet se primo enti, cum non sit ens nisi ab ipso et per ipsum, et per hoc manifestum est, quod universum est fluxus et exuberantia esse eius, quod est fons universalis essendi.’ Cf. ibid., p. 45, line 11; p. 47, line 46. 52 Cited in: William of Auvergne, De trinitate, c. 7, p. 48. 53 William of Auvergne, De anima, c. V.1, p. 112a-b: ‘[sc. ignis qui dicitur generans ad generationem] materiam praeparat removendo ab ea dispositiones quae repugnant formae generati et prohibent eam a materia in qua sunt. Praesto autem est largissima bonitas ac virtuosissima creatoris quae in materiae parte statim dat formam convenientem illi. Et hoc apparet evidenter in generationibus animalium; ubicumque enim materia parata est ad recipiendum vitam vel animam, statim eam illi creator infundit. Non enim est qualiscunque desipientia delirare vel fingere virtutem aliquam in caseo vel ligno esse vel rupe durissima quae vitam vel animam praenominatis animalibus infundere valeat vel praestare.’

Avicenna’s ‘Giver of Forms’ in Latin Philosophy 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

235

It is apparent that Avicenna’s theory of the giver of forms had several advantages for William: it is part of a system of causation in which the entire universe is understood as permanently dependent upon God; it explains generation with the preparedness of matter for the reception of life; and it offers a solution for the problem of spontaneous generation.54 I suspect that these advantages were also seen in the later scholastic tradition. In the fourteenth century, commentaries on Aristotle’s Metaphysics book 7 sometimes discuss a question with the title: ‘Whether because of the generation of inferior substances it is necessary to posit separate substances’. Unfortunately, only a few Metaphysics commentaries have been published, so that it is difficult to spot Avicennian influences. I am aware of two authors addressing the question directly: John of Jandun and John Buridan. John of Jandun (d. 1328) flatly rejects Avicenna’s theory, in the footsteps of Averroes’ critique: In view of this, one has to answer to this question in accordance with Aristotle and the Commentator that it is not necessary to posit abstract substances, such as ideas or a giver of forms, for the sake of the generation of inferior beings, and this is shown by four arguments of the Commentator …55

John Buridan (d. 1361), in contrast, takes the opposite position: One has to answer to this question that the most important reason, it seems, for concluding that there are separate substances (or at least one separate substance) can be drawn and inferred from the generation of the sense-perceptible substances.56

The principal argument in support of this conclusion is that spontaneous generation cannot be explained without assuming the existence of separate substances. It is not sufficient to assume that material principles in combination with heavenly bodies are responsible for the generation of the forms of inferior substances. The material principles do not have the degree of perfection which a substantial form has, and hence there must exist an immaterial generating principle (principale generans), which produces the substantial forms. This principle is God57 – and not ideas, as Plato thought. 54 For the history of the spontaneous generation problem in Greek, Arabic and Latin philosophy, see Hasse, Spontaneous Generation, pp. 150 – 75 (on William pp. 162 – 3). 55 John of Jandun, In duodecim libros Metaphysicae, lib. 7 q. 22, p. 101vb : ‘His visis dicendum ad quaestionem secundum intentionem Aristotelis et Commentatoris quod non oportet ponere substantias abstractas, ut ideas vel datorem formarum, propter generationem inferiorum, et hoc probatur quatuor rationibus Commentatoris …’ 56 Buridan, In Metaphysicen, lib. VII q. 9, fol. 46va : ‘Ad questionem respondendum est quod sicut mihi videtur ratio maxima ad concludendum substantias separatas vel saltem substantiam potest sumi et argui ex generatione substantiarum sensibilium.’ 57 Buridan, In Metaphysicen, lib. VII q. 9, fol. 46vb : ‘Illa substantia separata assistit presentialiter et indistanter toti mundo et cuilibet eius parti, et sic erat sufficienter simul

236 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Dag Nikolaus Hasse

The thrust of this argument is that substantial forms, even if they are the forms of inferior substances, cannot be generated by material principles only. Authors of the scholastic mainstream position would reply that Buridan mistakenly thinks that forms are generated – whereas in fact only the compound of form and matter is generated. This argument comes from Aristotle’s Metaphysics Zeta 8 (1033b17 – 18). Since there is no generation of forms, there is no need to posit a giver of forms. All we need are material principles, because inferior substances like animals, plants and stones, always remain within the limits of natural agency, as Thomas Aquinas puts it (De potentia q. 3 a. 11, Summa theologiae Ia q. 118 a.1). Buridan would probably reply that, even in a compound, the formal information has to have an origin that is not material. When turning to the Renaissance, it is difficult to encounter a favourable attitude towards the Avicennian theory of the giver of forms, even within Renaissance Platonism. The dator formarum theory is mentioned regularly, but is usually refuted. Marsilio Ficino (d. 1499) is an exception to this trend. He praises Avicenna as ‘the prince of the Arabic theologians’58 and Avicenna and al˙ aza-lı¯ as thorough friends of Plato.59 In the Platonic Theology, he refers G approvingly to Avicenna’s theory that the substantial forms are imprinted by ‘some divine mind’, a producer of forms (formatrix), into properly disposed matter and that likewise the human mind, when properly disposed, turns toward a divine intelligence, by which it is ‘informed’. Ficino’s true interest, however, is not Avicenna’s theory of substantial generation, but of intellectual knowledge. The context is epistemological: Ficino seeks support for his claim that we cannot attain intellectual knowledge if we are not ‘informed’ by divine ideas.60 cum materia rane, dico simul per indistantiam, ita quod ipsa posset ex illa materia producere formam substantialem rane. Et credo quod illa substantia separata est ipse deus omnipotens’. 58 Ficino, Theologia platonica, c. XV.2, p. 26: ‘Avicenna theologorum arabum princeps’. Avicenna is cited as holding that the mind is a form of the body while also being incorporeal. His hierarchy of intelligences is approvingly referred to a little later (ibid., p. 40): ‘Praeterea multum probanda videtur distinctio illa platonica in Metaphysicis Avicennae, videlicet in mundo intelligibili procedendum esse ab intelligibili summo ad intellectus multos, tamquam a formatore ad vires inde formabiles.’ 59 See next n. 60 Ficino, Theologia platonica, c. XII.1, pp. 10 – 11: ‘Huic autem Platonicorum mysterio similis ex quadam parte videtur esse Avicennae Algantelisque opinio. Opinantur enim materiam tum elementalem tum intellectualem sub luna divinae cuidam menti tamquam formatrici subesse, cuius instrumenta sint ad elementalem materiam disponendam formae corporeae, sed ipsa tandem praeparatae materiae substantiales imprimat formas. Similiter humanam mentem per imagines corporum per sensus phantasiamque acceptas ita saepe disponi, ut in divinam illam intelligentiam se convertat atque ab illa quatenus convertitur eatenus formari quotidie … Sed mittamus Arabes in praesentia, quamvis Platoni satis amicos. Ad Platonica redeamus.’

Avicenna’s ‘Giver of Forms’ in Latin Philosophy 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

237

One author, however, shows open sympathies for the ontological side of Avicenna’s theory: Tiberio Russiliano (d. after 1519). Russiliano, in a series of public disputations in 1519, defended a number of provocative philosophical theories: on the value of magical knowledge about Christ, on the eternity of the world, and on the trinity. He barely escaped the inquisitorial prosecution which followed. His fifth disputation discusses phenomena of spontaneous generation, among which he counts the first human being ever – at least, he says, ‘if we discuss the case in purely natural terms’ (cum phisice tantum disputemus) – and Americans, because the human beings on these newly discovered islands cannot have reached them by boat.61 In this context, he defends Avicenna’s theory of the spontaneous generation of human beings as most probable philosophically. It is sensible to assume that spontaneous generation is the result of material mixtures that trigger the deliverance of forms from the first craftsman and creator, which, Russiliano finds, is equivalent to Avicenna’s colcodea: In accordance with Aristotle, we can argue in two ways, either by holding that the rational soul is perishable and mortal … or by holding that it is immortal, and then we will say that just like the Colcodea necessarily creates appropriate forms due to a certain disposition of matter, likewise the first craftsman produces a rational soul, be it in the semen or in decay, as long as both preparations are sufficient for attaining it.62

Again, as in Buridan, spontaneous generation is a problem that led Western thinkers to adopt Avicenna’s giver of forms theory.63 The alternative was to say with Averroes and Thomas Aquinas that spontaneous generation is due to the influence of the stars. But then it remains unclear where the formal information comes from that explains the generation of a specific animal.

61 Russiliano, Apologeticus, disp. 5, p. 174: ‘… hominibus noviter in insulis incognitis repertis …’, and p. 175: ‘Unde secundum omnem philosophie semitam cogimur dicere hominem et cuncta animalia ex terra habuisse originem’. Cf. also p. 177: ‘… cuncta animalia tum perfecta tum imperfecta prima generatione ex putrefactione prodiere’. 62 Russiliano, Apologeticus, disp. 5, p. 177: ‘Secundum vero Aristotelis mentem dupliciter dicere possumus vel tenendo illam [sc. animam rationalem] esse caducam et mortalem … vel illam esse immortalem, et sic dicemus quod sicut colcodea necessario habita materiae dispositione creat convenientes formas, sic etiam erit de primo opifice, quod ita indifferenter producit animam rationalem in semine et putrefactione, dummodo ambae praeparationes sint sufficientes ad illam capescendam.’ 63 See Hasse, Spontaneous Generation, and, with special reference to the Renaissance: Hasse, Arabic Philosophy and Averroism, pp. 125 – 9. Cf. also Thorndike, History, vol. 5, p. 236, on the mentioning of Avicenna’s datrix formarum in an astrological prediction for 1521.

238 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Dag Nikolaus Hasse

III Albertus Magnus: the Origin of the Intellectual Soul Albertus Magnus’ attitude towards the giver of forms theory is interesting because it is not clearcut and because over his long and very productive life Albertus comes back to the topic several times.64 He was instrumental in formulating the scholastic mainstream position on the generation of forms: that only the rational soul comes from outside, ab extrinseco. This position is formulated already in his early De homine of 1243,65 but it also appears in later works, e. g., in the commentary on the Metaphysics, which dates ca. 1262.66 The formula ab extrinseco derives from Aristotle’s Generation of Animals, chapter II.3: ‘It remains that intellect (nous) alone enters from outside (thurathen) and it alone is divine’ (736b27 – 28). One problem for Albertus is to specify this external cause. In De homine, he says, like many scholastics after him, that the rational soul is created by God.67 But Albertus changes his position on this point, and one can observe that the theological interpretation that God is the origin of the souls is rivalled by an Avicennian interpretation that the origin is the separate active intellect. In De animalibus (dating 1256 – 60), Albertus interprets Aristotle’s term ab extrinseco variously as meaning: ‘from the light of the active intellect’, or: ‘from the light of the first active intellect’, or: ‘from the principle of generation, which is the intellect whose work is the work of nature’.68 Some interpreters conclude that the reference here is to the divine

64 Albertus’ reception of this Avicennian doctrine is discussed in: de Libera, Albert le Grand et le Platonisme, pp. 101 – 2; de Libera, Albert le Grand ou l’Antiplatonisme, pp. 260 – 62; de Libera, Mtaphysique et notique, pp. 156 – 68; Anzulewicz, Pseudo-Dionysius, pp. 258 – 64; Hasse, Avicenna’s ‘De anima’, pp. 188 – 9; Hasse, Plato-Arabico-Latinus, pp. 42 – 5; Takahashi, Nature, p. 476. On the textual traces of Avicenna’s Prima philosophia in Albertus Magnus’ Metaphysica see Bertolacci, ‘Subtilius speculando’, pp. 261 – 339, and id., Le citazioni implicite, pp. 179 – 274. 65 See n. 67 below. 66 Albertus, Metaphysica, lib. 2 tr. 1 c. 9, p. 473b: ‘Ex his etiam patet, cum omnes formae educantur de potentia ad actum, sicut diximus, et solus intellectus adeptus sit ingrediens ab extrinseco, quod …’. 67 Albertus, De homine, c. 3.3, p. 141, line 7: ‘Dicimus quod anima rationalis, hoc est anima hominis cum omnibus potentiis suis, hoc est vegetabilibus et sensibilibus et rationabilibus, non est in semine sicut in effectivo neque per substantiam, sed creatur a deo et infunditur corpori.’ 68 Albertus, De animalibus, lib. XVI tr. 1 c. 11, p. 1094, § 63: ‘Propter quod a toto extrinseco materiae spermatis et virtutum eius a luce intellectus qui secundum Anaxagoram et Aristotelem est primum agens in omnibus praeinductis virtutibus, in conceptum inducitur anima rationalis et intellectualis’; ibid., § 64: ‘Et ideo principium ipsius nichil aliud est nisi lux primi intellectus agentis. Intellectus enim hic purus est et immixtus et impassibilis omnino, sicut ostendimus in libro tertio De anima’; ibid., c. 12, p. 1096, § 67: ‘Sequitur necessario quod ipse solus sit ab extrinseco materiae datus a

Avicenna’s ‘Giver of Forms’ in Latin Philosophy 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

239

intellect, which is called ‘active’ in this context.69 But Albertus himself adds that this is the intellect which he has described as purus, immixtus and impassibilis (following Aristotle, 430a17 – 18) in his own De anima, where the active intellect is clearly distinct from the First Cause.70 Albertus, in his own De anima, which was written in the same years as De animalibus, that is, in the late 1250s, explicitly and approvingly uses the term dator formarum: Likewise, the claim (of Alexander of Aphrodisias) that the intellectual soul is educed from semen, is completely false. Rather, (the intellectual soul) enters from outside from a giver (dator) into matter, and it is the likeness of the giver of forms (similitudo datoris formarum), which is the first intelligence and unmixed with the body, whereas the power or form of the body is mixed, and hence this (argument) too is wrong. It is true, however, that (the intellectual soul) is the aim of generation, but this aim is not brought about in matter through the power of primary qualities which transform matter, but an intelligence is giving it when the matter is (properly) disposed through natural principles.71

This is one of the very few passages in scholastic literature where the ontological concept of the giver of substantial forms is used approvingly. Note that it is combined with Avicenna’s doctrine of the preparedness of matter: quando materia est diposita per principia naturalia. Earlier in the same treatise, Albertus had formulated the ab extrinseco theory not in Avicennian terms, but with reference to ‘some philosophers’, in fact to the Neoplatonic Liber de causis translated from Arabic, holding that ‘the soul is created by mediation through an intelligence’.72

69 70 71

72

principio generationis quod materiae non commiscetur, et hoc est intellectus cuius est opus naturae sicut primo moventis et causantis.’ Weisheipl, The Axiom, p. 451. See n. 68 above. Albertus, De anima, lib. 3 tr. 2 c. 4, p. 183: ‘Similiter autem, quod dicit, quod intellectualis anima educatur de semine, falsum est omnino, sed potius ipsa est ingrediens ab extrinseco a datore in materiam, et est similitudo datoris formarum, qui est intelligentia prima et non commixta corpori, sicut commiscetur virtus vel forma corporis; et ideo hoc etiam est falsum. Verum est tamen, quod ipsa est finis generationis, sed hic finis non efficitur in materia virtute qualitatum primarum transmutantium materiam, sed potius intelligentia dat eum, quando materia est disposita per principia naturalia.’ Albertus, De anima, lib. 1 tr. 2 c. 13, p. 54: ‘[sc. intellectualis anima] est similitudo quaedam agentis primi. Propter quod dixerunt philosophi quidam mediante intelligentia animam creari’. Cf. Pseudo-Aristotle, Liber de causis, § 3, p. 166: ‘causa prima creavit esse animae mediante intelligentia’. On this passage in Albertus see de Libera, Mtaphysique et notique, pp. 278 – 81.

240 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Dag Nikolaus Hasse

Some years later, in the Metaphysics commentary (dating 1262 – 63), Albertus arrives at a terminological clarification. He now distinguishes an active intellect in the soul and a separate intellect outside.73 About the latter he writes: The entirely separate intellect is the intellect of the intelligence, of which the human intellect is an image (imago), that surrounds its movable (carrier), which is the human body, just as the separate intelligence surrounds its movable (carrier), which is the sphere. The human intellect is in (the separate intellect), just like the inferior lights are in the superior light, from which they receive both the forms and the movement (formas et motum) by way of influence. This influence of form and movement continues until the first cause, which moves the first, universally influential sphere and is thoroughly pure light that does not receive anything from any other (light).74

In this passage, Albertus does not call the origin of the forms a ‘giver of forms’ or an ‘active intellect’ anymore. The Avicennian influence is nevertheless apparent. The separate intellect is not the first cause, but an inferior intelligence which belongs to a heavenly sphere. In Avicennian terms, this would be the active intellect, and indeed, the separate intellect gives ‘forms and movement’ to the human intellect. It is true that ‘forms’ here means ‘intelligible forms’, but the term imago indicates that the passage is a comment also on the origin of the intellectual soul: The human intellect is the imago of the separate intelligence, and it exists in it. The term imago is important, because it is a repeated claim of Albertus that the intellectual soul is the likeness (imago or similitudo) of an intelligence – and not the likeness of God, as one might expect. This claim appears in the commentaries on De divinis nominibus, De animalibus and Metaphysics. One such passage, in the commentary on De divinis nominibus, will be discussed below. The similitudo theory shows that Albertus in these works does not adopt either a naturalistic standpoint or the theological view on the origin of the soul. The intellectual soul’s most intimate link is with an inferior intelligence, even in terms of origin. The strength of Avicenna’s position is apparent in these passages: the intelligence is the source of intelligible forms and of movement in the sublunar world. It is the proximate cause of many effects in the sublunar 73 For further passages in Albertus’ works on the active intellect, within or outside the soul, see Anzulewicz, Entwicklung und Stellung, pp. 198 – 9 and 207 – 8. 74 Albertus, Metaphysica, lib. II tr. 1 c. 9, pp. 472 – 3: ‘Intellectus autem omnino separatus est intellectus intelligentiae, cuius intellectus hominis est imago quaedam ambiens suum mobile, quod est corpus hominis, sicut intelligentia separata ambit suum mobile, quod est sphaera. Et intellectus hominis est in illo, sicut lumina inferiora sunt in lumine superiori, a quo recipiunt et formas et motum per influentiam. Et continuatur huiusmodi influentia formae et motus usque ad causam primam, quae movet sphaeram primam universaliter influentem, quae est penitus lux pura, quae a nulla aliquid recipit.’

Avicenna’s ‘Giver of Forms’ in Latin Philosophy 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

241

world. It is reasonable, therefore, not to attribute the origin of the soul to the remote cause, God, but to the more proximate cause, the intelligence. Avicenna thus enables Albertus to move halfway from a theological towards a naturalistic theory of the origin of the rational soul. The full way would have been Alexander of Aphrodisias’ element theory or Aristotle’s ‘man is begotten by man and by the sun’ (Physics, c. II.2, 194b14).75

IV Albertus: the Origin of the Forms of Inferior Substances We have seen that Avicenna’s theory of the giver of forms falls on fertile ground with Albertus when it comes to the origin of the intellectual soul. In one work, De anima, Albertus even openly adopts the giver of forms theory. When we turn our attention to the other substantial forms – that is, to those of inferior beings such as animals, plants, minerals etc. – the case seems to be more straightforward, because here Albertus clearly favours the theory of inchoatio formae: everything comes to be out of indeterminate beginnings of its essence, which preexist in matter.76 The origin of form lies in matter’s never-ending desire for successive forms.77 This theory owes much to Averroes.78 Nevertheless, Avicenna’s theory appeals to Albertus in three contexts. (1) The first context is the question of whether God could have created things better. In his commentary on the first book of the Sentences (dating ca. 1245), Albertus says that with respect to substantial being the answer is: ‘no’. God could not have given a greater capacitas to the things (that is, greater powers tied to the forms) because the giving of forms is dependent upon the disposition of matter, ut dicit Avicenna. The dator formarum deus fills everything with forms according to the disposition (of matter): forms of elements, forms of elementary compounds, forms of plants, animals and human beings, depending upon the degree in which the material mixture reaches a balance. Albertus 75 Albertus’ theory of the rational soul is indebted to Avicenna also in other ways; see Hasse, The Early Albertus Magnus, pp. 232 – 52. 76 To quote a representative passage: Albertus, Metaphysica, lib. 2 tr. 1 c. 8, p. 470: ‘Quartum est quod nihil fit ex nihilo penitus secundum naturam, sed quaecumque fiunt procedunt ex indeterminatis et confusis incohationibus suarum essentiarum, quae indita sunt materiae’. On Albertus’ doctrine see Nardi, La dottrina, pp. 69 – 101; Snyder, Albert, pp. 63 – 82; Takahashi, Nature, pp. 451 – 81. 77 Albertus, De generatione et corruptione, I, tr. 1 c. 22, p. 130: ‘[sc. materia] non desiderat formam unam tantum, sed omnem formam successive, cum simul eas habere non possit. Hoc autem desiderium formae incohatio est in materia, quae educitur de ipsa’. 78 Cf. Averroes, Commentarium medium in De generatione, c. 17, pp. 26 – 7: ‘tunc necesse est ut generatio non abscindatur quoniam per successionem formarum super subiectum quod est materia non denudatur illud ex quo generatio fit simpliciter …’.

242 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Dag Nikolaus Hasse

concludes: Et per hoc patet solutio ad totum. 79 This is pure Avicennian philosophy – employed, however, for a theory not of the active intellect, but of the Christian creator God. The advantage of Avicenna’s theory is that it offers an explanation of the diversity and variety of sublunar nature. The variety is the result of the combination of forms with corresponding material mixtures. Some years later, in the commentary on Pseudo-Dionysius’ De divinis nominibus (dating 1250), Albertus has changed his attitude. He now rejects the Avicennian (and, in his view, Platonic) principle that all forms are given according to the preparedness of matter, because the position cannot account for the origin of matter. What proceeds from the first cause, is only forms. Hence, one would have to posit an eternal matter, but this is against faith.80 (2) However, the giver of forms theory is still present in De divinis nominibus, albeit in a different context: Albertus extends the above-mentioned similitudo theory to all forms, not only to the intellectual soul. All forms arise from matter when the light of the intelligence meets matter and turns it into actuality. The form is a likeness of the first cause: It has to be said that, according to Avicenna and to those who posit the giver of forms, the form of each thing is nothing else but the ray of the intelligence, or of the first cause … We, in contrast, maintain in a manner more appropriate for theology and philosophy according to the opinion of Aristotle that all forms are educed from the potentiality of matter. Hence, the form, in virtue of its essence, is not the embodied light of the first cause. However, since all actions of those things which are composed of a mover and a moved, have in them the power of the mover and the moved, as is obvious in the action of the natural heat which digests that which is not turned into ashes, but moves towards the form of flesh according to the power of the soul, likewise in the action of the heaven, insofar as it educes forms from the potentiality of matter, as it is said that ‘man begets man and the sun’, there is the power of the first mover, towards whose likeness matter rises through being turned into actuality, as far as it can. Hence, the form is not the embodied light of the first cause, but its likeness caused by it (similitudo eius causata ab ipsa). And it is in this way that one should understand what Dionysius says, that the divine ray appears

79 Albertus, In I Sententiarum, dist. 44 B a. 2, p. 392: ‘Et sic est etiam de datore formarum deo quod omnia implet esse substantiali secundum eorum capacitatem, quaedam forma elementi, quaedam forma mixti, quaedam autem anima vegetabili et quaedam sensibili et quaedam rationali, secundum quod magis recedunt ab actu contrarietatis ad temperamentum et uniformitatem complexionis, quia in hoc accedunt ad naturam coeli, ut ipse [sc. Avicenna] dicit. Et per hoc patet solutio ad totum.’ 80 Albertus, De divinis nominibus, c. 2, p. 73: ‘Et hoc est quod Plato dixit quod formae omnes dantur a datore secundum meritum materiae … Secundum hoc autem … non inveniretur modus quo procedert materia ab ipso; unde oporteret ponere materiam aeternam, quod est contra fidem. Et ideo sequimur opinionem Aristotelis, quae magis videtur catholica’. This passage is discussed in Anzulewicz, Pseudo-Dionysius, pp. 258 – 64.

Avicenna’s ‘Giver of Forms’ in Latin Philosophy 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

243

above all beings, not as embodied, but so that each thing rises towards its (sc. the divine ray’s) likeness, as far as it can.81

Remember that Albertus, in the Metaphysics commentary, has described the human soul as the imago of an intelligence associated with a heavenly sphere. In the present passage, the likeness is not with an inferior intelligence, but with the divine cause. Radius divinus is the term Albertus adopts from Pseudo-Dionysius Areopagita; he likens it to Avicenna’s ray of the intelligence and to the power of the first mover in the heaven. In Albertus’ view, it is not enough to say with Aristotle that the forms of inferior substances arise from the potentiality of matter; they are also caused by an immaterial supralunar cause, whose likeness they are.82 Hence, it is true that Albertus rejects Avicenna’s giver of forms theory and the idea that forms are ‘embodied light’ (lux incorporata), but he follows Avicenna’s basic assumption that the origin of forms has both sublunar and supralunar causes. (3) The third context is causation theory. In his Physics commentary (dating 1251 – 57), Albertus contrasts Avicenna’s giver of forms theory with the position of ‘most Peripatetics’ that the forms are educed from matter. ‘This is the position of the two parties’, Albertus says, ‘and everybody may choose as he likes. We, however, say, as it appears to us, that both opinions are true in some way.’83 Albertus argues as follows: In all things moved by the first cause, there is one essence (essentia), but the being (esse) is manifold. Avicenna maintains that in the 81 Albertus, De divinis nominibus, c. 1, p. 15: ‘Solutio: Dicendum quod secundum Avicennam et ponentes datorem formarum forma uniuscuiusque nihil aliud est quam radius intelligentiae sive causae primae … Nos autem aliter dicimus convenientius theologiae et philosophiae secundum opinionem Aristotelis, quod formae omnes educuntur de potentia materiae. Unde forma per suam essentiam non est lux primae causae incorporata. Sed cum omnis actio eius quod compositum est ex motore et moto, habeat se in virtutem motoris et moti, sicut patet in actione caloris naturalis digerentis quod non incineratur, sed agit ad formam carnis secundum virtutem animae, ita in actione caeli, secundum quam educit formas de potentia materiae, sicut dicitur, quod “homo generat hominem et sol”, est virtus primi motoris, in cuius similitudinem consurgit materia per reductionem in actum, quantum potest. Unde forma non est lux primae causae incorporata, sed similitudo eius causata ab ipsa. Et sic est intelligendum, quod dicit Dionysius, quod superapparet radius divinus omnibus existentibus, non tamquam incorporetur eis, sed ut in cuius similitudinem consurgit unaquaeque res, quantum potest.’ 82 Compare the parallel passage in: Albertus, De ecclesiastica hierarchia, c. 5, p. 123: ‘Quicquid autem sit verum de hoc [sc. Avicenna’s theory], in hoc tamen est simile, quod similiter divinus radius, secundum quod se tenet ex parte infundentis, supereminet et manet in sua simplicitate, diversificatur autem secundum quod recipitur in diversis dissimiliter proportionatis ad ipsum’. 83 Albertus, Physica, lib. II tr. 2 c. 3, p. 103, line 29: ‘Ecce, haec est sententia utrarumque opinionum, et eligat unusquisque quod vult. Nos autem dicimus, prout nobis videtur, quod utraque istarum opinionum vera est secundum aliquem modum.’

244 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Dag Nikolaus Hasse

universe there must be one being which, in virtue of its essence, is the efficient cause of everything.84 It is the essence of the first being which exerts all efficient causality in the world, just like warm objects and lucid objects are such only because the essence of heat warms in them and the essence of light shines in them.85 Hence we want to say that there is one essence through which the first mover and all subsequent things are moving … Therefore, with respect to the essence through which the first mover moves, Avicenna is right, because in this way the first mover alone educes from potentiality to actuality and perfects matter.86

But the other Peripatetics are right too in saying that the essence has manifold existence (diversum habet esse). 87 And, in fact, it is closer to the truth (verius) to say that matter is perfected by natural causes, and not by divine causes.88 It is typical of Albertus that he tries to harmonize Aristotle and the Arabic philosophers. But it remains astonishing that he does so in this question, since the scholastics usually treat the two alternatives – giver of forms and material principles – as antagonistic. Albertus shows his preferences for the proximate cause, which is the material principles, rather than the first cause, and he does not adopt the giver of forms theory. But he is convinced that a theory of generation has to integrate a first essence which is the cause of all efficient causality in the world. In these three contexts, Albertus qualifies his often repeated theory of inchoatio formae: that everything is generated from preexisting formal information in matter. His qualification is due to Avicennian influence and may be summed up as follows: The generation of forms also depends upon a higher immaterial cause. The relation between the forms and the higher cause (a relation called similitudo by Albertus) is of varying degrees due to the different degrees of preparedness of matter. The natural causes, which determine the generation process, would not work without an essence that is the principle of all efficient causality. The essence of things receives greater variety with increasing distance from the first cause, as a result of the diversity of matter.

84 Albertus, ibid., p. 102, lines 43 – 51. 85 Albertus, ibid., p. 102, lines 51 – 6. 86 Albertus, ibid., p. 103, lines 67 – 76: ‘Sic ergo intendimus dicere quod una est essentia qua movet primum movens et omnia consequenter moventia … Ed ideo quantum ad essentiam qua movet primum movens, verum dicit Avicenna, quia sic solum primum movens educit de potentia ad actum et perficit materiam’. 87 Albertus, ibid., p. 103, line 69. 88 Albertus, ibid., p. 103, line 83.

Avicenna’s ‘Giver of Forms’ in Latin Philosophy 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

245

In view of this, we realize that Albertus did not fully take over the eductio formarum theory from Averroes, as is sometimes claimed.89 It is true that Albertus distances himself from Plato and Avicenna and that he criticizes the dator formarum theory.90 But his position nevertheless remains close to Avicenna.91 From Albertus’ standpoint, matter has its own dynamism and is in need of a supralunar immaterial causality in order to be actualized and to reach the similitudo which is the result of the generation process. Note, however, that Albertus discusses the functions of the giver of forms under the label causa divina, as distinct from causa naturalis, and that this is in disaccord with Avicenna’s concept of a giver of forms, which is not a supernatural entity. The scholastics were not able to, or did not want to, integrate Avicenna’s basic idea that the sublunar and supralunar worlds are united in one system of the causality of existence. It remains remarkable, though, that Albertus, in very few passages, identifies the cause of the human soul not with God, but with a separate intelligence surrounding a heavenly sphere. We are now in a better position to understand why the scholastics bothered to discuss the theory they attributed to Plato and Avicenna, which almost all of them rejected. First, of course, because Plato and Avicenna are famous philosophers. But a second reason appears to be that it seemed, after all, a good philosophical idea to make the lowest intelligence, and not God, responsible for the origin of souls, and to assume that the generation of inferior substances was dependent not only on lower but also on higher causes. If the scholastics did not see the advantages themselves, they found them lurking in the much read works of Albertus Magnus, where the attraction of the Avicennian standpoint is clearly felt.

Bibliography Albertus Magnus, De anima, ed. C. Stroick, in Opera omnia … edenda … curavit Institutum Alberti Magni Coloniense, 7,1, Mnster: Aschendorff, 1968. 89 de Libera, Mtaphysique et notique, p. 167. But I agree with the following (de Libera, Albert le Grand et le Platonisme, p. 102): ‘Tout en rejetant le Dator formarum, le Colonais [sc. Albertus Magnus] n’a aucune raison de rejeter l’intgralit d’une doctrine – mÞme imparfaite – de la cration; une doctrine qui, faisant place  un Dieu des dieux et  une influence des toiles et des sphres, laisse ouverte la possibilit d’une thologie des Intelligences qui, on le sait, reste,  ses yeux, la fine pointe du pripattisme. Une fois amend par l’eductio formarum le platonisme est une philosophie  peu prs viable’. 90 Especially in: Albertus, De divinis nominibus, c. 2, p. 73 (n. 81 above), and id., Metaphysica, lib. II tr. 1 c. 8, pp. 468 – 71. 91 As was also pointed out by de Libera (as in n. 89 above) and Takahashi, Nature, p. 476: ‘… it [sc. the giver of forms theory] is evidently not so far removed from his own theory of the generative role of the prime intellect of the universe’.

246

˘

˘

˘

˘

˘ ˘ ˘

˘

˘

˘

˘

–– , De generatione et corruptione, ed. P. Hossfeld, in Opera omnia … edenda … curavit Institutum Alberti Magni Coloniense, 5,2, Mnster: Aschendorff, 1980. –– , De homine, ed. H. Anzulewicz and J.R. Sçder, in Opera omnia … edenda … curavit Institutum Alberti Magni Coloniense, 27,2, Mnster: Aschendorff, 2008. –– , Commentarii in primum Sententiarum (dist. XXVI–XLVIII), ed. A. Borgnet, in Opera omnia, vol. 26, Paris: Vivs, 1893. –– , Metaphysica, ed. B. Geyer, in Opera omnia … edenda … curavit Institutum Alberti Magni Coloniense, 16, Mnster: Aschendorff, 1964. –– , Physica, ed. P. Hossfeld, in Opera omnia … edenda … curavit Institutum Alberti Magni Coloniense, 4, 2 vols, Mnster: Aschendorff, 1987/1993. –– , Super Dionysium De divinis nominibus, ed. P. Simon, in Opera omnia … edenda … curavit Institutum Alberti Magni Coloniense, 37,1, Mnster: Aschendorff, 1972. –– , Super Dionysium De ecclesiastica hierarchia, ed. M. Burger, in Opera omnia … edenda … curavit Institutum Alberti Magni Coloniense, 36, 2, Mnster: Aschendorff, 1999. Anonymous (Van Steenberghen), Quaestiones de anima, ed. F. Van Steenberghen, in B. Bazn, M. Giele and F. Van Steenberghen, eds, Trois commentaires anonymes sur le trait de l’me d’Aristote, Louvain: Publications universitaires, 1971, pp. 123 – 348. Anonymous (de Vaux), De causis primis et secundis et de fluxu qui consequitur eas, ed. R. de Vaux, (Notes et textes sur l’avicennisme latin), Paris: Vrin, 1934, pp. 88 – 140. H. Anzulewicz, Pseudo-Dionysius Areopagita und das Strukturprinzip des Denkens von Albert dem Grossen, in T. Boiadjiev et al., eds, Die Dionysius-Rezeption im Mittelalter, Turnhout: Brepols, 2000, pp. 251 – 95. –– , Entwicklung und Stellung der Intellekttheorie im System des Albertus Magnus, Archives d’histoire doctrinale et littraire du moyen ge, 70, 2003, pp. 165 – 218. Pseudo-Aristotle, Liber de causis, ed. O. Bardenhewer, Freiburg im Breisgau: Herder, 1882, reprinted Frankfurt a.M.: Minerva, 1961. Averroes, Tafsı¯r ma- ba da at-tabı¯ a, ed. M. Bouyges, 3 vols, Beirut: Imprimerie Catholique, 1938 – 48. ˙ ˙ –– , Commentarium medium in Aristotelis De generatione et corruptione libros, eds F.H. Fobes and S. Kurland, Cambridge, Massachusetts: The Mediaeval Academy of America, 1956. –– , Commentarium in libros Metaphysicorum, in Aristotle/Averroes, Aristotelis Stagirite omnia quae extant opera … Averrois … commentarii aliique ipsius in logica, philosophia et medicina libri, Venice: Giunta, 1562, reprinted Frankfurt a.M.: Minerva, 1962, vol. 8. –– , Tahafut al-tahafut (The Incoherence of the Incoherence), transl. and ed. S. Van den Bergh, 2 vols, London: Luzac, 1954. Avicenna, The Metaphysics of ‘The Healing’, transl. and ed. M.E. Marmura, Provo, Utah: Brigham Young University Press, 2005. –– , asˇ-Sˇifa- , at-Tabı¯ iyya-t, al-Kawn wa-l-fasa-d (On Generation and Corruption), ed. M. Qassem,˙ ˙Cairo: al-Hay a al-Misriyya al- A¯mma, 1969. –– , asˇ-Sˇifa- , at-Tabı¯ iyya-t, Fı¯ l-af a-l ˙ wa-l-infi a-la-t (On Actions and Passions), ed. ˙ M. Qassem,˙Cairo: al-Hay a al-Misriyya al- A¯mma, 1969. ˙ –– , Liber tertius naturalium de generatione et corruptione, ed. S. van Riet, Louvain-laNeuve/Leiden: Brill, 1987. A. Bertolacci, ‘Subtilius speculando’. Le citazioni della Philosophia Prima di Avicenna nel Commento alla Metafisica di Alberto Magno, Documenti e Studi sulla Tradizione Filosofica Medievale, 9, 1998, pp. 261 – 339. ˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Dag Nikolaus Hasse

247

Avicenna’s ‘Giver of Forms’ in Latin Philosophy

–– , Le citazioni implicite testuali della Philosophia prima di Avicenna nel commento alla Metafisica di Alberto Magno: analisi tipologica, Documenti e Studi sulla Tradizione Filosofica Medievale, 12, 2001, pp. 179 – 274. –– , The Reception of Averroes’ Long Commentary on the Metaphysics in Latin Medieval Philosophy until Albertus Magnus, in L. Honnefelder et al., eds, Via Alberti. Texte – Quellen – Interpretationen, Mnster: Aschendorff, 2009, pp. 457 – 80. Bonaventure, Commentaria in quatuor libros Sententiarum, in Doctoris Seraphici S. Bonaventurae … opera omnia, 4 vols, Quaracchi: Typogr. Coll. S. Bonaventurae, 1882 – 89. Campanella, Tommaso, De homine (Theologicorum liber IV), ed. R. Amerio, 2 vols, Rome: Centro Internaz. di Studi Umanistici, 1960 – 61. R.C. Dales, The Problem of the Rational Soul in the Thirteenth Century, Leiden/New York/Cologne: Brill, 1995. Dominicus Gundisalvi, De processione mundi, ed. G. Blow (Beitrge zur Geschichte der Philosophie und Theologie des Mittelalters, Band 24, Heft 3), Mnster: Aschendorff, 1925. Ficino, Marsilio, Theologia platonica (Platonic Theology), transls and eds M. J. B. Allen, J. Hankins, 6 vols, Cambridge, Massachusetts: Harvard University Press, 2001 – 06. ˙ aza-lı¯, Maqa-sid al-fala-sifa, ed. S. Dunya-, Cairo: Da-r al-Ma a-rif, 1960. al-G Ch. Genequand,˙ Ibn Rushd’s Metaphysics: A Translation with Introduction of Ibn Rushd’s Commentary on Aristotle’s Metaphysics, Book La¯m, Leiden: Brill, 1986. H. Goldstein, Dator formarum: Ibn Ruschd, Levi ben Gerson, and Moses ben Joshua of Narbonne, in I. al-Faruqi, ed., Islamic Thought and Culture, Washington DC: International Institute of Islamic Thought, 1982, pp. 107 – 21. D.N. Hasse, Avicenna’s ‘De anima’ in the Latin West: The Formation of a Peripatetic Philosophy of the Soul, 1160 – 1300, London/Turin: The Warburg Institute/Nino Aragno Editore, 2000. –– , Plato arabico-latinus: Philosophy-Wisdom Literature-Occult Sciences, in S. Gersh and M.J.F.M. Hoenen, eds, The Platonic Tradition in the Middle Ages: A Doxographic Approach, Berlin/New York: deGruyter, 2002, pp. 31 – 65. –– , Spontaneous Generation and the Ontology of Forms in Greek, Arabic, and Medieval Latin Sources, in P. Adamson, ed., Classical Arabic Philosophy: Sources and Reception, London/Turin: The Warburg Institute/Nino Aragno Editore, 2007, pp. 150 – 75. –– , Arabic Philosophy and Averroism, in J. Hankins, ed., Cambridge Companion to Renaissance Philosophy, Cambridge: Cambridge University Press, 2007, pp. 113 – 36. –– , The Early Albertus Magnus and his Arabic Sources on the Theory of the Soul, Vivarium, 46, 2008, pp. 232 – 52. H. Hirai, Atomes vivants, origine de l’ me et gnration spontane chez Daniel Sennert, Bruniana & Campelliana, 13, 2007, pp. 477 – 95. J. Janssens, Le Da¯nesh-Na¯meh d’Ibn Sı¯na¯ : un texte  revoir?’, Bulletin de philosophie mdivale, 28, 1986, pp. 163 – 77, reprinted in id., Ibn Sı¯na¯ and his Influence on the Arabic and Latin World, Ashgate: Ashgate Variorum, 2006, art. VII. –– , The Notions of Wa-hib Al-Suwar (Giver of Forms) and Wa-hib Al- Aql (Bestower of intelligence) in Ibn Sı¯na-,˙ in M.C. Pacheco, J.F. Meirinhos, eds, Intellect et imagination dans la Philosophie Mdivale, 3 vols, Turnhout: Brepols, 2006, vol. 1, pp. 551 – 62. H.S. Jantz, Goethe’s Faust as a Renaissance Man: Parallels and Prototypes, Princeton: Princeton University Press, 1951. ˘

˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

248 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Dag Nikolaus Hasse

John Buridan, In Metaphysicen Aristotelis questiones argutissimae, Paris: Badius Ascensius, 1518, reprinted Frankfurt a.M.: Minerva, 1964 (with wrong date 1588 on the titlepage). John of Jandun, In duodecim libros Metaphysicae, Venice: Scotus, 1553, reprinted Frankfurt a.M.: Minerva, 1966. A. de Libera, Albert le Grand et le platonisme. De la doctrine des ides  la thorie des trois tats de l’universel, in E.P. Bos, ed., On Proclus and His Influence in Medieval Philosophy, Leiden et al.: Brill, 1992, pp. 89 – 119. –– , Albert le Grand ou l’antiplatonisme sans Platon, in M. Dixsaut, ed., Contre Platon, 2 vols, Paris: Vrin, 1993, vol. 1, pp. 247 – 71. –– , Mtaphysique et notique: Albert le Grand, Paris: Vrin, 2005 M.E. Marmura, The Metaphysics of Efficient Causality in Avicenna (Ibn Sina), in M.E. Marmura, ed., Islamic Theology and Philosophy: Studies in Honor of George F. Hourani, Albany, New York: State University of New York Press, 1984, pp. 172 – 87. C.A. Nallino, La ‘Colcodea’ d’Avicenna e T. Campanella, Giornale critico della filosofia italiana, 6, 1925, pp. 84 – 91. B. Nardi, La dottrina d’Alberto Magno sull’ ‘inchoatio formae’, in id., Studi di filosofia medievale, Rome: Edizioni di Storia e letteratura, 1960, pp. 69 – 101. –– , Origine dell’ anima humana, in id., Studi su Pietro Pomponazzi, Florence: Le Monnier, 1965, pp. 231 – 46. Nifo, Agostino, In librum Destructio destructionum Averroys commentationes, Venice: Bonatus Locatellus, 1495. –– , In librum De anima Aristotelis et Averoys commentatio, Venice: Petrus de Quarengiis Bergomensis, 1503. –– , Expositiones in libros Metaphysices, Venice: Scotus, 1559, reprinted Frankfurt a.M.: Minerva, 1967. Pietro d’Abano, Conciliator controversiarum quae inter philosophos et medicos versantur, Venice: Giunta, 1565, reprinted Padua: Antenore, 1985. P. Porro, Colcodea, in V. Melchiorre, ed., Enciclopedia filosofica, Milan: Bompiani, 2006, vol. 3, pp. 2009 – 2010. Russiliano, Tiberio, Apologeticus adversus cucullatos. Una reincarnazione di Pico ai tempi di Pomponazzi, ed. P. Zambelli, Milan: Il Polifilo, 1994. Scaliger, Julius Caesar, Exotericarum exercitationum liber XV de subtilitate ad Hieronymum Cardanum, Frankfurt: Wechel, 1582. Sennert, Daniel, Hypomnemata physica, Frankfurt: Schleich, 1636. S.C. Snyder, Albert the Great, incohatio formae, and the Pure Potentiality of Matter, American Catholic Philosophical Quarterly, 70, 1996, pp. 63 – 82. Suarez, Francisco, Disputationes metaphysicae, in Opera omnia, ed. C. Berton, vols XXV–XXVI, Paris: Vivs, 1866, reprinted Hildesheim: Olms, 2009. A. Takahashi, Nature, Formative Power and Intellect in the Natural Philosophy of Albert the Great, Early Science and Medicine, 13, 2008, pp. 451 – 81. R.J. Teske, William of Auvergne on the Individuation of Human Souls, Traditio, 49, 1994, pp. 77 – 93, reprinted in id., Studies in the Philosophy of William of Auvergne, Bishop of Paris (1228 – 1249), Milwaukee, Wisconsin: Marquette University Press, 2006, pp. 121 – 43. –– , Studies in the Philosophy of William of Auvergne, Bishop of Paris (1228 – 1249), Milwaukee, Wisconsin: Marquette University Press, 2006. Thomas Aquinas, Summa contra gentiles, in Opera omnia iussu Leonis XIII P.M. edita, vol. XIII–XV, Rome: Editori di San Tommaso, 1918 – 30.

Avicenna’s ‘Giver of Forms’ in Latin Philosophy 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

249

–– , Quaestiones disputatae de veritate, in Opera omnia iussu Leonis XIII P.M. edita, vol. XXII, Rome: Editori di San Tommaso, 1970 – 76. –– , De potentia, in Quaestiones disputatae, eds R. Spiazzi and P.M. Bazzi, vol. 2, Turin/ Rome: Marietti, 1953. L. Thorndike, A History of Magic and Experimental Science, 8 vols, New York: Columbia University Press, 1923 – 58. Tignosi, Niccol , In libros Aristotelis de anima commentarii, Florence: Ex Bibliotheca Medicea, 1551. J.A. Weisheipl, The Axiom ‘Opus naturae est opus intelligentiae’ and its Origins, in G. Meyer OP and A. Zimmermann, eds, Albertus Magnus – Doctor Universalis 1280/1980, Mainz: Matthias-Grnewald-Verlag, 1980, pp. 441 – 63. –– , Aristotle’s Concept of Nature: Avicenna and Aquinas, in L.D. Roberts, ed., Approaches to Nature in the Middle Ages, Binghamton, New York: Center for Medieval and Early Renaissance Studies, 1982, pp. 137 – 60. William of Auvergne, Opera omnia, 2 vols, ed. F. Hotot, with Supplementum, ed. B. Le Feron, Orlans/Paris: Billaine, 1674, reprinted Frankfurt a.M.: Minerva, 1963 (vol. 1: De universo, vol. 2: De anima et al.). –– , De Trinitate, ed. B. Switalski, Toronto: Pontifical Institute of Mediaeval Studies, 1976. H.A. Wolfson, Colcodea, 4=79K@9K, in id., Studies in the History of Philosophy and Religion, 2 vols, Cambridge, Massachusetts: Harvard University Press, 1973 – 77, vol. 2, pp. 573 – 6 (originally: The Jewish Quarterly Review, n.s. 36, 1945, pp. 179 – 82). Zimara, Marcantonio, Contradictionum solutiones in hos Metaphysicorum libros, in Aristotle and Averroes, Aristotelis opera cum Averrois commentariis, Venice: Giunta, 1562, reprinted Frankfurt a. M.: Minerva, 1962, vol. 8, fols 401 – 24.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Avicenna and Aquinas on Form and Generation Kara Richardson Avicenna’s Agent Intellect – the last of a series of productive intellects in his emanationist creation story – is implicated in several controversies in the Medieval Latin West.1 One of these debates has to do with substantial generation, i. e., the coming to be of new corporeal substances, such as plants and animals. According to Aquinas, Avicenna and the Platonists hold that the substantial form needed for the generation of a corporeal substance is produced ex nihilo by a creative agent.2 For Avicenna, this agent is the ‘Giver of Forms’ (wa¯hib al-suwar/dator formarum), i. e., the Agent Intellect.3 (I will refer to this ˙ Infusion Model of substantial generation. On this model, when a view as the new compound of form and matter is made, an incorporeal substance produces its form and bestows it upon matter made ready to receive it by corporeal causes.) Aquinas rejects the Infusion Model in favour of a view he attributes to Aristotle: substantial forms are educed from the potentiality of matter by corporeal causes.4 (I will refer to this view as the Eduction Model of substantial generation. On this model, when a new compound is made, its form is produced per accidens in the process of eduction; the compound is produced per se and is the terminus of the process of eduction.) Thus Aquinas claims for himself the territory of Aristotelian naturalism and casts Avicenna’s Agent Intellect in the role of deus ex machina.

1

2 3

4

This paper is developed in part from my doctoral dissertation. I would like to thank my dissertation advisors, Deborah Black and Marleen Rozemond, as well as Martin Pickav, all of the University of Toronto, for very helpful comments on that work. I would also like to thank the participants in the Villa Vigoni conference on the reception of Avicenna’s Metaphysics for their engagement with an earlier version of this paper. Finally, I am especially grateful to Dag Nikolaus Hasse and Amos Bertolacci for their comments and suggestions on the paper in its final form. Hasse, Plato arabico-latinus, discusses Aquinas’ association of Plato and Avicenna in De potentia q. 3, a. 8 and elsewhere. See especially pp. 42 – 5. Hasse, Avicenna’s ‘De Anima’, p. 188 points out that Avicenna does not explicitly identify the ‘Giver of Forms’ (wa¯hib al-suwar) and the ‘bestower of intelligibles’ (wa¯hib al-‘aql), ˙ The Notions, p. 558 also raises this issue, but he i. e., the Agent Intellect. Janssens, concedes that the roles played by the Giver of Forms indicate that this principle is the Agent Intellect. Aquinas attributes this view to Aristotle in De potentia q. 5 a. 1 ad 5: ‘Si autem ponamus formas substantiales educi de potentia materiae, secundum sententiam Aristotelis, …’.

252 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Kara Richardson

This paper raises several objections to Aquinas’ reckoning of the matter, which has been very influential.5 First, it distorts Avicenna’s view of the unity of a hylemorphic composite substance. Second, it invites an occasionalist reading of Avicenna, which is foreign to his thinking. Third, it conceals Aquinas’ agreement with Avicenna that an incorporeal agent causes the existence of the composite through its form, and his subsequent use of Avicennian ideas to reconcile this view with the Eduction Model of generation. Aquinas’ account of Avicenna’s view of substantial generation suggests a very negative reception of Avicenna’s metaphysics. This paper argues that Aquinas’ criticisms of Avicenna are based on a weak reading of Avicenna’s discussions of the causal roles of the Agent Intellect, and shows that Aquinas’ reception of several aspects of these discussions was quite positive.

I Aquinas’ Argument against the Creation of Form in Generation Since Aquinas’ most detailed argument against the view of generation he attributes to Avicenna appears in De potentia q. 3 a. 8, I will focus primarily on that text.6 In De potentia q. 3 a. 8, Aquinas treats the question whether God works in nature by creating. ‘Creation’ is a technical term for Aquinas. It refers not to any act of making, such as the builder’s act of making a house out of bricks, but rather to acts of making ex nihilo. Thus the question of De potentia q. 3 a. 8 is not the broad one whether God works in nature at all – which Aquinas answers in the affirmative in De potentia q. 3 a. 7, Summa theologiae 1a q. 105 and Summa contra gentiles lib. 3 c. 67 – but the narrower one whether natural processes, such as the generation of plants and animals, involve the making of something from nothing by God. Aquinas himself denies that God works in nature by creating. Of the great many arguments Aquinas recounts against his own view, two are especially instructive for our purposes. The first relies on the following premises: (1) ‘whatever has no material part cannot be made of matter’ and (2)

5

6

See, for example, Suarez, Disputationes Metaphysicae 18.1 – 2, Leibniz, De ipsa natura, p. 505, Zedler, Saint Thomas, Lee, St. Thomas, Weisheipl, Aristotle’s Concept of Nature, and Brand, Book of Causes. Aquinas’ interpretation of Avicenna has roots in Averroes’ Long Commentary on the Metaphysics, Book Zeta, comm. 31. I discuss Averroes’ view of Avicenna below on p. 268. Also relevant are Albert the Great’s arguments against an Infusion Model of generation he attributes to ‘Platonist’ philosophers. See Snyder, Albert the Great, especially pp. 65 – 71. Aquinas also presents this view of Avicenna in Summa contra gentiles lib. 3, c. 69, and in Scriptum super Sententiis lib. 2 d. 1 q. 1 a. 4 ad 4.

Avicenna and Aquinas on Form and Generation 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

253

‘a form has no material part’.7 (The second premise reflects the Aristotelian view that form is distinguished both from matter and from the composite.) These premises seem to yield the conclusions that substantial forms cannot be made of matter and that if they are made, they must be made ex nihilo. Thus the argument suggests that some natural processes involve creative activity: namely, cases of generation, in which matter acquires a substantial form. It also suggests that the Infusion Model of generation, which Aquinas attributes to Avicenna and the Platonists, is correct.8 A second argument threatens the Eduction Model of generation, which Aquinas champions. This argument relies on the principle that ‘nothing is educed from that in which it is not’.9 If this is true, then new forms cannot be educed from matter unless they pre-exist in it. This argument supports a position referred to as Anaxagorean, which holds apparently new forms to be previously latent in matter and construes generation as their uncovering.10 Aquinas does not reject the principle behind the Anaxagorean threat to the Eduction Model – sc., that ‘nothing is educed from that in which it is not’ – but he claims that it suffices (in cases of generation, at least) that what is educed exist potentially in that from which it is educed. Thus Aquinas counters Anaxagoras on the ground that he fails to distinguish potentiality and actuality: For he considered it necessary that what is generated pre-exist in act. But it is necessary that it pre-exist in potency and not in act. For if it did not pre-exist in potency, it would come to be out of nothing. And if it did pre-exist in act it would not come to be, since what is does not come to be.11

The force of this argument, as I understand it, depends on a prior commitment to the reality of substantial change. If truly new material substances come to be, then either the principle ‘nothing is educed from that in which it is not’ is satisfied by the potential existence of form in matter prior to generation or new forms are produced ex nihilo. 7 Aquinas, De potentia, q. 3 a. 8 arg. 6: ‘Praeterea, id quod non habet materiam partem sui, non potest ex materia fieri. Sed formae non habent materiam partem sui: quia forma distinguitur et contra materiam et contra compositum, ut patet in principio secundi de Anima. Cum ergo formae fiant quia de novo esse incipiunt, videtur quod non fiant ex materia; et sic fiunt ex nihilo, et per consequens creantur.’ English translations from Aquinas’ De potentia are my own, but I have benefited from consulting On the Power of God, transl. English Dominican Fathers, 1952. 8 Aquinas does not identify the source of the objection at De potentia, q. 3 a. 8 arg. 6. 9 Aquinas, De potentia, q. 3 a. 8 arg. 8: ‘nihil educitur de aliquo quod non est in eo’. 10 Aquinas, De potentia, q. 3 a. 8 arg. 8 – 9. Aristotle attributes this view to Anaxagoras at Physics, 187a26–b7. 11 Aquinas, De potentia, q. 3 a. 8: ‘putabat enim oportere quod actu praeextiterit [sic] illud quod generatur. Oportet autem quod praeexistat potentia et non actu; si enim non praeexisteret potentia, fieret ex nihilo; si vero praeexisteret actu, non fieret: quia quod est, non fit.’

254 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Kara Richardson

Aquinas then describes the Infusion Model of generation he attributes to Plato and Avicenna: But since a generated thing is in potency through matter and in act through its form, others held that a thing becomes with respect to form, while its matter preexists. And since the work of nature cannot be out of nothing, and thus must be out of something presupposed, nature, according to them, works only on the side of matter by disposing it for form. But form, which must become and is not presupposed, must be from an agent who does not presuppose anything but can make something out of nothing. And this is a supernatural agent, which Plato posited to be a giver of Forms. And Avicenna called this the lowest intelligence among separated substances. But certain moderns following them consider this to be God.12

This view employs a distinction between potential and actual being: a hylomorphic composite substance exists potentially prior to generation in the sense that its matter exists prior to generation. On this last point, it diverges from the Eduction Model, which holds that the form of the composite exists potentially in matter prior to generation. Thus on the Infusion Model, the role of nature is not to educe form from matter but rather to dispose matter for form. Form itself is produced ex nihilo by a creative agent, which for Avicenna is the Agent Intellect, but for some of Aquinas’ contemporaries is God.13 Aquinas contends that the proponents of the Infusion Model misunderstand form: if a creative principle produces ex nihilo the form of the composite, then this form must exist per se, i. e., through itself or in its own right. (If the form is made in its own right, it must be something in its own right.) Aquinas claims that the form of the composite does not exist per se, but rather it exists because something else – the composite – exists by it. Since the form of the composite 12 Aquinas, De potentia, q. 3 a. 8: ‘Sed quia res generata est in potentia per materiam, et in actu per suam formam; posuerunt aliqui, quod res fiebat quantum ad formam materia praeexistente. Et quia operatio naturae non potest esse ex nihilo, et per consequens oportet quod sit ex praesuppositione, non operabatur, secundum eos, natura, nisi ex parte materiae disponendo ipsam ad formam. Formam vero, quam oportet fieri et non praesupponi, oportet esse ex agente qui non praesupponit aliquid, sed potest ex nihilo facere: et hoc est agens supernaturale, quod Plato posuit datorem formarum. Et hoc Avicenna dixit esse intelligentiam ultimam inter substantias separatas. Quidam vero moderni eos sequentes, dicunt hoc esse Deum.’ While Aquinas says here that Plato posited a dator formarum, he claims in De potentia, q. 5 a. 1 arg. 5 that Plato held that forms in matter are from forms without matter and that Avicenna posited a dator formarum. 13 Aquinas attributes this view to ‘moderni’ in De potentia q. 3 a. 8. Hasse, Avicenna’s ‘De Anima’, pp. 203 – 23, argues that the ‘moderni’ Aquinas refers to are Jean de la Rochelle and Alexander of Hales (or rather: the anonymous author of the Summa fratris Alexandri, book 2), and also identifies William of Auvergne as one of those who take up the view that the agent who produces the form in generation is God. See also Hasse, Plato arabico-latinus, p. 43, and Hasse, Spontaneous Generation, p. 163.

Avicenna and Aquinas on Form and Generation 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

255

does not exist per se, it cannot be produced ex nihilo by a creative agent. In support of his view, Aquinas invokes the following principle: ‘that which is made is said to become according to the way in which it is: because its being is the term of its making; whence that which properly becomes per se is the composite’.14 Since the way things are made must accord with the way they are, the form of the composite ‘is not properly made per se but is that whereby a thing is made, i. e., that through whose acquisition something is said to be made’.15 Aquinas’ argument against the Infusion Model of generation relies on the following account of the unity of a hylemorphic substance. The formal and material constituents of the composite are one in the way that a piece of wax and its shape are one.16 Neither the shape of the wax, nor the form of the composite exist per se: rather the shape together with the wax and form together with matter comprise a single per se existent. It follows from this view that we must deny that the shape of the wax can be made per se, and that we must deny that the form of the composite substance can be made per se. 17 Thus, the form 14 Aquinas, De potentia, q. 3 a. 8: ‘Unumquodque autem factum, hoc modo dicitur fieri quo dicitur esse. Nam esse est terminus factionis: unde illud quod proprie fit per se, compositum est’. 15 Aquinas, De potentia, q. 3 a. 8: ‘Forma autem non proprie fit, sed est id quod fit, id est per cuius acquisitionem aliquid dicitur fieri.’ Here I read ‘quo’ instead of the ‘quod’ of the Marietti edition. 16 I use Aristotle’s example from De anima. Having determined that the soul is ‘a substance in the sense of the form of a natural body having life potentially within it’ and that the soul qua substance is the actuality of a body, Aristotle says, ‘That is why we can dismiss as unnecessary the question whether the soul and the body are one: it is as though we were to ask whether the wax and its shape are one, or generally the matter of a thing and that of which it is the matter. Unity has many senses (as many as “is” has), but the proper one is that of actuality’ (Aristotle, De anima 2.1 412a20 – 21; 412b5 – 9). 17 Averroes, in his Long Commentary on the Metaphysics, Book Zeta, comm. 31, makes a similar argument on behalf of Aristotle: ‘The forms are not generated in themselves, because if they were, then the generation would be without the matter of the enmattered thing. Consequently, what is generated is something informed, but if that is so then what generates it is that which moves the matter until it receives the form, that is, that which causes [the form] to emerge from potency to act. Now what moves matter must be either a body possessing an active quality or a power of a substance that acts through a body possessing an active quality. If what generates the subject of the form were other than what generates its form, then the subject and its form would be actually two things, which is impossible. Thus, the subject does not exist without the form, unless it is said by homonymy. So because the subject of the form has existence only through the form, the agent’s activity is associated with [the subject] only due to [the subject’s] association with the form. Since the agent’s activity is neither associated with the form alone, nor with the subject without the form, consequently then, the agent’s activity is clearly associated with the subject only on the part of its association with the form. So what generates the form’s subject is what generates the form; in fact, there would be no subject if it were not for

256 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Kara Richardson

of the composite cannot be produced ex nihilo by a creative agent; it is generated only per accidens through the process of eduction in which the composite is generated per se. Generation does not involve creative acts by God. Aquinas’ claim that proponents of the Infusion Model of generation misunderstand form is somewhat ambiguous. He may mean that they hold the mistaken view that form exists per se. Or he may mean that they hold the mistaken view that the generation of form per se is compatible with an account of hylemorphic unity on which form and matter comprise a single per se existent. In Section II, I will consider Avicenna’s own views in this light.

II Avicenna’s Hylemorphism Avicenna discusses form, matter and their relationship in a composite substance in Book 2, chapter one of the Metaphysics of the Sˇifa¯’, which focuses on substance. He defines substance as what is not in a subject.18 This definition encompasses the form of the composite and its matter, the composite itself, as well as separate souls and intellects.19 We might assume that since a substance is not in a subject, every substance exists per se. (I.e., we might assume that what is

[the agent’s] generating the form and generating both of them simultaneously. If the subject of the form were to be generated from one agent and the form from another agent, then a single effect insofar as it is one would be generated from two agents, which is impossible; for it will not be associated with a single act, unless it is an act of a single agent. So one should rely on this [demonstration] in this situation, namely, that on which Aristotle relied’ (Averroes, Commentary on Metaphysics, Zeta 9, p. 334). This argument is discussed in Davidson, Alfarabi, pp. 245 – 50. Earlier in his career, Averroes himself held the view that when a new compound is generated, its form is produced by an incorporeal agent. The merits of Averroes’ later explanation of generation, which does not appeal to an incorporeal producer of form, are discussed in Freudenthal, The Medieval Astrologization. 18 He says that every existent belongs to one of two divisions: ‘One of them is the existent in some other thing, which is a thing having subsistence and species in itself. Its existence isn’t that of a part of this other thing, but it cannot be separated from that other thing. And this is the existent in a subject. And the second is the existent which is not in some other thing in this way. So it isn’t in a subject at all. And this is substance’ (Avicenna, Metaphysics 2.1, p. 45 (my transl.); Sˇ. Il., p. 57; Pr. ph., p. 65). English translations from the Metaphysics of the Sˇifa¯’ are Marmura’s unless otherwise noted. 19 He says, ‘Every substance is either a body or not a body. And if it is not a body, then it is either a part of a body or not a part of a body. And if it is not part of a body, then it is separate from bodies entirely. And if it is a part of a body, then it is either form or matter. And if it is separate, rather than a part of a body, then either it has an administrative relation which is in bodies by means of motion, and is called soul, or it is free from matter in every respect and is called intellect’ (Avicenna, Metaphysics 2.1, p. 48 (my transl.); Sˇ. Il., p. 60; Pr. ph., pp. 68 – 9).

Avicenna and Aquinas on Form and Generation 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

257

not in a subject exists through itself or in its own right.) But Avicenna holds that some substances – namely, form and matter – do not exist per se. Avicenna’s view that form and matter, although substantial, do not exist per se is illustrated in his account of the way form exists in matter. This account relies on the following distinction between a subject (mawdu¯‘) and a receptacle ˙ (mahall): ˙ By ‘subject’ is meant that which becomes subsistent in itself and, in terms of being the species, becomes thereafter a cause for something to subsist in it ([but] not as a part of it); and that the receptacle is anything in which something dwells [and which] becomes, by virtue of that [indwelling] thing, [the possessor] of a certain state.20

A subject is a subsistent thing of some species, and is the bearer of accidents. A receptacle is merely a repository: something can exist in a receptacle even though that receptacle is not a subsistent thing of some species but rather becomes one through what exists in it.21 This is the way form exists in matter: it exists in a receptacle, which is not a subject; through its reception of form, this receptacle becomes a subsistent thing of some species, i. e., a subject.22 Here the view that matter does not exist per se is quite evident: if matter is not a subsistent thing of some species, it cannot exist per se, i. e., through itself or in its own right. The view that form does not exist per se is less clear, but it is suggested by the claim that form exists in matter as in a receptacle, as well as by two later claims. First, form and matter are conceptually distinct but coexistent; neither can exist apart from the other (Metaphysics 2.4). Second, form is a part (g˘uz’) of a body rather than something separate and not part of a body, such as a separate soul or an intellect (Metaphysics 2.1).23 Thus on Avicenna’s view, the form of the composite is a substance (as is its matter), but the composite, not its form or its matter, is a subsistent thing of some species, i. e., a subject (mawdu¯‘). Indeterminate matter becomes a subsistent thing of some species through˙ its reception of form. With respect to this claim, it is worth noting that Avicenna describes prime matter as that which is ‘in potency, all [material] things’.24 This account of form, matter and their relationship suggests that Avicenna denies that form exists per se, and it suggests that he shares with Aquinas the following view of the structure of substantial change: matter first has a form 20 Avicenna, Metaphysics 2.1, pp. 46 – 7; Sˇ. Il., p. 59; Pr. ph., p. 67. 21 Avicenna, Metaphysics 2.1, p. 47; Sˇ. Il., p. 59; Pr. ph., p. 67. 22 He says, ‘[a]s for establishing this thing which exists in a receptacle but not in a subject, this is something incumbent on us to show shortly. Once we establish it, [it will be seen] that it is the thing to which, in this place, the name ‘form’ is properly attributed’ (Avicenna, Metaphysics 2.1, p. 47; Sˇ. Il., p. 59; Pr. ph., p. 68). Stone, Simplicius and Avicenna, pp. 77 – 8 discusses Avicenna’s denial that prime matter is a subject. 23 Avicenna, Metaphysics 2.1, p. 48; Sˇ. Il., p. 60; Pr. ph., pp. 68 – 9. 24 Avicenna, Metaphysics 4.2, p. 134; Sˇ. Il., p. 175; Pr. ph., p. 200.

258 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Kara Richardson

potentially and later, through the activity of some agent or agents, it has it actually.25 This account of form, matter and their relationship conflicts with the Infusion Model of generation.26 If Avicenna endorses the Infusion Model, then his hylemorphism conflicts with his view of generation. In Section III, I consider whether Avicenna’s discussions of causal roles played by the incorporeal ‘Giver of Forms’ support the view that he endorses an Infusion Model of generation.27

25 This account of the structure of substantial change is a cornerstone of the Eduction Model of generation: a form is educed from matter in which it pre-exists potentially. A second feature of Aquinas’ Eduction Model is that eduction is the work of a corporeal agent. Aquinas suggests in De potentia q. 5 a. 1 that incorporeal agents cannot move matter and thus that corporeal agents are needed to dispose and transmute matter; he also equates transmuting matter and educing form from matter. Averroes remarks explicitly that eduction must be the work of a corporeal agent: the generator is ‘that which moves the matter until it receives the form, that is, that which causes [the form] to emerge from potency to act. Now what moves matter must be either a body possessing an active quality or a power of a substance that acts through a body possessing an active quality’ (Averroes, Commentary on Metaphysics, Zeta 9, p. 334). 26 Avicenna’s doctrine of ‘corporeal form’ (su¯ra g˘ismiyya) might seem to complicate his views on the unity of the form/matter ˙compound. This doctrine aims to explain a defining feature of body qua body, namely, its receptivity to the three dimensions of length, breadth and depth. Avicenna explains this signal characteristic by positing in body the form of corporeity. Prime matter and this form together comprise body considered in this very general way. But no individual body is a composite of prime matter and the form of corporeity alone. Individual bodies are members of some species. This might seem to suggest that Avicenna holds that a body has two distinct forms, namely, the form of corporeity, which it shares with all other bodies, and an additional form, which makes it a certain kind of body. I will call this second form the ‘species form’. Hyman, Aristotle’s ‘First Matter’, p. 404 seems to advocate this view, but he notes that it is not clear how Avicenna understood this doctrine of the multiplicity of forms; see p. 404, n. 85. This interpretation of Avicenna’s doctrine of corporeal form is incompatible with the account of hylemorphic unity I attribute to him based on his discussion of form and matter in Metaphysics 2.1. I reject the view that Avicenna holds that a body has two distinct forms. In my view, Avicenna holds the form of corporeity to be merely conceptually distinct from the species form. So every species form confers on the composite the power to take on the three dimensions of length, breadth and depth, as well as its specific features. (Zedler, Saint Thomas, pp. 127 – 8 and Stone, Simplicius and Avicenna, p. 100 also take this position. Stone argues: ‘If a substantial form were in another substantial form as its subject, then it would be an accident. But as (according to Avicenna) the same thing can never be both substance and accident, it follows that substantial form is never in any subject whatsoever, but always in prime matter’ Stone, Simplicius and Avicenna, p. 100.) 27 I examine texts from three parts of al-Sˇifa¯’: Metaphysics, Physics and Generation and Corruption. I exclude Avicenna’s discussions of the causal roles of the Agent Intellect in his De anima, since these are not directly relevant to the issue of the generation of compounds of form and matter.

Avicenna and Aquinas on Form and Generation 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

259

III Avicenna on the Causal Roles of the Incorporeal ‘Giver of Forms’ Avicenna refers three times to an incorporeal Giver of Forms in the Metaphysics of the Sˇifa¯’. He mentions a separable cause of the existence of form and matter in Metaphysics 2.4, a separable Giver of Forms, which causes the existence of the composite in Metaphysics 6.2 and a separate intellect from which the forms of our lower world emanate in Metaphysics 9.5. In Physics 1.10 of the Sˇifa¯’, he says that ‘it seems that that which gives the constitutive form belonging to natural species is extrinsic to natural things’.28 And in his account of elemental transformation in On Generation and Corruption, Avicenna states that ‘it is of the nature of matter when prepared with a complete preparation for a form that it attains this form from the Giver of Forms to matter and accepts it’.29 Taken together, these passages provide an ambiguous view of the division of labour between corporeal agents and the Agent Intellect in generation. In Metaphysics 2.4, where Avicenna aims to establish the ontological priority of form to matter, he also determines the cause of the existence of form and matter. He argues first that matter cannot be the cause of the existence of form.30 He then argues that form alone cannot be the cause of the existence of matter.31 He concludes that ‘the cause of the existence of matter is something conjoined with form, so that the existence of matter emanates only from that thing’.32 The reason the cause must be conjoined with form is that matter and 28 Avicenna, al-Sˇifa¯’: al-Sama¯‘ al-tabı¯‘i [Physics], 1.10, p. 49. ˙ 29 Avicenna, al-Sˇifa¯’: al-Kawn wa-’l-fasa ¯ d [Generation and Corruption], c. 14, p. 190. 30 Avicenna rejects the claim that matter is the proximate cause of the existence of form for three reasons. He says first that ‘matter is only matter because it has the potency for reception and preparedness. But the prepared insofar as it is prepared is not a cause of existence of what it is prepared for. If it were a cause of existence [of what it is prepared for], then it would have this continuously, without preparedness’ (Avicenna, Metaphysics 2.4, p. 65 (my transl.); Sˇ. Il., p. 83; Pr. ph., p. 96). His second argument relies on the absurdity that ‘the same thing be a cause for something in actuality while it remains in potentiality’ (Avicenna, Metaphysics 2.4, p. 65 (my transl.); Sˇ. Il., p. 83; Pr. ph., p. 96). Matter, as the principle of potentiality in change, cannot actualize itself. Third, ‘if matter were the proximate cause of form, and [if ] matter has no variance in itself, and [if ] whatever was necessitated by that which has no variance has no variance at all in it, it would then necessarily follow that corporeal forms have no variance in them’ (Avicenna, Metaphysics 2.4, p. 65; Sˇ. Il., p. 84; Pr. ph., p. 97). 31 This would be possible were a form never separated from its matter, but is not possible given the corruptibility of the composite: for if ‘this form were alone in itself a cause, matter would cease to exist after [the form] ceases to exist and the commencing form would have another matter that comes to exist through it. That matter would then be originated, and [this origination] would require another matter’ (Avicenna, Metaphysics 2.4, p. 67; Sˇ. Il., p. 85; Pr. ph., p. 98). Avicenna takes this scenario to entail a vicious regress and rejects it. 32 Avicenna, Metaphysics 2.4, p. 67; Sˇ. Il., p. 85; Pr. ph., p. 98.

260 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Kara Richardson

form cannot exist apart from one another. So form is a necessary condition for the existence of matter. Thus he refers to form as the ‘partner’ of a ‘separate’ thing, which causes the existence of matter.33 This separate thing – i. e., the Agent Intellect – is an efficient causal principle of the existence of both form and matter; but it causes the existence of matter through form. The Agent Intellect’s role as cause of existence is also discussed in Book 6 of the Metaphysics of the Sˇifa¯’. In Metaphysics 6.2, Avicenna discusses the relative roles of corporeal agents and the Agent Intellect with respect to the existence of corporeal substances. This discussion occurs in the context of an analysis of efficient causality, which aims to correct misconceptions of the efficient cause or agent (al-fa¯‘il), especially the following one: Someone may think that the agent and the cause are needed only for a thing to have existence after nonexistence and that, once a thing is brought into existence, the thing would [continue to] exist as sufficient unto itself [even] if the cause is no longer present. Thus, someone has thought that a thing is in need of the cause only for its origination, but that, once it is originated and comes to exist, it no longer needs the cause. For such a person, the causes are thus only the causes of origination, being [temporally] prior [to their effects], not simultaneous with them.34

This ordinary but mistaken view holds that that role of an efficient cause is to make something, which persists through itself once it is made. Avicenna argues that what is made needs a cause of its existence so long as it exists. This argument establishes a distinction between causes of existence (wug˘u¯d/esse) and causes of origination (hudu¯t/fieri). The cause of the existence of sublunar things ¯ is the separate Giver of˙ Forms, i. e., the Agent Intellect. Avicenna develops this point in Metaphysics 6.2, where he answers two questions raised by the distinction between causes of existence and causes of coming to be: What sort of causal relationship holds between corporeal agents, such as builders and fathers, and the existence of their effects? What is the efficient cause of a thing’s existence post-origination?35 Avicenna identifies builders as causes of motion: [a]s for the builder, his movement is the cause of a certain motion. Thereafter, his immobility and refraining from motion, or his ceasing to move and affect transportation after having transported, constitute a cause for the termination of that motion. [Now,] that very act of transporting and the termination of this 33 Avicenna, Metaphysics 2.4, p. 68; Sˇ. Il., p. 87; Pr. ph., p. 100. Lizzini notes, and I agree, that although Avicenna mentions there only that the cause of the existence of form and matter is a ‘separable principle’, the role of this separable principle indicates that he refers to the Agent Intellect (Lizzini, The Relation, p. 183). 34 Avicenna, Metaphysics 6.1, p. 198; Sˇ. Il., p. 261; Pr. ph., pp. 296 – 7. 35 Avicenna’s answers to these questions are discussed in Marmura, The Metaphysics, and Druart, Metaphysics, pp. 327 – 48.

Avicenna and Aquinas on Form and Generation 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

261

motion are a cause of a certain combination, and that combination is a cause for a certain shape taking place; and each of [the things] that constitutes a cause coexists with its effect.36

The movements of the builder aim at the form or shape of the building: they terminate in a certain combination of materials, and the combination of materials is the cause of an artifact with a certain shape.37 A father too is identified as a cause of motion: ‘[a]s for the father, he is the cause of the movement of the sperm’, and the (proper) termination of this movement results in a new human being.38 According to this analysis, builders and fathers are efficient causes of movements, which temporally precede, and terminate in, the existence of buildings and sons. In terms of the distinction drawn in Metaphysics 6.1 between causes of existence and causes of origination, builders and fathers count as causes of origination: they bring something into being. But such causes are merely accidental or helping causes of existence. This view accords with a principle Avicenna holds true: every cause coexists with its effect.39 He concludes that a correct analysis of the causes of the existence of a thing reflects the simultaneity principle: Thus, the true causes coexist with the effect. As for those that are [temporally] prior, these are causes either accidentally or as helpers. For this reason, it must be believed that the cause of the building’s shape is combination, the cause of [the latter] being the natures of the things being combined and their remaining in the way they are composed, the cause of [these natures] being the separable cause that enacts the natures. The cause of the son is the combination of his form with matter through the cause that endows forms…We thus find that the causes coexist with [their] effects.40

36 Avicenna, Metaphysics 6.2, p. 201; Sˇ. Il., p. 264; Pr. ph., p. 301. 37 In Metaphysics 6.4 Avicenna says, ‘[y]ou have known that one and the same thing may, in different respects, be form, purpose, and efficient principle. This is also the case in art. For art is the form of the artifact in the soul. For building is in itself the movement toward the form of the house. This is the principle from which the realization of the form in the matter of the house proceeds’ (Avicenna, Metaphysics 6.4, p. 219; Sˇ. Il., p. 283; Pr. ph, p. 325). 38 Avicenna, Metaphysics 6.2, p. 201; Sˇ. Il., p. 264; Pr. ph, p. 301. 39 The simultaneity principle has Aristotelian origins. Aristotle claims in Physics 2.3 that ‘causes which are actually at work and particular exist and cease to exist simultaneously with their effect, e. g. this healing person with this being-healed person and that housebuilding man with that being-built house; but this is not always true of potential causes – the house and the housebuilder do not pass away simultaneously’ (Aristotle, Physics 2.3. 195b16 – 21). Avicenna argues for the simultaneity principle in Metaphysics 4.1. 40 Avicenna, Metaphysics 6.2, p. 202; Sˇ. Il., p. 265; Pr. ph., p. 302.

262 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Kara Richardson

The existence of a building is due to its components and to the efficient cause of the natures of its components. The existence of a human being is due to the combination of form and matter and to a separable Giver of Forms, i. e., the Agent Intellect. He then argues that the ‘assisting and preparatory causes’ of the existence of a thing are infinite, one [temporally] preceding the other’, but its ‘essential causes’ are finite and coexist with it.41 Since Avicenna’s discussions of the causal roles of the Agent Intellect in Metaphysics 2.4 and 6.2 are concerned with existence, rather than coming to be, they do not provide any clear evidence of his view of substantial generation. That he calls the incorporeal cause of the existence of form, matter and the composite a Giver of Forms is unremarkable given his view of form as that by which indeterminate matter becomes a subsistent thing of some species; this epithet does not indicate that he endorses the Infusion Model of generation. Moreover, Avicenna’s distinctions between causes of origination and causes of existence, and between accidental, assisting or preparatory causes of existence and essential causes of existence, which are a crucial part of his account of the Agent Intellect’s role as the cause of the existence of composite substances, are also a crucial part of Aquinas’ reconciliation of the Eduction Model of generation with the view that God is the cause of the existence of composite substances. This issue is discussed in Section IV. Avicenna’s reference to the Agent Intellect in Physics 1.10 of the Sˇifa¯’ is also ambiguous with respect to his model of generation. In that text, Avicenna first defines the efficient cause in natural things as a principle of motion in another, and states that motion is the going out from potentiality to actuality in matter.42 He then draws a distinction between the principle of motion considered as a ‘preparer’ (muhayyi’) and the principle of motion considered as a ‘perfecter’ (mutammim). The preparer ‘is that which puts the matter in order’, or, in other words disposes the matter.43 Avicenna says that the perfecter ‘is that which gives the form’.44 The referent of ‘perfecter’ here is a matter of controversy, since after stating that the perfecter is that which gives the form, Avicenna says that ‘it seems that that which gives the constitutive form belonging to natural species is extrinsic to natural things’.45 Here Avicenna clearly states that the form shared by members of a natural species has a cause extrinsic to natural things. But he stops short of claiming that whenever a new compound of form and matter is made, an incorporeal substance produces its form ex nihilo. 46 In my view, 41 42 43 44 45 46

Avicenna, Metaphysics 6.2, p. 202; Sˇ. Il., p. 266; Pr. ph., p. 303. Avicenna, al-Sˇifa¯’: al-Sama¯‘ al-tabı¯‘i [Physics], 1.10, p. 48. Avicenna, al-Sˇifa¯’: al-Sama¯‘ al-t˙ abı¯‘i [Physics], 1.10, p. 49. Avicenna, al-Sˇifa¯’: al-Sama¯‘ al-t˙ abı¯‘i [Physics], 1.10, p. 49. Avicenna, al-Sˇifa¯’: al-Sama¯‘ al-t˙ abı¯‘i [Physics], 1.10, p. 49. ˙ Brand, Book of Causes, p. 46, n. 7 identify Physics 1.10 Lee, St. Thomas, p. 40, n. 1 and as support for the view that Avicenna endorses what I call the Infusion Model of

Avicenna and Aquinas on Form and Generation 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

263

Avicenna here refers to the Agent Intellect as a cause of species, rather than a cause of individuals. Avicenna distinguishes these two types of cause in Metaphysics 6.3. In Metaphysics 6.3, Avicenna challenges the causal likeness principle – the view that agent and effect are always alike in kind – as well as the belief that agents are always ‘worthier and stronger’ than their patients with respect to whatever it is the agent gives the patient.47 Avicenna argues that this ‘is neither evident nor true in every respect, unless what it bestows is existence itself, and reality. For then the giver is more worthy of what it bestows than the recipient of what is bestowed’.48 He elaborates the view that the agent is superior to the patient when the agent bestows existence and reality by drawing a distinction between efficient causes of a thing insofar as it is a member of a species and efficient causes of a thing insofar as it is an individual. He argues first that the causes of a thing insofar as it is a member of a species must differ from it in kind. His support for this view rests on the idea that if the cause of a thing insofar as it is a member of some species X were itself a member of species X, then the cause of species X would remain unexplained: the two species [i.e., the species of the cause and the species of the effect] are not one [and the same], since what is being sought after is the cause of that species … These [latter] would be essential causes of the thing absolutely caused with respect to the species of the effect.49

By contrast, the cause of a thing qua individual may be the same in kind as it: this fire is not the cause of that fire in that it is the cause of the specificity of fire, but in that it is the cause of some fire. If considered in terms of specificity, it would be the cause of specificity accidentally. The case is similar with [the causal relation] of father to son, not inasmuch as this is a father and that a son, but with respect to the existence of humanity.50

47 48 49 50

generation. Lee’s and Brand’s interpretation is contested by Robert Wisnovsky in his Ph.D. dissertation, Avicenna on Final Causality. Wisnovsky states his own position on the respective roles of corporeal agents and the Agent Intellect in substantial generation briefly: he says that ‘the Agent Intellect gives existence to the form/matter compound, but the matter is set in motion toward the form by its natural principle of motion, i. e. the perfecter. It is thus through the mediation of the perfecter in nature (i. e. the form itself ) that the Agent Intellect causes the existence of the compound’ (Wisnovsky, Avicenna, pp. 98 – 9). My own interpretation of Physics 1.10 is generally in accord with this one. But I question the view that the perfecter in nature is the form itself, since Avicenna says that the perfecter gives the form. I consider semen and seeds to be the ‘perfecters in nature’ to which Wisnovsky refers. Avicenna, Metaphysics 6.3, p. 205; Sˇ. Il., p. 268; Pr. ph., p. 307. Avicenna, Metaphysics 6.3, p. 205; Sˇ. Il., p. 268; Pr. ph., p. 307. Avicenna, Metaphysics 6.3, p. 207; Sˇ. Il., p. 270; Pr. ph., p. 310. Avicenna, Metaphysics 6.3, p. 208; Sˇ. Il., p. 271; Pr. ph., p. 310.

264 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Kara Richardson

He argues that agents of the second type are not superior to their patients: for example, when this fire causes that fire, agent and patient are equal, since the fiery form does not admit of degree.51 But agents of the first type are superior to their patients: though ‘existence inasmuch as it is existence does not vary in terms of strength and weakness’, still it varies in terms of ‘priority and posteriority, absence of need and need, and necessity and possibility’.52 In these ways, the cause of thing qua member of a species is superior to that thing, since it is the source of its very existence and reality. The distinction between causes of species and causes of individuals suggests that a superior principle – in my view, Avicenna’s Agent Intellect – is the cause of a type replicated by tokens of the type. This view does not entail an Infusion Model of generation. (Aquinas himself holds this view to be compatible with the Eduction Model of generation. This issue is discussed in Section IV.) If Avicenna’s claim in Physics 1.10 that ‘that which gives the constitutive form belonging to natural species is extrinsic to natural things’ refers to an incorporeal cause of the species of natural things, then this claim is compatible with the view that corporeal agents educe form from the potency of matter in generation.53 The latter view also accords with Avicenna’s claims about the powers of semen and seeds in generation, which suggest that he considers these corporeal agents to be ‘perfecters’ who play a form-giving role, rather than ‘preparers’, who dispose matter for form.54 51 52 53 54

Avicenna, Metaphysics 6.3, p. 206; Sˇ. Il., p. 269; Pr. ph., p. 308. Avicenna, Metaphysics 6.3, p. 213; Sˇ. Il., p. 276; Pr. ph., pp. 317 – 18. Avicenna, al-Sˇifa¯’: al-Sama¯‘ al-tabı¯‘i [Physics], 1.10, p. 49. ˙ he claims that that what is in potency depends on a For example, in Metaphysics 4.2, proximate mover which is for the most part something of the same kind (mug˘a¯nis) ‘existing in actuality before the action – as in the case of the hot that heats and the cold that cools’ (Avicenna, Metaphysics 4.2, p. 142; Sˇ. Il., p. 184; Pr. ph., p. 212). Furthermore, ‘it is often the case that that which is in potency, inasmuch as it is the subject bearing potency, comes into existence through that which is in act, where the act is prior to the potency in time, not [contemporaneous] with it. For sperm comes to be from the human, and seed from the tree, so that, from [the former], a human comes to be, and, from [the latter], a tree. Hence, the supposition that the act in these matters precedes potency has no priority to the supposition that potency precedes the act’ (Avicenna, Metaphysics 4.2, p. 142; Sˇ. Il., p. 184; Pr. ph., p. 212). Here Avicenna seems to endorse Aristotle’s belief that biological organisms generally come to be from corporeal things, which are like them in kind. In Metaphysics 5.9, he argues that sperm has the potency for the formation of the human. He says so while illustrating a point about potencies, namely that a potency can be in potency: ‘potency, inasmuch as it is, is the potency of an existence in act. Potency sometimes also exists in potency, this being the potency which is remote from act but which becomes in act a proximate potency. For the proximate potency for the formation of the human exists potentially in nourishment. Then, if [the nourishment becomes] sperm, this proximate potency exists in act. It is only that its action [while it is still a proximate potency] does not exist’ (Avicenna,

Avicenna and Aquinas on Form and Generation 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

265

The passages that offer the most compelling textual evidence for the view that Avicenna adopts an Infusion Model have to do with elemental bodies. In Metaphysics 9.5, Avicenna discusses the generation of the elements by the first causes in the context of his emanationist account of creation. He claims first that since the elements are generable and corruptible, ‘their proximate causes must be things that receive a species of change and motion’ and so ‘that which is a pure intellect is not alone a cause for their existence’.55 He identifies these proximate causes as the celestial spheres, or, more specifically, their motions. These motions play a role in the generation of the elements because they contribute to the preparedness of some portion of matter for the reception of one elemental form, rather than another, e. g., for the reception of the form of fire, rather than the form of water. Avicenna refers to this preparatory function as the act of ‘specifying matter’: [t]he things that specify matter are the things that prepare it. The preparer is that through which there comes to be, in the thing prepared, something by virtue of which its appropriateness for [the reception] of a specific thing is more appropriate than [the reception of some] other thing. This act of preparing renders preponderant the existence in it of the more appropriate [form] from the principles that bestow forms.56

He then illustrates preparedness by the example of the transformation of water into fire: This is similar to water when its warming is made excessive, whereby the alien warmth and the watery form combine, [the former] being remote in appropriateness from the watery form [but] greatly appropriate for the fiery form. If that [warming] is rendered excessive and the appropriateness intense, the preparedness becomes intense. It thus becomes aright for the fiery form to emanate and aright for this [watery form] to cease.57

In these passages Avicenna illustrates his view that the elements turn into one another when they surpass a certain degree of warmth/coldness or moistness/ dryness. This view is clarified in On Generation and Corruption: Metaphysics 5.9, p. 193; Sˇ. Il., p. 252; Pr. ph., pp. 289 – 90). Here Avicenna claims that nourishment, which can become sperm, has the power to acquire the power proper to sperm, i. e., the power for human formation. In this passage, the sperm’s potency seems clearly to be an active power: Avicenna claims that the sperm’s potency is for the formation of the human. Moreover, the view that male semen has active power is the ground for Avicenna’s distinction between the roles of male semen and female semen in the Kita¯b al-Hayawa¯n (Book of Animals). On this aspect of Avicenna’s account of sexual ˙ Musallam, The Human Embryo, pp. 32 – 4 and McGinnis, On the generation see Moment, p. 55. 55 Avicenna, Metaphysics 9.5, p. 334; Sˇ. Il., p. 410; Pr. ph., p. 488. 56 Avicenna, Metaphysics 9.5, p. 335; Sˇ. Il., p. 411; Pr. ph., pp. 489 – 90. 57 Avicenna, Metaphysics 9.5, p. 336; Sˇ. Il., p. 411; Pr. ph., p. 490.

266 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Kara Richardson

[e]ach one of the elements has latitude for receiving increase and decrease in its quality. For it may increase and decrease in its natural or accidental quality. And it may do this while maintaining still its form and species. But increase and decrease in this has two extreme limits. When they are surpassed, the complete disposition for its form is annulled in matter. And it is prepared with a complete preparation for another form. And it is of the nature of matter when prepared with a complete preparation for a form that it attains this form from the giver of forms to matter and accepts it. And it is for this reason that indistinct matter is distinguished such that it is matter for successive forms. And this from the giver of forms.58

Avicenna’s account of the roles played by preparers and by the principles that bestow form in these passages are suggestive of the Infusion Model of substantial generation, and seem to support the view that he endorses this model, at least in cases of elemental transformation. But Avicenna’s intention in such passages may be to identify the Agent Intellect as the efficient cause of the four species of elemental bodies, which are the basis of the system in which elemental transformation occurs. Within this system, fire, for example, has the power to heat water to such a degree that water is completely prepared for the airy form. But the emanation of the airy form is not attributed to the fire that heats it because the fire is not the cause of the fact that water heated to a certain degree becomes air. On this view, elemental transformation is more the product of a system (and so of the efficient cause of this system), than of the causal activity of any individual corporeal agent. The view that Avicenna adopts the Infusion Model of generation, at least in cases of elemental transformation, might find additional support in his account of elemental forms.59 Avicenna denies that the form of an elemental body can be conflated with the two primary qualities, which are characteristic of that body, i. e., he denies that the form of fire can be conflated with the qualities of heat and dryness.60 Avicenna’s view relies on the following claims. First, qualities are accidents and accidents are ontologically posterior to their subjects. Second, substantial forms are ontologically prior to their subjects. Given that substantial forms cannot be identified with their sensible qualities, we cannot know them directly: they are occult entities.61 Stone argues that this view of the elements, as well as a related one about the mixtures they form, supports Avicenna’s ‘occasionalist’ solution to the question of the origins of ‘secondary qualities’, such as colours, of ‘faculties’, such as the magnet’s power to attract iron, as well 58 Avicenna, al-Sˇifa¯’: al-Kawn wa-’l-fasa¯d [Generation and Corruption], c. 14, p. 190. 59 This was suggested to my by Jon McGinnis in his appraisal of my dissertation and by Gad Freudenthal at the Villa Vigoni conference on the reception of Avicenna’s Metaphysics. 60 On this aspect of Avicenna’s view, see Avicenna, al-Sˇifa¯’: al-Kawn wa-’l-fasa¯d [Generation and Corruption], c. 6. This issue is also discussed in Stone, Simplicius and Avicenna, and Stone, Avicenna’s Theory. 61 This point is identified as an Avicennian innovation in Stone, Avicenna’s Theory.

Avicenna and Aquinas on Form and Generation 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

267

as of substantial forms.62 Since these occult entities cannot emerge from mixtures of the elements, they must be ‘called down’ from the heavens.63 Stone has shown that Avicenna holds that forms, faculties and qualities such as colours cannot be reduced to the primary qualities (hot, cold, wet and dry) and so that their existence is not explained by the existence of these qualities but rather requires independent explanation. But these claims are compatible with the view that through possession of these non-derivative entities, composite substances exercise causal powers, including generative ones, and thus that generation occurs through natural processes.64 (The fact that these entities are occult does not bar them from acting or being principles of activity, even if their occultness renders them useless for explanatory purposes.) In my view, Avicenna’s elemental theory and his related account of mixture do not suffice to show that he offers an ‘occasionalist’ account of generation. Moreover, I find questionable the application of the term ‘occasionalist’ to Avicenna. It seems to me that he has no occasionalist axe to grind. Even where he emphasizes the superiority of divine causality, he recognizes lesser agents as genuine causes.65 He very clearly holds that corporeal substances are endowed with natures, i. e., 62 Stone, Avicenna’s Theory, pp. 117 – 8. 63 Stone, Avicenna’s Theory, p. 118. 64 Stone’s claim that Avicenna adopts an occasionalist view of the origin of substantial forms may overlook the possibility that the Eduction Model of generation is compatible with a commitment to creation. For he claims that in De potentia q. 5 a. 1, Aquinas too adopts an occasionalist view: ‘although Thomas is generally known as an opponent of Avicenna’s occasionalism, he takes over intact the key part of it: namely, that sublunar causes can only dispose the matter to receive substantial forms, while the actual source of those forms is the angels and celestial bodies’ (Stone, Avicenna’s Theory, p. 119). The phrase ‘actual source’ is ambiguous: Aquinas thinks that God is the creative source of form, but he also thinks, as we have seen, that substantial forms are educed from matter by corporeal agents in generation. This issue is discussed in Section IV. 65 For example, in Metaphysics 6.2, Avicenna argues that creative efficient causality, i. e., causing existence after absolute non-existence, is superior to non-creative efficient causality, i. e., causing existence after privation in matter; but he does not deny that noncreative efficient causes have power: ‘The originated in the sense [of the thing] that does not require time must be such that either its existence is after absolute nonexistence or its existence is after nonexistence which is not absolute, being after a specific opposing privation in existing matter, as you have known. If its existence were after absolute nonexistence, then its proceeding from the cause in this manner would be ‘creation,’ and it would represent the highest mode of the giving of existence, because nonexistence would have been utterly prevented, existence being fully empowered over it. If nonexistence is firmly established in a manner whereby it precedes existence, then its formation would be impossible except out of matter. In this case, the empowerment of bringing into existence – I mean, the existence of one thing from another – would be weak, of short duration, and intermittent’ (Metaphysics 6.2, p. 204; Sˇ. Il., p. 267; Pr. ph., p. 305).

268 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Kara Richardson

internal principles of motion and change. And he appears to reject the epistemological arguments for occasionalism.66 Avicenna’s discussions of the causal roles played by the Agent Intellect do not provide a clear answer to the question whether he endorses the Infusion Model of generation. In my view, these texts suggest that he points to this incorporeal agent to explain the existence of composite substances, and their species, rather than to explain individual cases of substantial generation. While Aquinas’ argument against the Infusion Model of generation in De potentia q. 3 a. 8 is persuasive, it is problematic insofar as it takes Avicenna as one of its targets. In my view, the attribution to Avicenna of the Infusion Model of generation is based on a coarse reading of Avicenna’s discussions of the causal roles played by the Agent Intellect, which conflicts with his account of the unity of form/matter compounds. This interpretation has roots in Averroes’ Long Commentary on the Metaphysics. According to Averroes, Avicenna’s alleged commitment to creation is the source of his (and al-Fa¯ra¯bı¯’s) divergence from Aristotle’s view that forms are educed from matter: It is only because the earlier group of men [namely, Ibn Sı¯na¯ and al-Fa¯ra¯bı¯] neither understood Aristotle’s demonstration [that the forms are not generated in themselves] nor accepted its truth that they were undone. The conceit does not belong to Ibn Sı¯na¯ alone, but also to al-Fa¯ra¯bı¯; for it is evident in his book on the two philosophers [Plato and Aristotle] that he had problems concerning this account. This earlier group of men was inclined toward the thought of Plato only because it was an opinion very much akin to that upon which the theologians of our religion rely in this account, namely, that the agent of all [generated] things is one, and that some of the [generated] things do not bring about an effect in others. In other words, they believed that from some of them creating others they would be committed to the infinite series of actual causes, and so they asserted an incorporeal agent.67

Averroes’ account of Avicenna’s fall is misleading. Avicenna does not posit an incorporeal agent in order to avoid an infinite regress of corporeal causes of generation. Indeed, as mentioned above, Avicenna endorses an infinite series of corporeal causes of generation in Metaphysics 6.2. There he posits an incorporeal agent in order to explain the very existence of generated things. Nevertheless, there is a tension between the view that there is an incorporeal cause of the existence of generated things and the view that corporeal agents educe form 66 As Marmura notes, Avicenna endorses the claim that sense experience establishes only concomitance, and not the necessary connections needed to establish causal relationships, but he does not conclude that we cannot know that natural things have causal powers. Marmura suspects that Avicenna’s arguments related to this issue are in response to Asˇ‘arite occasionalism. See Marmura, The Metaphysics, pp. 183 – 7. 67 Averroes, Commentary on Metaphysics, Zeta 9, comm. 31, p. 334.

Avicenna and Aquinas on Form and Generation 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

269

from the potency of matter in generation. This tension is not peculiar to Avicenna’s metaphysics; it is a feature of Aquinas’ metaphysics as well. In Section IV, I examine Aquinas’ treatment of this tension in De potentia q. 5 a. 1, and in Summa theologiae 1a q. 104 a. 1. I show that these texts reveal a very positive reception of some of Avicenna’s discussions of the causal roles of the Agent Intellect, and that Aquinas’ resolution of the tension between the Eduction Model of generation and God’s role as the cause of the existence of generated things integrates several Avicennian ideas.

IV Aquinas’ Arguments for Divine Conservation Aquinas’ argument in De potentia q. 5 a. 1 for the view that God preserves all things in being relies on the following account of the relationship between a thing made and an efficient cause: The existence of a thing made depends on an efficient cause insofar as it depends on the form itself of the thing made. But there is some efficient cause on which the thing made depends not per se and according to the account (ratio) of [its] form but only per accidens: thus the form of the generated fire depends on the generating fire not per se and according to the account of its species, since they hold the same rank in the order of things, nor is the form of fire otherwise in the thing generated than in the generator, but rather it is distinguished from it only by a material distinction (divisio) according as it is in different matter. Whence since the form of the generated fire has some cause, the form itself must depend on a higher principle, which is the cause of the form itself per se and according to the account proper to its species.68

Aquinas’ first premise – namely, that ‘the existence of a thing made depends on an efficient cause insofar as it depends on the form itself of the thing made’ – reflects his view that the form of the composite is that by which the composite exists and thus that the existence of the composite depends on its form. This suggests that the efficient cause of the existence of the composite and its form are identical. Aquinas then distinguishes two types of efficient cause, one on which a thing made depends per se and another on which it depends per accidens. In the case of a generated fire, for example, the form of that fire depends on the 68 Aquinas, De potentia, q. 5 a. 1: ‘Secundum hoc ergo esse rei factae dependet a causa efficiente secundum quod dependet ab ipsa forma rei factae. Est autem aliquod efficiens a quo forma rei factae non dependet per se et secundum rationem formae, sed solum per accidens: sicut forma ignis generati ab igne generante, per se quidem, et secundum rationem suae speciei non dependet, cum in ordine rerum eumdem gradum teneat, nec forma ignis aliter sit in generato quam in generante; sed distinguitur ab ea solum divisione materiali, prout scilicet est in alia materia. Unde cum igni generato sua forma sit ab aliqua causa, oportet ipsam formam dependere ab altiori principio, quod sit causa ipsius formae per se et secundum propriam speciei rationem.’

270 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Kara Richardson

generating fire only accidentally. It cannot depend on the generating fire essentially, since the generating and the generated fire share the same type of form. Thus it must depend on a ‘higher’ principle, the per se cause of the form itself (God). Here Aquinas adapts Avicenna’s argument in Metaphysics 6.3 for a distinction between causes of species and causes of individuals. (This argument is discussed in Section III.) The ground for Avicenna’s distinction between causes of species and causes of individuals is the idea that the cause of an effect’s specificity cannot be another member of its species. Aquinas offers the following explanation in Summa theologiae 1a q. 104 a. 1: Now it is clear that of two things in the same species one cannot be essentially the cause of the other’s form as such, since it would then be the cause of its own form, since both forms have the same nature; but it can be the cause of this form inasmuch as it is in matter – in other words, it may be the cause that this matter receives this form. And this is to be the cause of becoming, as when man begets man, and fire causes fire.69

Aquinas argues that if the generating fire were the essential cause of the existence of any fiery form, then it would be the cause of the existence of its own form, which is impossible. (Nothing is the cause of itself.) He then alludes to a distinction between causes of coming to be and causes of existence, which Avicenna defends in Metaphysics, Book 6, chapters 1 – 2: corporeal agents cause the becoming, but not the existence of generated things.70 The cause of the existence of generated things, Aquinas argues (in Summa theologiae 1.104.1, as well as in De potentia 5.1), is a ‘higher’ principle, which is the essential or per se cause of form: this is God.71 69 Aquinas, Summa theologiae, 1a q. 104 a. 1: ‘Manifestum est autem quod, si aliqua duo sunt eiusdem speciei, unum non potest esse per se causa formae alterius, inquantum est talis forma, quia sic esset causa formae propriae, cum sit eadem ratio utriusque. Sed potest esse causa huiusmodi formae secundum quod est in materia, idest quod haec materia acquirat hanc formam. Et hoc est esse causa secundum fieri; sicut cum homo generat hominem, et ignis ignem.’ 70 The provenance of the distinction is noted by Doolan, The Causality, p. 399, n. 29 and Wippel, Thomas Aquinas, p. 202. 71 In De potentia 5.1, the argument for this conclusion – sc., that the essential or per se cause of the existence of the form of the composite is God – relies on two key claims. First, ‘the existence of a form in matter implies no movement or change except accidentally’ and, second, bodies act only by being moved (Aquinas, De potentia, q. 5 a. 1 co.): ‘Cum autem esse formae in materia, per se loquendo, nullum motum vel mutationem implicet, nisi forte per accidens; omne autem corpus non agat nisi motum, ut philosophus probat, necesse est quod principium ex quo per se dependet forma, sit aliquod principium incorporeum; per actionem enim alicuius principii dependet effectus a causa agente.’ I take Aquinas to reason as follows: if bodies act only by being moved, then the effects of their actions will be movements or changes. Since the existence of a

Avicenna and Aquinas on Form and Generation 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

271

The claim that corporeal agents cause the becoming but not the existence of corporeal things, together with the claim that that existence of the composite depends on its form might seem to provide support for the Infusion Model of generation. Aquinas raises this objection to his claim that God preserves all things in being in De potentia q. 5 a. 1: Someone might say that lower agents are causes of becoming and not of being. For this reason the existence of an effect remains once the action of its cause of becoming has ceased. But God causes not only the becoming but also the being of things. For this reason the existence of things cannot remain if divine action ceases. – On the contrary, every generated thing has existence through its form. If therefore lower generating causes are not causes of existence, they will not be causes of form and thus the forms which are in matter are not from forms which are in matter, as according to the opinion of the Philosopher, who says that the form which is in this flesh and bones is from the form which is in those flesh and bones, but rather it follows that forms in matter are from forms without matter, as according to the opinion of Plato, or from the giver of forms, as according to the opinion of Avicenna.72

Here the objector assumes as true the view of the relative roles of lower generating agents and God with respect to the existence of a thing, which Aquinas adopts from Avicenna: lower generating agents are causes of becoming, but not of being. The objector suggests that on this view, God is the cause of the form of the thing generated: for if God is the cause of the existence of a thing, and the existence of that thing is due to its form, then God is the cause of its form. This conclusion seems to conflict with the view that corporeal agents educe form from the potency of matter, and to support the Infusion Model of generation. Aquinas rejects this line of reasoning. He replies that as agents of becoming, lower generating causes educe form from matter. In this way (i. e., by educing), they are causes of substantial forms and thus they are ‘principles of existence considered in terms of its beginning not as considered absolutely’.73 In De potentia q. 5 a. 1, Aquinas supports his claim that the Eduction Model is compatible with God’s role as the cause of the existence of generated things through a cooperative account of the roles of God and corporeal agents in generation. Insofar as a corporeal agent is a cause of form it acts in the power of form in matter does not involve movement or change, no body can be the cause of the existence of a form in matter. So the cause of the existence of a form in matter must be an incorporeal principle, which Aquinas identifies as God. 72 Aquinas, De potentia, q. 5 a. 1 arg. 5 73 Aquinas, De potentia, q. 5 a. 1 ad 5: ‘Si autem ponamus formas substantiales educi de potentia materiae, secundum sententiam Aristotelis, agentia naturalia non solum erunt causae dispositionum materiae, sed etiam formarum substantialium; quantum ad hoc dumtaxat quod de potentia educuntur in actum, ut dictum est, et per consequens sunt essendi principia quantum ad inchoationem ad esse, et non quantum ad ipsum esse absolute.’

272 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Kara Richardson

an incorporeal principle as if it were its instrument.74 This agent thus makes a genuine causal contribution to generation, even though it is merely a per accidens cause of the existence of generated things. In Section III, I discussed Avicenna’s very similar claims Metaphysics 6.2: due to their causal roles in generation, corporeal agents are identified as accidental or helping causes of the existence of the composite, while the Agent Intellect is identified as the essential cause of the existence of the composite. According to Aquinas, the difference between his view of generation and Avicenna’s is that when Avicenna identifies builders and fathers as accidental or helping causes of existence, he means to deny that they educe form from matter. In Section II, I argued that this interpretation conflicts with Avicenna’s view of hylemorphic unity, and in Section III, I argued that it is based on a questionable reading of Avicenna’s discussions of the causal roles of the incorporeal Giver of Forms. Aquinas’ arguments for divine conservation borrow several distinctions from Avicenna’s metaphysics in order to support the view that there is an incorporeal agent who causes the existence of the composite by causing its form, and to explain how the latter view accords with the Eduction Model of generation. Aquinas’ use of these distinctions shows that key parts of Avicenna’s view of the causal roles of the Agent Intellect are adopted as important features of Thomistic metaphysics. Aquinas’ positive reception of these aspects of Avicenna’s metaphysics may be obscured by his very negative critique of Avicenna’s account of generation in De potentia q. 3 a. 8, which accuses Avicenna of serious metaphysical errors. Through examining Avicenna’s account of hylemorphic unity, and his discussions of the causal roles of the Agent Intellect, I hope to have shown the demerits of this critique, and also to have challenged the view that Avicenna endorses the Infusion Model of generation.

Bibliography Thomas Aquinas, Summa contra gentiles, Book Three: Providence, Part 1, transl. and ed. V.J. Bourke, Notre Dame: University of Notre Dame Press, 2001. –– , Summa theologiae, Part One (Basic Writings of Saint Thomas Aquinas, vol. 1), ed. A. Pegis, Indianapolis: Hackett, 1997. –– , On the Power of God, transl. and ed. English Dominican Fathers, Westminster, Maryland: The Newman Press, 1952. –– , Quaestiones disputatae, 9th rev. ed., Rome: Marietti, 1953. –– , Opera omnia iussu impensaque Leonis XIII P. M. edita, t. 4 – 5: Pars prima Summae theologiae, Rome: Ex Typographia Polyglotta S.C. de Propaganda Fide, 1888 – 9. 74 Aquinas, De potentia, q. 5 a. 1: ‘Et si aliquod principium corporeum est per aliquem modum causa formae, hoc habet in quantum agit virtute principii incorporei, quasi eius instrumentum.’

Avicenna and Aquinas on Form and Generation 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

273

Aristotle, The Complete Works of Aristotle, ed. J. Barnes, Princeton: Princeton University Press, 1984. Averroes, Commentary on Metaphysics, Zeta 9, in Classical Arabic Philosophy, transl. and ed. J. McGinnis and D.C. Reisman, Indianapolis: Hackett, 2007, pp. 330 – 35. –– , Tafsı¯r ma¯ ba‘d at-tabı¯‘iyya, 3 vols, ed. M. Bouyges, Beirut: Imprimerie Catholique, ˙ 1938 – 52. Avicenna, al-Sˇifa¯’: al-Ila¯hiyya¯t, vol. 1, eds G. C. Anawati and S. Za¯yed, Cairo: Organisation Gnrale des Imprimeries Gouvernementales, 1960. –– , al-Sˇifa¯’: al- Ila¯hiyya¯t, vol. 2, eds M.Y. Moussa, S. Dunya¯ and S. Za¯yed, Cairo: Organisation Gnrale des Imprimeries Gouvernementales, 1960. –– , al-Sˇifa¯’: al-Kawn wa-’l-fasa¯d, ed. M. Qa¯sim, Cairo: Organisation Gnrale des Imprimeries Gouvernementales, 1983. –– , al-Sˇifa¯’: al-Hayawa¯n, ed. ‘A. Montasir, S. Za¯yed and ‘A. Isma¯‘ı¯l, Cairo: Organisation Egyptienne˙ Gnrale du Livres, 1970. –– , al-Sˇifa¯’: al-Sama¯‘ al-tabı¯‘i [Physics], ed. S. Za¯yed, Cairo: Organisation Gnrale des ˙ Imprimeries Gouvernementales, 1983. –– , Avicenna Latinus: Liber de philosophia prima sive Scientia divina, 3 vol., ed. S. Van Riet, Leiden: Peeters; Louvain-Brill, 1977 – 83. –– , Avicenna Latinus: Liber tertius naturalium de generatione et corruptione, ed. S. Van Riet, Leiden: Peeters; Louvain-Brill, 1987. –– , The Metaphysics of ‘The Healing’, transl. and ed. M. Marmura, Provo, Utah: Brigham Young University Press, 2005. D.J. Brand, transl. and ed., Book of Causes, 2nd rev., Milwaukee: Marquette University Press, 1984. H.A. Davidson, Alfarabi, Avicenna, Averroes, on Intellect, Oxford: Oxford University Press, 1992. G.T. Doolan, The Causality of the Divine Ideas in Relation to Natural Agents, International Philosophical Quarterly, vol. 44, 2004, pp. 393 – 409. T. Druart, Metaphysics, in The Cambridge Companion to Arabic Philosophy, ed. P. Adamson, Cambridge: Cambridge University Press, 2005, pp. 327 – 48. G. Freudenthal, The Medieval Astrologization of Aristotle’s Biology: Averroes on the Role of the Celestial Bodies in the Generation of Animate Beings, Arabic Sciences and Philosophy, vol. 12, 2002, pp. 111 – 37. D.N. Hasse, Avicenna’s ‘De anima’ in the Latin West. The Formation of a Peripatetic Philosophy of the Soul, 1160 – 1300, London/Turin: The Warburg Institute/Nino Aragno Editore, 2000. –– , Plato arabico-latinus: Philosophy-Wisdom Literature-Occult Sciences, in The Platonic Tradition in the Middle Ages. A Doxographic Approach, eds S. Gersh, M.J.F.M. Hoenen, Berlin/New York: Walter de Gruyter, 2002, pp. 31 – 65. –– , Spontaneous Generation and the Ontology of Forms in Greek, Arabic, and Medieval Latin Sources, in Classical Arabic Philosophy: Sources and Reception, ed. P. Adamson, London/Turin: The Warburg Institute/Nino Aragno Editore, 2007, pp. 150 – 175. A. Hyman, Aristotle’s ‘First Matter’ and Avicenna’s and Averroes’ ‘Corporeal Form’, in Harry Austryn Wolfson Jubilee Volume, English Section, vol. 1, ed. S. Lieberman, Jerusalem: American Academy for Jewish Research, 1965, pp. 385 – 406. J. Janssens, The Notions of Wa¯hib al-Suwar (Giver of Forms) and Wa¯hib al-‘Aql (Bestower of Intelligence) in Ibn Sı¯na¯, in Actes du XIe Congrs International de Philosophie Mdivale de S.I.E.P.M., 2002, eds M. Pacheco et J. Meirinhos, Turnhout: Brepols, 2006, pp. 551 – 62.

274 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Kara Richardson

P. Lee, St. Thomas and Avicenna on the Agent Intellect, Thomist, 45, no. 1, 1981, pp. 41 – 61. G.W. Leibniz, De ipsa natura, in Die philosophischen Schriften von Gottfried Wilhelm Leibniz, vol. 4, ed. C.I. Gerhardt, Berlin: Weidmannsche Buchhandlung, 1880, pp. 504 – 16. O. Lizzini, The Relation Between Form and Matter: Some Brief Observations on the ‘Homology Argument’ (Ila¯hiyya¯t, II.4) and the Deduction of Fluxus, in Interpreting Avicenna: Science and Philosophy in Medieval Islam, ed. J. McGinnis, Leiden: Brill, 2004, pp. 175 – 85. M. Marmura, The Metaphysics of Efficient Causality in Avicenna (Ibn Sı¯na¯), in Islamic Theology and Philosophy, ed. M. Marmura, Albany, New York: SUNY Press, 1984, pp. 172 – 87. J. McGinnis, On the Moment of Substantial Change: A Vexed Question in the History of Ideas, in Interpreting Avicenna: Science and Philosophy in Medieval Islam, ed. J. McGinnis, Leiden: Brill, 2004, pp. 42 – 61. B. Musallam, The Human Embryo in Arabic Scientific and Religious Thought, in The Human Embryo: Aristotle and the Arabic and European Traditions, ed. G.R. Dunstan, Exeter: University of Exeter Press, 1990, pp. 32 – 48. S.C. Snyder, Albert the Great, Incohatio Formae, and the Pure Potentiality of Matter, American Catholic Philosophical Quarterly, LXX, no. 1, 1996, pp. 63 – 82. A. Stone, Avicenna’s Theory of Primary Mixture, Arabic Sciences and Philosophy, vol. 18, 2008, pp. 99 – 119. –– , Simplicius and Avicenna on the Essential Corporeity of Material Substance, in Aspects of Avicenna, ed. Robert Wisnovsky, Princeton: Princeton University Press, 2001, pp. 73 – 130. J.A. Weisheipl, Aristotle’s Concept of Nature: Avicenna and Aquinas, in Approaches to Nature in the Middle Ages, ed. L. D. Roberts, Binghampton, New York: Centre for Medieval & Early Renaissance Studies, 1982, pp. 137 – 60. J.F. Wippel, Thomas Aquinas on Creatures as Causes of Esse, International Philosophical Quarterly, 40, 2000, pp. 197 – 213. R. Wisnovsky, Avicenna on Final Causality, Ph.D. Dissertation, Princeton University, 1994. B.H. Zedler, Saint Thomas and Avicenna in the De Potentia Dei, Traditio, 6, 1948, pp. 105 – 59.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Immateriality and Separation in Avicenna and Thomas Aquinas Pasquale Porro If we consider the traditional, Scholastic classification of the speculative sciences, it seems rather obvious that first philosophy, or metaphysics, has little to do with matter or with material beings, and has much to do, or has only to do, with beings separate from matter and motion. This is explicitly affirmed by many medieval masters, including Thomas Aquinas. Just to mention a few examples: For since each thing has intellective power by virtue of being free from matter, those things must be intelligible in the highest degree which are altogether separate from matter. … Now those things are separate from matter in the highest degree which abstract not only from designated matter, ‘as the natural forms taken universally, of which the philosophy of nature treats’, but from sensible matter altogether; and these are separate from matter not only in their intelligible constitution, as the objects of mathematics, but also in being, as God and the intelligences.1 (Thomas de Aquino, In Metaph., Prooemium, transl. Rowan). Consequently, separation from matter and motion, or connection with them, essentially belongs to an object of speculation, which is the object of speculative science. As a result, the speculative sciences are differentiated according to their degree of separation from matter and motion. … There are still other objects of speculative knowledge that do not depend upon matter for their being, because they can exist without matter; either they never exist in matter, as in the case of God and the angels, or they exist in matter in some instances and not in others, as in the case of substance, quality, being, potency, act, one and many, and the like. The science that treats of all these is theology or divine science, which is so called because its principal object is God.2 (Thomas de Aquino, Super Boetium De Trinitate, q. V, a. 1, transl. Maurer). 1

2

Thomas de Aquino, In Metaph., Prooemium, eds Cathala and Spiazzi, p. 1: ‘cum unaquaeque res ex hoc ipso vim intellectivam habeat, quod est a materia immunis, oportet illa esse maxime intelligibilia, quae sunt maxime a materia separata. … Ea vero sunt maxime a materia separata, quae non tantum a signata materia abstrahunt, sicut formae naturales in universali acceptae, de quibus tractat scientia naturalis, sed omnino a materia sensibili. Et non solum secundum rationem, sicut mathematica, sed etiam secundum esse, sicut Deus et intelligentiae.’ Thomas de Aquino, Super Boetium De Trinitate, q. V, a. 1, ed. leon., p. 138, lines 135 – 63: ‘Sic ergo speculabili, quod est obiectum scientie speculatiue, per se competit separatio a materia et motu, uel applicatio ad ea; et ideo secundum ordinem remotionis a materia et motu scientie speculatiue distinguntur. Quedam ergo speculabilium sunt que

276 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Pasquale Porro

It must be understood, therefore, that there are some things whose existence depends upon matter, and which cannot be defined without matter. Further there are other things which, even though they cannot exist except in sensible matter, have no sensible matter in their definitions. … And this is true of all the mathematicals, such as numbers, magnitudes and figures. Then, there are still other things which do not depend upon matter either according to their existence or according to their definitions. And this is either because they never exist in matter, such as God and the other separated substances, or because they do not universally exist in matter, such as substance, potency and act, and being itself. Now metaphysics deals with things of this latter sort. Whereas mathematics deals with those things which depend upon sensible matter for their existence but not for their definitions. And natural science, which is called physics, deals with those things which depend upon matter not only for their existence, but also for their definition.3 (Thomas de Aquino, In Phys., I, lect. 1, transls Blackwell, Spath, Thirlkel).

Of course, we could say that this view had already been expressed in Aristotle’s Metaphysics, and especially in E 1, which is nevertheless a text to be read with some caution, as we shall see. Yet, if the medieval masters generally agree on the fact that metaphysics concerns what is immaterial, they disagree about the proper subject of metaphysics, i. e., the question of whether it coincides with God and other separate substances or with being qua being.4 There is no doubt that this alternative can also be found in Aristotle, for his new science or ‘the science we are searching for’ (1pifgtoul]mg 1pist^lg, as Aristotle says in B 1, 995a24) has at least a twofold structure: on the one hand, in C, as is wellknown, the subject of this science is identified with being qua being; on the other, for instance in E 1, the same science is called heokocij^: in other words, it

3

4

dependent a materia secundum esse, quia non nisi in materia esse possunt. … Quedam uero speculabilia sunt que non dependent a materia secundum esse, quia sine materia esse possunt, siue numquam sint in materia, sicut Deus et angelus, siue in quibusdam sint in materia et in quibusdam non, ut substantia, qualitas, ens, potentia, actus, unum et multa, et huiusmodi; de quibus omnibus est theologia, id est scientia diuina, quia precipuum in ea cognitorum est Deus. Que alio nomine dicitur metaphisica …’ [The italics are mine]. Thomas de Aquino, In Phys., I, lect. 1, ed. Maggiolo, p. 3, §§ 2 – 3: ‘Sciendum est igitur quod quaedam sunt quorum esse dependet a materia, nec sine materia definiri possunt: quaedam vero sunt quae licet esse non possint nisi in materia sensibili, in eorum tamen definitione materia sensibilis non cadit. … Quaedam vero sunt quae non dependent a materia nec secundum esse nec secundum rationem; vel quia numquam sunt in materia, ut Deus et aliae substantiae separatae; vel quia non universaliter sunt in materia, ut substantia, potentia et actus, et ipsum ens. De huiusmodi igitur est Metaphysica: de his vero quae dependent a materia sensibili secundum esse sed non secundum rationem, est Mathematica: de his vero quae dependent a materia non solum secundum esse sed etiam secundum rationem, est Naturalis, quae Physica dicitur.’ Suffice it to consider the material discussed in this regard by Zimmermann in his well known volume Ontologie oder Metaphysik. See also Porro, ed., Metaphysica–sapientia–scientia divina.

Immateriality and Separation in Avicenna and Thomas Aquinas 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

277

is a theology (or divine science) which concerns what is separate from matter and motion (peq· wyqist± ja· !j_mgta). The history of the reception of Aristotle’s Metaphysics is above all the history of all the attempts – from the ancient commentators to Heidegger’s famous theses on the onto-theological constitution of Western metaphysics – to find a way to balance or to combine these two distinct subject-matters: being qua being and the divine. It is generally agreed that this twofold structure originates in the fact that Aristotle attributes to his new science two distinct features: metaphysics should be both universal and first (or primary), and so it should deal, at one and the same time, with that which is not a genus, because it encompasses all possible genera (being qua being), and with that which is a genus, the supreme or highest genus, the divine, i. e. that which is eternal, immaterial and immovable. But how can a qualified subject (being qua immovable, eternal and immaterial, or being qua divine) coincide with an unqualified subject (being qua being)? Or, to reverse the question, how can being qua being exclude from itself material being, if matter is one of the principles which is included in the universality of being? In the thirteenth century, Thomas Aquinas sought out a possible solution to this question, and found it in Avicenna. But while borrowing from Avicenna the essential part of the solution (i. e. the distinction between two different meanings of ‘immaterial’ and ‘immateriality’, as we shall see), some of his conclusions seem radically different from those of Avicenna. This is precisely the issue I would like to deal with here. My exposition consists of five parts. (1) I will show how the question concerning the immateriality of the subject of metaphysics is raised in Aristotle and in his ancient commentators. (2) I will explain the way in which Thomas Aquinas deals with this issue in his commentary on the Metaphysics and, above all, in his commentary on Boethius’ De trinitate, in which he introduces a distinction between two degrees of abstraction proper to physics and mathematics, and a further distinct operation, called separatio, proper to divine science. (3) I will show how Aquinas draws this distinction from Avicenna, and yet distorts the intentions of Avicenna himself in order to reach a different and perhaps divergent conclusion. (4) I will consider the main logical tool used by Avicenna in his re-interpretation of metaphysics and his criticism of Platonism: the distinction between plain negation and metathetic affirmation (or negation by equipollence). (5) I will briefly investigate the possible reasons for the peculiar attitude Aquinas adopts towards this issue, with respect to Avicenna’s legacy.

278 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Pasquale Porro

1. Ancient Problems: Aristotle and his Commentators As mentioned above, and as is well-known, Aristotle admits two conflicting requirements for the new science: universality and eminence. That is: the new science must deal with the highest and supreme causes and with the most universal things. This conflict can be easily detected in A or in the first 5 aporiai of B, and explodes, of course, in the transition from C to E : But if there is something which is eternal and immovable and separable (eQ d³ t_ 1stim !ýdiom ja· !j_mgtom ja· wyqist|m), clearly the knowledge of it belongs to a theoretical science, – not, however, to natural science (for natural science deals with certain movable things), nor to mathematics, but to a science prior to both. For natural science deals with things which are inseparable from matter but not immovable [sed cf. Schwegler: peq· wyqist± l³m !kkûoqj !j¸mgta], and some parts of mathematics deal with things which are immovable, but probably not separable, but embodied in matter (peq· !j¸mgta l³m oq wyqist± d³ Usyr !kkû¡r 1m vk,); while the first science deals with things which are both separable and immovable (peq· wyqist± ja· !j¸mgta). … There must, then, be three theoretical philosophies, mathematics, natural science and theology (heokocij¶) … (Arist., Metaph., VI, 1, 1026a10 – 19; transl. Ross).

I shall leave aside all the problems concerning the well-known correction suggested by Schwegler and accepted by many subsequent editors, according to which, in order to preserve the grammatical balance of the sentence, the objects of physics should be characterized, in this passage, as wyqist², just like the object of divine science.5 As a matter of fact, whether we accept Schwegler’s correction or maintain the reading of the manuscripts and of the Bekker edition, the concept of separation, in this crucial text, and in Aristotle generally, is rather unclear, especially if we consider that ‘separation’ (wyqislºr) is precisely the core of Aristotle’s criticism towards Plato’s theory of forms (Metaph., XIII, 9, 1086b6 – 7). Aristotle, for his part, seems to use wyqislºr, wyq_feim and wyqistºr, -^, -|m in four different principal meanings:

5

See Aristotle, Die Metaphysik, ed. Schwegler, pp. 14 – 16; Metaphysics, ed. Ross, ad loc.; Metaphysica, ed. Jaeger, ad loc.; see also Jaeger, Aristoteles, p. 225, n. 1. Against the correction see for instance: Apelt, Beitrge zur Geschichte der griechischen Philosophie, p. 231; Cousin, A Note on the Text of Metaphysics 1026a14, pp. 495 – 6; Trpanier, La philosophie de la nature porte-t-elle sur des spars ou des non-spars?, pp. 206 – 9; Owens, The Doctrine of Being in Aristotelian Metaphysics, p. 296, n. 44; Wundt, Untersuchungen zur Metaphysik des Aristoteles, p. 49; Dcarie, La physique porte-t-elle sur des ‘non-spars’ ( !w¾qista)?, pp. 466 – 8; Halper, One and Many in Aristotle’s Metaphysics, pp. 258 – 9; Zekl, Topos: Die aristotelische Lehre von Raum, p. 85, n. 94; Pietsch, Prinzipienfindung bei Aristoteles, esp. pp. 276 – 7, n. 93; Cleary, Emending Aristotle’s Division of the Theoretical Sciences, pp. 33 – 70; Cleary, Aristotle & Mathematics, esp. pp. 426 – 32.

Immateriality and Separation in Avicenna and Thomas Aquinas 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

– – – –

279

with respect to place (time, quantity); with respect to definition (formula, logos); with respect to matter; as a capacity of independent existence (‘separate’ is what is capable of separate existence).6

If we do not accept Schwegler’s correction, we should read the text quoted above according to the third meaning (according to which ‘separate’ means ‘separate from matter’, and thus immaterial). In this case, physics concerns those things which are material and movable, mathematics those things which are material but immovable, and divine science, or theology, concerns those things which are immaterial and immovable. If we accept Schwegler’s correction, we should read the use of wyqist² in this text, at least insofar as physics is concerned, according to the fourth meaning: separate is what is capable of independent existence. Consequently, physics would concern those beings which possess independent existence but are movable; mathematics would concern those beings which do not possess independent existence and are immovable, while divine science, or theology, would concern those beings which possess independent existence and are immovable. According to this reading, the distinction between the three theoretical sciences is not immediately deduced from the degree of separation from matter, but has mainly to do with different kinds of substantial and accidental being. In other words, we would have here a distinction between these three different modes of being: substantial being in motion; accidental being not in motion; and substantial being not in motion. Immateriality would still play a role in this distinction, but only in an indirect way, since the possibility of motion is always rooted in the potentiality of matter. However, in the famous conclusion of E 1, Aristotle himself draws an explicit connection between the study of universal being and the study of supreme being. This science, so Aristotle writes, is universal because it is primary: One might indeed raise the question whether first philosophy is universal, or deals with one genus, i. e. some one kind of being. […] We answer that if there is no substance other than those which are formed by nature, natural science will be the 6

For the different meanings and textual occurrences in Aristotle see Philippe, *va_qesir, pq|shesir, wyq_feim dans la philosophie d’Aristote, pp. 461 – 79; Berger, ‘Wegnemen’ en ‘scheiden’, pp. 207 – 78; Morrison, Wyqistºr in Aristotle, pp. 89 – 105. See also the debate on the exact meaning of separation in Aristotle between Gail Fine and Donald Morrison, and the subsequent contributions in this regard by Charlotte Witte and Lynne Spellmann: Fine, Separation, pp. 31 – 87; Morrison, Separation in Aristotle’s Metaphysics, pp. 125 – 57; Fine, Separation: A Reply to Morrison, pp. 159 – 65; Morrison, Separation: A Reply to Fine, pp. 167 – 73; Witt, Substance and Essence in Aristotle; Spellmann, Substance and Separation in Aristotle. For a survey, see also my Astrazione e separazione: Tommaso d’Aquino e la tradizione greco-araba, pp. 527 – 80.

280 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Pasquale Porro

first science; but if there is an immovable substance, the science of this must be prior and must be first philosophy, and universal in this way, because it is first (jahºkou ovtyr fti pq¾tg). And it will belong to this to consider being qua being – both what it is and the attributes which belong to it qua being (Arist., Metaph., VI, 1, 1026a23 – 32; transl. Ross).

Now, the main problem in this twofold structure of first philosophy seems to be represented precisely by matter. A science of being qua being cannot exclude matter, since matter is one of the principles which both constitute and explain many substances (as is stated, for instance, in F 3). But in the text of E 1 mentioned above, Aristotle clearly states that theological science concerns what is eternal, immaterial and immovable. It was precisely this contradiction which lead Jaeger to postulate that the final passage of E 1 was a later attempt to reconcile Aristotle’s Urmetaphysik (A, B, C, E 1, L 9 – 10, M) with the Sptmetaphysik (F, G, H), or which lead Natorp to the conviction that – due to this unleidlicher Widerspruch, this unbearable contradiction – E 1 and J must be expunged from the authentic text of the Metaphysics. 7 If the sphere of being qua being includes material beings, we cannot say that first philosophy has only to do with immaterial beings, with those beings which are separate from matter. I shall leave aside the debate amongst contemporary scholars, though I shall briefly consider the position of the ancient commentators. As is well-known, we have three ancient commentaries on the Metaphysics: one by Alexander of Aphrodisias, one by Syrianus and one by Asclepius (i. e. a commentary by Ammonius reported by his pupil Asclepius). But since the commentary attributed to Alexander is authentic only for books A–D, we could add to this list a fourth commentary, that of Pseudo-Alexander, i. e. books E–M included in Hayduck’s edition of Alexander’s commentary. There are very good reasons to suspect, together with Karl Praechter and Concetta Luna, that PseudoAlexander is in actual fact Michael of Ephesus, who was a twelfth century Byzantine commentator.8 Now, only two of these four commentaries include book E and discuss the above-mentioned passage of E 1, and these are the commentaries by Pseudo-Alexander and Asclepius. Their interpretation on this particular issue is more or less uniform: ‘theology’ deals with what is separate (from matter) and immovable.9 7 8 9

See Natorp, Thema und Disposition der aristotelischen Metaphysik, pp. 37 – 65 and 540 – 74; Jaeger, Aristoteles, esp. pp. 170 – 99 (Die Urmetaphysik) and 200 – 36 (Die Entwicklung der Metaphysik). See Praechter, Review of Michaelis Ephesii In libros De partibus animalium, De animalium motione, De animalium incessu, pp. 861 – 907; Luna, Trois tudes sur la tradition des commentaires anciens  la ‘Mtaphysique’, esp. pp. 53 – 71, 197 – 212. See for instance [Ps.-]Alexander Aphrodisiensis, In Metaph., E 1, p. 446, line 35–p. 447, line 3; Asclepius, In Metaph., E 1, p. 360, line 31–p. 361, line 6; Syrianus, In Metaph., C 2, p. 61, lines 17 – 28. We find this very same tri-partition in many Neoplatonic

Immateriality and Separation in Avicenna and Thomas Aquinas 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

281

We might be tempted to discard Pseudo-Alexander, since he is probably a very late commentator, yet his commentary deserves some consideration because it often reports many theses coming from Syrianus.10 As already mentioned, we do not possess Syrianus’ own commentary on E, but we can detect in C a threefold distinction between different kinds of being based on their respective degree of separation from matter and motion. To summarize, we can say that, roughly speaking, the standard interpretation of Aristotle by his ancient commentators relied upon two main theses: i) first philosophy is mainly theology; ii) the subject of this science is obtained through a real separation from matter and motion, while abstraction, i. e. the possibility of conceiving something apart from its real existence in matter, gives us the subject of mathematics.

2. From Boethius to Thomas Aquinas In the Latin West, too, even before Aristotle’s Metaphysics came to be known and read, the choice appears already made. The most significant text concerning the division of the sciences before the introduction of the Greek and Arabic works is undoubtedly one of Boethius’ small theological tracts – the De trinitate. Here Boethius suggests a tripartition of the sciences, which apparently recalls that of Aristotle in E 1, but is actually consistent with the interpretation of the Neoplatonic commentators: Nam cum tres sint speculativae partes, naturalis, in motu inabstracta, !mupena¸qetor (considerat enim corporum formas cum materia, quae a corporibus actu separari non possunt: quae corpora in motu sunt, ut cum terra deorsum ignis sursum fertur, habetque motum j forma materiae coniuncta), mathematica, sine motu inabstracta (haec enim formas corporum speculatur sine materia ac per hoc sine motu: quae formae cum in materia sint, ab his separari non possunt), theologica, sine motu commentaries on the Categories or on the Isagoge: see for instance Ammonius, In Isag., p. 11, line 22–p.12, line 11; Olympiodorus, In Categ., p. 7, lines 30 – 34; Ioannes Philoponus, In Categ., p. 5, lines 1 – 6; Elias, In Isag., p. 27, line 36–p. 28, line 7 and In Categ., p. 115, lines 18 ff.; David, In Isag., p. 57, line 26–p.58, line 25. See also Simplicius, In Phys., p. 1, line 14–p. 2, line 2. On the Neoplatonic interpretation of Aristotle’s Metaphysics see Kremer, Der Metaphysikbegriff in den Aristoteles-Kommentaren der Ammonius-Schule; O’Meara, Le problme de la mtaphysique dans l’antiquit tardive, pp. 3 – 22; Steel, Theology as First Philosophy. The Neoplatonic Concept of Metaphysics, pp. 3 – 21. 10 In addition to the contributions by Luna and O’Meara quoted above [notes 5 and 6] see also Luna, Syrianus dans la tradition exgtique de la Mtaphysique d’Aristote. I. Syrianus entre Alexandre d’Aphrodise et Asclpius, pp. 301 – 309 and D’Ancona, Syrianus dans la tradition exgtique de la Mtaphysique d’Aristote. II. Antcdents et postrit, pp. 311 – 27.

282 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Pasquale Porro

abstracta atque separabilis (nam Dei substantia et materia et motu caret): in naturalibus igitur rationabiliter, in mathematicis disciplinaliter, in divinis intellectualiter versari oportebit …. (Boethius, De trinitate, II, ed. Moreschini, p. 168 – 9, lines 68 – 80).

Interestingly enough, the qualifications are here attributed to the sciences themselves, and not to their objects: thus, natural philosophy is in motion, is not abstract and ‘cannot be removed’; mathematics is not in motion and is not abstract; divine science, or theology, is not in motion, is abstract and separable.11 The denomination here perfectly fits the thing: metaphysics is, strictly speaking, only theology or divine science. But this text too is puzzling. We find !mupena¸qetor instead of the troublesome oq wyqist², but the new term increases, if anything, the confusion, since if its meaning is ‘what exists in matter’ (literally: ‘what cannot be removed or taken away’), it should also be attributed to mathematics, although Boethius does not use it in this case. Furthermore the same term inabstracta means two opposite things, depending on whether it refers to physics or to mathematics: in the first case, it means that the objects of physics are considered together with their matter (considerata cum materia); in the second case, it means that the objects of mathematics are investigated without matter (speculata sine materia). Finally, we can observe that this text does not even mention a possible science of being qua being: there is no room for an ontological science in such a system. Now, the text by Boethius is also crucial because it has been commented by Aquinas (Aquinas was the only great master of the thirteenth century to write commentaries on Boethius’ Opuscula sacra) and, together with the famous prologue to his commentary on Aristotle’s Metaphysics, it is the most significant text in allowing us to understand Aquinas’ position on the issue of the true, or proper, subject of metaphysics. The commentary on the De trinitate is well-known to all readers and scholars of Aquinas due to a textual peculiarity: we possess, at least for part of the work, an autograph version by Aquinas himself, contained in ms. Vat. Lat. 9850, fols 90r–103v. And this manuscript shows a laborious transition from a first redaction, in which Thomas Aquinas still admitted three different kinds of abstraction (corresponding to the three kinds of being mentioned above),12 to a final redaction, in which we find only two degrees of abstraction, while the third degree is replaced by a different operation of the intellect called separatio. It is precisely this distinction between abstraction and separation that is usually taken as one of the most distinctive features of Aquinas’ metaphysics: while 11 On Boethius’ division of sciences see d’Onofrio, La scala ricamata, pp. 11 – 63 (esp. pp. 52 – 9); Speer, The Hidden Heritage: Boethian Metaphysics and its Medieval Tradition, pp. 163 – 81. See also Merlan, From Platonism to Neoplatonism, esp. c. 3. 12 See Thomas de Aquino, Super Boetium De Trinitate, ed. leon., p. 146, app. ad lin. 86.

Immateriality and Separation in Avicenna and Thomas Aquinas 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

283

natural philosophy and mathematics, in order to define their own subject, require two different kinds, or degrees, of abstraction (respectively, the abstraction of the universal from the particular, and the abstraction of form from sensible matter), ‘divine science’, in order to define its own subject, requires a different operation, namely separation; and this consists in the negative judgement by which our intellect states that being is free of any dependency upon matter, not only in thought, but also in reality. Aquinas himself bases this distinction on two different operations of the human intellect: the first is the so-called ‘understanding of indivisibles’ (intelligentia indivisibilium), by which the intellect knows what a thing is (i. e., the quiddity or nature of a given thing); the second is that by which the intellect joins or divides its own contents, by forming affirmative or negative statements. Thus, the first operation concerns the nature itself of a thing, while the second has to do with the being of a thing (esse). Sic ergo intellectus distinguit unum ab altero aliter et aliter secundum diuersas operationes: quia secundum operationem qua componit et diuidit distinguit unum ab alio per hoc quod intelligit unum alii non inesse, in operatione uero qua intelligit quid est unumquodque, distinguit unum ab alio dum intelligit quid est hoc, nichil intelligendo de alio, neque quod sit cum eo, neque quod sit ab eo separatum; unde ista distinctio non proprie habet nomen separationis, set prima tantum. Hec autem distinctio recte dicitur abstractio, set tunc tantum quando ea quorum unum sine altero intelligitur sunt simul secundum rem: non enim dicitur animal a lapide abstrai si animal absque intellectu lapidis intelligatur. Vnde cum abstractio non possit esse proprie loquendo nisi coniunctorum in esse, secundum duos modos coniunctionis predictos, scilicet quo pars et totum uniuntur, uel forma et materia, duplex est abstractio: una qua forma abstraitur a materia, alia qua totum abstraitur a partibus (Thomas de Aquino, Super Boetium De trinitate, q. 5, a. 3, ed. leon., p. 148, lines 159 – 79).

There has been a long and heated debate over the correct interpretation of what separation means and implies in Aquinas. Roughly speaking, the doctrinal focus of the debate could be summarised as follows: does separation imply that we must first prove the effective existence of separate substances (God and the angels) in order to show that being, as the proper subject of first philosophy, can be considered apart from matter? Or rather, is it sufficient, for this same aim, to show that matter, or materiality, is not included in the definition of being qua being? If we support the first thesis (maintained by Geiger, for instance),13 we

13 See Geiger, Abstraction et sparation d’aprs S. Thomas in De trinitate q. 5, a. 3, pp. 3 – 40. But see also, more or less along the same interpretative line, Robert, La mtaphysique, science distincte de toute autre discipline philosophique, pp. 206 – 22; Merlan, Abstraction and Metaphysics in St. Thomas’ Summa, pp. 284 – 91; Moreno, The Nature of Metaphysics, pp. 109 – 35; Owens, Metaphysical Separation in Aquinas, pp. 287 – 306; Nordberg, Abstraction and Separation in the Light of the Historical

284 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Pasquale Porro

emphasise the theological aspect of metaphysics; while, if we endorse the second thesis (suggested by J.F. Wippel and Jan Aertsen, for instance),14 we stress above all the ‘transcendental’ aspect of first philosophy: that is, the proper subject of metaphysics, being qua being, or common being (though we must note that the two expressions do not strictly coincide in Aquinas), can be obtained without a preliminary theological investigation. In other words, we are not obliged to admit God’s existence in order to prove that being can effectively be separate from matter, precisely as in the case of God. But I shan’t deal any longer with this question. Rather, I would like to raise some doubts about the real importance of the doctrine of separatio within Aquinas’ metaphysical project. The way in which Aquinas arrives at the distinction between abstraction and separation is in itself laborious. First of all, we must attend to an oft-neglected detail: the text of the De trinitate used by Aquinas in his commentary presents a slight, yet significant, difference when compared with the text we usually read today. Where Boethius writes that theology is without motion, abstract and separable (sine motu abstracta atque separabilis), the text possessed by Aquinas gives: without motion, abstract and inseparable (sine motu abstracta atque inseparabilis). Thus, in his literal commentary on this section of Boethius’ text, Aquinas is forced to distinguish initially between abstraction and separation (or separability) by referring the first term to the real relation of a thing to matter, and the second to a simple mental relation: Theologia, id est tertia pars speculatiue, que dicitur diuina, uel metaphisica, uel philosophia prima, est sine motu, in quo conuenit cum mathematica et differt a naturali; abstracta, scilicet a materia, atque inseparabilis, per que duo differt a mathematica: res enim diuine sunt secundum esse abstracte a materia et motu, set mathematice inabstracte, sunt autem consideratione separabiles, set res diuine inseparabiles, quia nichil est separabile nisi quod est coniunctum; unde res hdiuinei non sunt secundum considerationem separabiles a materia, set secundum esse abstracte, res uero mathematice e contrario (Thomas de Aquino, Super Boetium De trinitate, hExpositio capituli secundii, ed. leon., p. 134, line 102–p. 135, line 14 [the italics are mine]).

According to this tentative and slightly awkward explanation, mathematical objects are not abstract from matter in existence, while they are separable in thought; divine things, on the contrary, are not separable from matter in Roots of Thomas’ Tripartition of the Theoretical Sciences, pp. 144 – 53; Tavuzzi, Aquinas on Resolution in Metaphysics, pp. 199 – 227. 14 See Wippel, Metaphysics and separatio according to Thomas Aquinas, pp. 431 – 70; Aertsen, Was heißt Metaphysik bei Thomas von Aquin?, pp. 217 – 39; Aertsen, Medieval Philosophy and the Transcendentals. The Case of Thomas Aquinas, esp. pp. 113 – 58. For further bibliographical indications and a survey of the debate see Porro, Metafisica e teologia nella divisione delle scienze speculative del Super Boetium De Trinitate, pp. 467 – 526.

Immateriality and Separation in Avicenna and Thomas Aquinas 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

285

thought, but they are abstract in existence, and therefore are ‘inseparable’, for nothing is separable unless it is conjoined. The use of inseparabilis is interpreted as a kind of lucus a non lucendo, as the grammarians would say: it is impossible, or at any rate meaningless and superfluous, to separate in thought what is already separate in existence. Now, if we compare this passage with the previous one (taken from q. 5, a. 3, which comes later in Aquinas’ commentary), we may note that the terminology has been completely reversed: abstraction can take place only for those things which are conjoined or one in reality (and thus inseparable), while separation is the operation by which we distinguish two things by understanding that one does not exist in the other. In other words, what in the literal commentary is qualified as ‘inseparable’ (divine things) has now become the object of separation. But this is not the only problem. In the first article of the same q. 5, Aquinas specifies that those things which do not depend upon matter for their being, and not only for their being understood, can be subdivided in two classes: there are some beings (or better, some objects of speculation) which never exist in matter, as in the case of God and the angels, and other objects which do not necessarily exist in matter, as in the case of being, substance, quality, potency and act, one and many, and the like. I refer to the same passage quoted at the very beginning of this article: Quedam uero speculabilia sunt que non dependent a materia secundum esse, quia sine materia esse possunt, siue numquam sint in materia, sicut Deus et angelus, siue in quibusdam sint in materia et in quibusdam non, ut substantia, qualitas, ens, potentia, actus, unum et multa, et huiusmodi, de quibus omnibus est theologia, id est scientia diuina (Thomas de Aquino, Super Boetium De trinitate, q. 5, a. 1, ed. leon., p. 138, lines 154 – 61).

The distinction is between a ‘positive’ sense of immateriality (God and other separate substances are positively immaterial), and a ‘negative’ or ‘neutral’ sense of immateriality: that is, matter does not figure in the definition of being qua being, and this means, on the one hand, that being can be conceived independently of or prior to matter and, on the other, that being qua being is in itself neither material nor immaterial, for in the first case we would have only corporeal beings, in the second only separate substances. Both classes of immaterial objects are considered, as is clearly stated in the above passage, by one and the same science – the science we have so far called theology. Thus, according to the distinction mentioned above, we should be able to obtain both classes of immaterial objects through the same operation; in other words, through separation. We must remember, however, that separation distinguishes truthfully only what is not united or conjoined in reality. Therefore it seems to apply, properly speaking, to only one of the classes mentioned above, that is, to the class constituted by God and other separate substances, while the other class consists in those objects which can also be in matter or with matter, though

286 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Pasquale Porro

matter does not figure in their definition. Yet it is precisely the latter indication (i. e. the fact that matter does not belong to the definition of these objects) which shows us how, for this class, abstraction is more suitable than separation. Aquinas explicitly states that we can only abstract what can be conceived independently, whether in reality two things are united or separated. There is no essential dependence between being and matter, and this is precisely what is required, according to Aquinas, in order to abstract something: Si uero unum ab altero non dependeat secundum id quod constituit rationem nature, tunc unum potest ab altero abstrai per intellectum ut sine eo intelligatur non solum si sint separata secundum rem … set etiam si secundum rem coniuncta sint (Thomas de Aquino, Super Boetium De Trinitate, q. 5, a. 3, ed. leon., p. 147, lines 147 – 52 [the italics are mine]).

Matter does not figure in the definition of being qua being and so, it would be tempting to say, being can be conceived independently of matter; that is, it can be abstracted from matter. Yet, as we have noted, Aquinas insists on saying that the subject of divine science, or metaphysics, is obtained through separation and not through abstraction. Finally, things become more complicated because what has so far been considered as belonging to one and the same science suddenly gives rise, in a. 4 of the same q. 5, to two different sciences: Sic ergo theologia siue scientia diuina est duplex: una in qua considerantur res diuine non tamquam subiectum scientie, set tamquam principia subiecti, et talis est theologia quam philosophi prosequntur, que alio nomine metaphisica dicitur; alia uero que ipsas res diuinas considerat propter se ipsas ut subiectum scientie, et hec est theologia que in sacra Scriptura traditur. Vtraque autem est de his que sunt separata a materia et motu secundum esse, set diuersimode, secundum quod dupliciter potest esse aliquid a materia et motu separatum secundum esse: uno modo sic quod de ratione ipsius rei que separata dicitur sit quod nullo modo in materia et motu esse possit, sicut Deus et angeli dicuntur a materia et motu separati; alio modo sic quod non sit de ratione eius quod sit in materia et motu, set possit esse sine materia et motu quamuis quandoque inueniatur in materia et motu, et sic ens et substantia et potentia et actus sunt separata a materia et motu, quia secundum esse a materia et motu non dependent sicut mathematica dependebant, que numquam nisi in materia esse possunt quamuis sine materia sensibili possint intelligi. Theologia ergo philosophica determinat de separatis secundo modo sicut de subiectis, de separatis autem primo modo sicut de principiis subiecti; theologia uero sacre Scripture tractat de separatis primo modo sicut de subiectis, quamuis in ea tractentur aliqua que sunt in materia et motu, secundum quod requirit rerum diuinarum manifestatio (Thomas de Aquino, Super Boetium De Trinitate, q. 5, a. 4, ed. leon., p. 154, lines 175 – 206).

Of course, Aquinas still assumes (as in his later Prologue to his commentary on the Metaphysics) that ‘philosophical theology’ deals with both classes of separate beings (respectively as subject and as principles of its subject), yet, insofar as the two classes of immaterial being now constitute the subjects of two different

Immateriality and Separation in Avicenna and Thomas Aquinas 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

287

sciences, the original tri-partition of Aristotle and Boethius appears absolutely inadequate. I shall not dwell upon the consequences of this reduplication, but I shall focus rather on the procedure followed by Aquinas, a procedure that can be summarised in three main steps, or theses: a) We obtain the subject of theology/divine science through a special operation called separation, which gives us access to real immateriality or transmateriality. b) There are, however, two different kinds, or types, of immateriality: the ‘positive’ immateriality of God and the angels, and the ‘negative’ or ‘neutral’ immateriality of being qua being. c) These two different classes of the immaterial objects of speculation give rise to two different sciences: the theology of Sacred Scripture, on the one hand, and ‘philosophical theology’, or metaphysics, on the other. Now, what is quite amazing in this threefold articulation is that the doctrine of separation (thesis a), which is considered by many scholars as the keystone of Aquinas’ metaphysics, is perhaps the least proper way both to designate the essential feature of metaphysics (philosophical theology) and to justify the third conclusion (thesis c). Indeed, it would have been much easier to distinguish from the beginning the two classes of immaterial objects, instead of associating them both under the operation of separation. One could even say that in order to reach the third conclusion, it would have been better for Aquinas to maintain three degrees of abstraction, as in the original redaction of the autograph text, for the three speculative sciences of the philosophical tradition, and to appeal to separation only for the fourth non-philosophical science – Christian theology. The hypothesis I would like to suggest is that Aquinas is confronted here with a conflict between what he draws from a given tradition (thesis a – insofar as it postulates a distinction between what is abstract and what is separate – and thesis b) and his own conclusion (thesis c), which is that there are two different theologies – a philosophical theology and a ‘theological’ theology, if one could concede a similar tautology. In other words, I assume that the latter thesis – thesis c – is what Aquinas has mainly at heart, and in order to reach it he is ready to quarrel, at least partly, with his own sources. I would like to verify my hypothesis by showing that theses a and b – contrary to the emphasis placed on separation in the field of Thomistic studies – are not original doctrines by Aquinas. And I would like to investigate their true origin by raising two parallel questions. First, is Aquinas really the only one, or the first one, to make use of the distinction between separation and abstraction? And second, is Aquinas really the only one, or the first one, to distinguish between two different kinds of immateriality or transmateriality?

288

3. Avicenna and the Immateriality of Being Luckily enough, the first of these two questions has already received an answer. Claude Lafleur and Joanne Carrier have shown beyond all doubt that the distinction between abstraction and separation was quite common in the writings of the Masters of the Parisian Faculty of the Arts in the first half of the thirteenth century, and especially in the so-called accessus, or introductions, to philosophy composed between 1230 and 1250: notandum est quod entium triplex est genus: quedam sunt entia coniuncta motui et materie sensibili secundum esse et secundum intellectum, abstracta tamen ab indiuiduis, et de talibus est naturalis philosophia; alia sunt que sunt coniuncta motui et materie sensibili secundum esse, abstracta tamen per intellectum, et de talibus est mathematica; alia sunt separata penitus a motu et materia secundum esse et secundum rationem, et de talibus est methaphisica.15

Thus we can at least say that in his commentary on Boethius’ De trinitate Aquinas complies with this common usage. But whence do the Masters of the Arts derive this distinction? Following Alain de Libera, Lafleur and Carrier suggest that the remote source of the distinction between abstraction and separation might be al-Fa¯ra¯bı¯, and they refer more precisely to al-Fa¯ra¯bı¯’s tract De intellectu et intelligibili, in which, however, the distinction is not clearly expressed. Yet, in another text by al-Fa¯ra¯bı¯, his short treatise Maqa¯la … fı¯ ag˙ra¯d ˙ al-hakı¯m fı¯ kull maqa¯la min al-kita¯b al-mawsu¯m bi-l-huru¯f (Treatise … on the ˙ ˙ 16 Goals of the Sage in Each Treatise of the Book Named by Means of Letters), – a text which was not translated into Latin, but the contents of which were nonetheless introduced in the Latin West through Avicenna – is explicitly stated that the objects of metaphysics (or, at least, some of them) ‘have no existence at all (be it imaginary or real) in natural things. It is not that imagination has abstracted them from natural things; rather, their existence and nature [itself ] is abstracted [i.e. immaterial].’17 The terminological distinction between abstraction and separation might thus be the result of the overlapping, in the Latin

˘

15 [Anon.], De communibus artium liberalium, § 204, pp. 190 – 91, lines 980 – 87. See also Lafleur and Carrier, Abstraction, sparation et tripartition de la philosophie thortique: quelques lments de l’arrire-fond farabien et artien de Thomas d’Aquin, pp. 248 – 71. 16 The text has been edited by F. Dieterici in Alfarabi’s philosophische Abhandlungen, pp. 34 – 8 (another anonymous edition has been published in Hyderabad: Da¯ irat alMa a¯rif, 1349H/1930). A complete English translation is now available in Bertolacci, The Reception of Aristotle’s ‘Metaphysics’ in Avicenna’s ‘Kita¯b al-Sˇifa¯ ’, pp. 66 – 72 (but for an analysis of the contents of the treatise and of its influence on Avicenna see the whole c. 3, pp. 65 – 103). A partial English translation was already available in Gutas, Avicenna and the Aristotelian Tradition, pp. 240 – 42. As is well-known, Avicenna refers explicitly to al-Fa¯ra¯bı¯’s treatise in his Autobiography. 17 See Bertolacci, The Reception of Aristotle’s ‘Metaphysics’ in Avicenna’s ‘Kita¯b al-Sˇifa¯ ’, p. 69. ˘

˘

˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Pasquale Porro

289

Immateriality and Separation in Avicenna and Thomas Aquinas

Scholastic vocabulary, of al-Fa¯ra¯bı¯’s idea that some objects are not only abstracted through ‘imagination’, but ‘totally abstracted’ in their existence and nature, and Aristotle’s use of separation in Metaphysics, E 1. Moreover, for the second of the above-mentioned questions (is Aquinas really the only one, or the first one, to distinguish between two different kinds of immateriality, or transmateriality?), I would like to suggest the same genealogy: indeed, even the distinction between two forms of immateriality is an elaboration by Avicenna of a theme mentioned by al-Fa¯ra¯bı¯ in the passage referred to above: Of the subject-matters of this science, on the other hand, some have no existence at all (be it imaginary or real) in natural things. It is not that imagination has abstracted them from natural things; rather, their existence and nature [itself ] is abstracted [i.e. immaterial]. Others exist in natural things, even though they are imagined as abstracted from them. However, they do not exist in natural things essentially, i. e. in such a way that their existence is not independent from these and they are things whose subsistence is due to natural things. Rather, they exist both in natural things and in non-natural things (these latter being separate either really or in imagination). Therefore the science which deserves to be called by this name is [only] the present one. It alone, all other sciences excluded, is ‘metaphysics’.18

Yet, since the Fı¯ ag˙ra¯d remained unknown to the Latins, and Avicenna’s ˙ treatment of these doctrines is much deeper and wider than al-Fa¯ra¯bı¯’s, we can consider directly Avicenna’s Metaphysics (Ila¯hiyya¯t, The Science of Divine Things, i. e. the metaphysical section of The Cure). Briefly, what is the role played by separation in Avicenna’s Metaphysics? In order to arrive at an answer to this question, we should probably turn to book VII of the Ila¯hiyya¯t, because the latter is exactly the place dedicated to an explicit and severe criticism of the Platonic theory of ideas and the Platonic notion of separation. Now, book VII has always been considered relatively eccentric with regard to the overall structure of Avicenna’s Metaphysics, if not completely out of place. After having established, in book I, the true subject of the divine science, and set forth all the prolegomena required for its treatment, the work follows the programme which is announced in I, 4, and which envisages first a consideration of the ‘quasi’-species of the existent, and then a consideration of its properties. With regard to the latter, Avicenna first deals, in book IV, with the pairs prior/posterior and potency/act, and with the concepts of ‘complete’, ‘incomplete’, ‘whole’ and ‘total’; then, in book V, with the notions of universal and particular; and finally, in book VI, with causality. Still according to the

18 See Bertolacci, The Reception of Aristotle’s ‘Metaphysics’ in Avicenna’s ‘Kita¯b al-Sˇifa¯ ’, p. 69. ˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

290

programme expounded in I, 4, Avicenna announces that, at this point, it will be necessary to introduce an analysis of the one and the many. And this is, in actual fact, what we find in book VII. But what puzzles the reader is that this discussion of the one and the many interrupts the transition from the consideration of causes in general to the demonstration of their finiteness, and consequently of the existence of a cause which is absolutely first. It would have been more congruous – so it seems – to place the analysis of the one and the many after that regarding the universal and the particular, and before the section dedicated to causality. Moreover, book VII appears to be rather heterogeneous even in itself. As is well-known, the treatise consists of three chapters, or sections. In the first chapter, Avicenna really deals with the one and the many (or unity and multiplicity); in the other two sections, he discusses and rejects the opinions of the Platonists and Pythagoreans, according to which the principles of all things should be found in ideal forms (mutu¯l) or in numbers (a da¯d). ¯ The second section, in particular, begins with a quite abrupt transition to a different topic: ˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Pasquale Porro

The time has come for us to devote ourselves exclusively to opposing opinions that have been uttered about forms, mathematics, separated principles, and universals that are contrary to our principles which we have established (Avicenna, The Metaphysics of ‘The Healing’, VII, 2, transl. Marmura, p. 243).

Again, such programmes refer more to the contents and aims of book V than to the first section of book VII. And in a way it is Avicenna himself who concedes that this discussion has already been partially raised in the earlier parts of his work: [We will engage in this] even though, in the correctness of what we have said and the rules we have provided, there is, for the discerning, a directing of attention toward the resolution of all their doubts and [toward] showing their falsehood and the contradictions of their doctrines. Nonetheless, we will help by undertaking this ourselves because of what we hope will ensue from this [endeavour] by way of benefits that we will mention in the course of combating them – [benefits] which we may have missed [mentioning] in what we had [previously] presented and explained (Avicenna, The Metaphysics of ‘The Healing’, VII, 2, transl. Marmura, p. 243).

Interestingly enough, Avicenna introduces the issue with a kind of general historiographical remark, which obviously echoes Aristotle’s attitude, but can also be considered as a possible source (though with some important differences) for the famous account of the historical progress of metaphysics given by Thomas Aquinas in his Summa theologiae, I, q. 44, a. 2: Every art has a genesis wherein it is raw and unripe, except that after a while it matures and after some more time, it develops and is perfected. For this reason, philosophy in the early period of the Greek’s occupation with it was rhetorical. It

Immateriality and Separation in Avicenna and Thomas Aquinas 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

291

then became mixed with error and dialectical argument. Of its divisions, it was the natural which first attracted the masses. They then began to give attention to the mathematical [division], then to the metaphysical. They were involved in transitions from one part [of philosophy] to another that were not sound. When they first made the transition from what is apprehended by the senses to what is apprehended by the mind, they became confused. [One] group thought that the division necessitates the existence of two things in each thing – as, for example, two humans in the idea of humanity: a corruptible, sensible human; and an intellectually apprehended, separate, eternal, and changeless human. For each of the two they assigned an existence. They termed the separable existence ‘exemplary existence’, and for each of the natural things they made a separable form that is intellectually apprehended, being the [very] one that the mind receives, since the intelligible is something that does not undergo corruption, whereas every sensible of these [natural things] is corruptible. They [further] rendered the sciences and demonstrative proofs move in the direction of [the incorruptible intelligibles], these being the ones they treat (Avicenna, The Metaphysics of ‘The Healing’, VII, 2, transl. Marmura, pp. 243 – 4).

According to this passage, the first mistake that ancient philosophers made in the transition from sensible to intellectual knowledge was that of considering the intelligibles as separate from sensible things, so that they had to admit two different kinds of existence for each thing. The responsibility for this mistake is clearly and explicitly indicated: It was known that Plato and his teacher, Socrates, went into excess in upholding this view, saying that there belongs to humanity one existing idea in which individuals participate and which continues to exist with their ceasing to exist. This [they held] is not the sensible, multiple, and corruptible meaning and is therefore the intelligible, separable meaning (Avicenna, The Metaphysics of ‘The Healing’, VII, 2, transl. Marmura, p. 244).

Another group, Avicenna continues, attributed a separate existence not to intelligible forms, but to mathematical entities: more precisely, they rendered mathematical entities that are separable in definition (magnitudes, shapes and numbers) as deserving to be separable in existence. Avicenna does not mention any particular thinker in this regard, but it is likely that he is referring to the first Academicians, such as Speusippus and Xenocrates. Others – and Avicenna refers here explicitly to the followers of Pythagoras – took mathematical entities as principles, but did not make them separate. The Pythagoreans composed everything from unity and duality: they made unity within the bounds of good and what is restricted (limited), and duality within the bounds of evil and what is unrestricted (unlimited). Still another group, which remains anonymous, considered ‘excess’, ‘defect’ and ‘equality’ as principles. It is evident that, in its structure, book VII of Avicenna’s Ila¯hiyya¯t corresponds in general to themes which Aristotle deals with in at least six different books of his Metaphysics: A, C, D, I, L, M. As is well-known, a scrupulous and detailed comparison between Aristotle’s Metaphysics and Avicenna’s Ila¯hiyya¯t

292

has been carried out by Amos Bertolacci,19 so that I can confine myself to interpreting his results (as far as book VII is concerned), and to making a couple of additional remarks. 1. First, the references to Aristotle are split into two well-defined blocks: C, D, I for section 1; A, L, M for sections 2 – 3. Thus, even the distribution of Aristotelian quotations shows the composite, not to say heterogeneous, structure of the book, which collects, or refers to, completely different parts and exigencies of Aristotle’s Metaphysics. This lack of homogeneity (in a book which is also relatively short) raises the problem of understanding the real purpose of Avicenna’s treatise – a question which is connected to that of its position, and which we shall postpone, as mentioned above, to our conclusions. 2. Second, it is interesting to note that – despite the fact that sections 2 – 3 are constructed in close reference to Metaphysics A, L, M – when Avicenna comes to describe explicitly the causes of the errors of the Platonists and Pythagoreans, he does not employ the chief arguments elaborated by Aristotle, but suggests five reasons which appear to be completely original, in the sense that they have no parallel in Aristotle’s text. Avicenna hence derives from Aristotle the doctrines he seeks to criticise, and some of the specific objections against the Pythagoreans and Academicians, but the real background of his criticism is rooted in something quite different and original: Avicenna’s doctrine of the indifference of the essences, and above all, his distinction between plain negation and negation by equipollence, or metathesis.20 We can thus move quickly to the consideration of the five roots indicated by Avicenna in order to explain the errors of the Pythagoreans and Academicians (‘If you give thought [to this matter], you will find the bases of the cause of error, in all the things wherein these people have gone astray, to be five [in number]’). We can perhaps confine ourselves to the consideration of the first and fifth ‘roots’, and of the corresponding arguments. The first runs as follows: One of them is their belief that, if a thing is abstracted such that the consideration of another thing is not connected with it, then it is separated from it in existence. It is as though, if attention is paid to the thing alone – [a thing] that has an associate – in a manner that gives no attention to its associate, [this] would render it not adjoining its associate. In short, if it is considered without the condition of [its] conjunction with another, then it is believed that it is considered with the condition that there is no conjunction [with another], so that [according to this view] it was only suitable to be examined because it was not conjoined, but separate. For this reason it was believed that, since the mind attains the intelligibles existing in the world without attending to what is conjoined to them, the mind attains nothing but 19 See Bertolacci, The Reception of Aristotle’s ‘Metaphysics’ in Avicenna’s ‘Kita¯b al-Sˇifa¯ ’, pp. 309 – 73, esp. p. 366. 20 We shall come back soon on the meaning of this distinction; see also below, n. 22.

˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Pasquale Porro

Immateriality and Separation in Avicenna and Thomas Aquinas 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

293

[what is] separate among them. This, however, is not the case. Rather, each thing has one consideration with respect to itself and another consideration with respect to its relatedness to something conjoining it. If we apprehend the form of the human intellectually – for example, inasmuch as it is the form of the human alone – we would have apprehended intellectually an existent alone with respect to its essence. But, inasmuch as we have [so] apprehended it, it does not follow that it alone is separate [in existence]. For that which is mixed with another, inasmuch as it is itself, is inseparable from [the other] by way of negation, not by way of equipollence, in terms of which separation in subsistence is understood. It is not difficult for us to direct attention through perception or some other state to one of the two things whose role is not to separate from its companion in subsistence – even though it separates from it in definition, meaning, and reality, since its reality is not entered in the reality of the other. For conjunction necessitates connectedness, not permeation in meanings (Avicenna, The Metaphysics of ‘The Healing’, VII, 2, transl. Marmura, p. 247).21

The first error of the Platonists was that of considering what is separable by the mind as necessarily separate in being. The error thus consists in confusing two different notions of separation: a purely eidetic separation, and a real ontological separation. But this confusion implies, in turn, a specific logical (and linguistic) mistake, i. e. the confusion between plain negation and negation by equipollence (i. e. metathetic affirmation), or better, between separation by negation and separation by equipollence, or metathesis, as Avicenna himself clearly and explicitly states. 21 See the text of the Avicenna Latinus, Liber de philosophia prima sive Scientia divina, VII, 2, pp. 363 – 4, lines 00 – 26: ‘Tu autem, cum diligenter consideraveris hoc, invenies quod radices occasionis omnis erroris in quem inciderunt isti viri sunt quinque. Una est opinio eorum quod, cum res est exspoliata ab aliquo nec est adiunctus ei respectus alius, profecto exspoliata est in esse ab eo, quemadmodum si id cui aliquid adiunctum est consideraveris per se sine consideratione eius quod sibi adiunctum est, iam enim considerasti illud non adiunctum illi; et omnino, cum consideraveris illud sine condicione coniunctionis, iam putabis te considerasse illud cum j condicione non coniunctionis, ita ut non oporteat considerare illud nisi non coniunctum, quamvis sit coniunctum. Sed, quia intellectus apprehendit intellecta quae sunt in mundo sine consideratione eius cui adiunguntur, ideo putaverunt quod intellectus non apprehendit nisi separata ab eis. Non est autem ita; immo omnis res, secundum quod in seipsa est, habet unum respectum, et, secundum quod coniuncta est alii, habet alium respectum. Nos enim cum intelligimus, verbi gratia, formam hominis inquantum est forma hominis solummodo, iam intelligimus aliquid quod solummodo est secundum quod est in se, sed ex hoc quod intelligimus, non oportet ut sit solum et separatum. Coniunctum enim, ex hoc quod est ipsum, est non separatum secundum modum negationis, non secundum modum privationis qua intelligitur separatio existentiae. Non est autem nobis difficile intelligere per apprehensionem vel per reliquas dispositiones unum ex duobus quorum unum est scilicet quod non est de natura eius separari a sibi coniuncto in existentia, quamvis separetur ab eo in definitione et intentione et certitudine, cum fuerit eius certitudo non contenta intra certitudinem alterius, quoniam esse cum illo facit debere esse coniunctionem non contineri in intentionibus.’ [The italics are mine].

294 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Pasquale Porro

The fifth ‘root of error’ concerns more the separation of mathematical principles than the separation of ideal forms: The fifth [error] is their belief that if material things are caused, then their causes are necessarily any of the things that can separate. For it is not the case that, if material things are caused and mathematical things are separate, then it follows necessarily that the mathematical things are their causes. Rather, [their causes] may be other substances that are not among the nine categories. Nor have they ascertained the core of the truth that the definitions of geometrical [figures] among mathematical [objects] absolutely do not dispense with matter, even though they can do without some kinds of matter (Avicenna, The Metaphysics of ‘The Healing’, VII, 2, transl. Marmura, pp. 248 – 9).

This error presupposes the confusion between abstraction and separation, and in this way it is also connected to the distinction between separation by (plain) negation and separation by metathesis, or equipollence. Moreover, there is no need to admit that the causes of material things should necessarily be separate things, whereas it seems rather awkward that an accidental category (quantity) could be the cause of substances. Finally, Pythagoreans and Academicians do not take into account that mathematical entities are not completely devoid of matter (they lack corporeal matter, but they do possess, in Aristotelian terms, intelligible matter).

4. The Main Logical Tool of Avicenna’s Criticism Against Platonism: the Distinction Between Plain Negation and Negation by Equipollence (Metathetic Affirmation) As we have seen, all five arguments are connected to the doctrine of the indifference of the essences. More precisely, we might add, they refer to the logical tool underlying this doctrine–that is, an accurate analysis of the role which a negation plays in a statement according to its position. In every negative judgement–and this is really a crucial point for Avicenna–the sense changes according to the position of the negation, or (what is really the same thing) according to the purpose (sjopºr) of the negation itself. The negation can refer only to that which is predicated of a thing, or it can refer to the whole predication. In other words (since in classical logic the verb is not usually included in the predicate), the negation can affect either the verb or only the predicate. This distinction is explicitly posed by Avicenna not so much in his Metaphysics, as for instance in a passage of his Remarks and Admonitions (Kita¯b al-Isˇa¯ra¯t wa l-tanbı¯ha¯t), which it might be useful to quote:

Immateriality and Separation in Avicenna and Thomas Aquinas

Remark: Concerning equipollence and positiveness. Sometimes the composition consists of a negative particle with another [term], as in the statement, ‘Zayd is non-sighted.’ By ‘non-sighted’ we intend ‘blind’, or a concept more general than that. In short, if ‘non’ is made as one thing with ‘sighted’, or with what resembles it, and, is, then, affirmed or negated, ‘non’ – as well as any other negative particle [in its place] – is, then, a part of the predicate. Thus if you affirm the whole, that would be an affirmation. And if you negate it, that would be a negation, as when you say, ‘Zayd is not non-sighted.’ […] [In Arabic], if a negative particle precedes the copula, as in the statement, ‘Zayd laisa huwa bas¯ıran’ (Zayd is not sighted), the negation has been applied to the ˙ eliminating it and negating it. But if the copula precedes the affirmation, thus negative particle, it makes a part of the predicate: And the proposition is an affirmation as in the statement, ‘Zayd huwa g˙ayr bas¯ır’ (Zayd is non-sighted) ˙ (Avicenna, Remarks and Admonitions. Part One: Logic, transl. Inati, pp. 83 – 4).

The passage refers to the doctrine of propositions (relying at least indirectly on Aristotle, De int., 10, 19b27 ff.; Anal. Pr., I, 46, 51b36 ff., and perhaps on alFa¯ra¯bı¯)22, and in particular to the composite and divided sense of a negation. As already mentioned, Avicenna makes a distinction between plain or simple negation and negation by equipollence (or perhaps, better, metathesis). Plain

˘

22 On the logical meaning of metathesis in al-Fa¯ra¯bı¯ and its possible Greek sources (presumably, Theophrastus) see al-Fa¯ra¯bı¯, Commentary and Short Treatise on Aristotle’s ‘De interpretatione’, ed. F.W. Zimmermann, esp. p. 98: ‘In languages, however, in which a noun is not predicated of a noun unless they are expressly connected by a hyparctic verb, negation is distinct from metathesis. For if a speaker attaches the negative particle to the hyparctic verb the result is a plain hnegationi; and if he attaches it to the predicate noun instead, the result is like Zaydun mawju¯dun la¯ a¯dilan, which is a metathetic affirmation.’ See also Zimmermann’s comments in this regards, esp. p. lxiii: ‘The bulk of al-Fa¯ra¯bı¯’s exposition of the bewilderingly ramified body of doctrine surrounding Aristotle’s cryptic remarks at the beginning of section three (ch. 10, 19b19 ff.) conspicuously resembles what may tentatively be called Theophrastus’ theory of metathesis, because Ammonius’ account … would suggest that to Theophrastus belonged not only the term “metathesis” for the shift from “… is P” to “… is not not-P” (or simply from “P” to “not-P”) but also the broader framework of ideas of which it forms a part’. The use of ‘metathesis’ in this particular context is explained by Zimmermann himself in a footnote on the same page, on the basis of Ammonius: ‘ “In arranging the propositions in a diagram …, once the indefinite negation has been placed underneath the plain affirmation, there is nothing left but to place the indefinite affirmation underneath the plain negation. And this is why Theophrastus has called them [sc. the pair with an indefinite predicate] “metathetic” [1j letah]seyr], for their order is reversed [letat]heitai] in the diagram – or else because the definite is replaced [letatehe_r] by an indefinite predicate” (Ammonius 161.24 – 32; cf. Stephanus 40.22 – 5, where the alternative explanation is that the negative particle is transferred [letat_hetai] from the copula to the predicate). Theophrastus’ 1j letah]seyr entered the Arabic tradition as ma du¯l [deviated, deflected], which in deference to the original I have rendered as “metathetic”. The opposite of ma du¯l is bası¯t “plain” [tr. "pkoOr].’ On single and composite negative expressions in Avicenna see˙ also S. Inati, Ibn Sina on Single Expressions, pp. 148 – 59. ˘

˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

295

296

negation denies the verb, and thus renders a proposition negative, as in the case: ‘Zayd is not sighted’; whereas negation by metathesis ( udu¯l: equipollence, according to Inati’s and Marmura’s translations) is that negation which denies the predicate and, in this sense, is equivalent to a (metathetic) affirmation such as ‘Zayd is non-sighted’.23 This logical tool – that is, the distinction between plain negation and negation by equipollence (or metathetic affirmation) – is employed with great consistency and efficacy throughout the whole system of Avicenna’s Metaphysics, and represents, in my opinion, one of its most profound, original and innovative features. I shall confine myself here to pointing out just three cases in which negation by equipollence, which implies a position and hence an ontological separation, is contrasted with plain negation, which implies only a form of eidetic separation. 1. The first basic example concerns the determination of the subject of metaphysics: the existent, or being, that represents the subject of metaphysics is not im-material in the sense of ‘on the condition of not being material’, but rather in the sense of ‘not on the condition of being material’, in other words, without all the concomitant conditions that accompany it, but that do not enter into the definition of the existent itself: ‘The primary subject matter of this science is, hence, the existent inasmuch as it is an existent; and the things sought after in [this science] are those that accompany [the existent], inasmuch as it is an existent, unconditionally.’ (I, 2; transl. Marmura, p. 10).24 As is wellknown, Avicenna enumerates many different senses of immateriality, while always maintaining a clear distinction between immateriality by equipollence or metathesis (separation which posits something, hence ‘positive’ immateriality – the immateriality which belongs to God and the separate substances) and immateriality by plain or simple negation, which only implies an eidetic separation (matter is not included in the intention of being qua being). 2. A second example is offered by the very same doctrine of the indifference of the essences: the indifferent essence is not a separate entity (a third entity with respect to the concept in the mind or the individuals in the physical world); it is indifferent only because it can be considered independently of the conditions that always, unavoidably, accompany it. Here again the distinction is ˘

23 Likewise, in his Kita¯b al-Nagˇat, Avicenna explains that the difference between a ‘plain’ proposition (bası¯ta) and a metathetic, or equivalent, proposition (ma du¯la) lies in the fact ˙ the negation is part of the predicate, so that, for instance, ‘nonthat, in the second, seeing’ or ‘non-sighted’ is equivalent to ‘blind’; see Ibn Sı¯na¯, Kita¯b al-Nagˇat, ed. M. Fahrı¯, pp. 54 – 5. ˘ Latinus, Liber de philosophia prima sive Scientia divina, I, 2, p. 13, lines 36 – 8: 24 Avicenna ‘Ideo primum subiectum huius scientiae est ens, inquantum est ens; et ea quae inquirit sunt consequentia ens, inquantum est ens, sine condicione.’ (The English translation in the text is that of Marmura). ˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Pasquale Porro

Immateriality and Separation in Avicenna and Thomas Aquinas 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

297

between essence taken on the condition of not being associated with its concomitants (the ontological separation of Platonic forms) and essence taken not on the condition of being associated with its concomitants (eidetic separation). In the Avicenna Latinus (i. e. in the Latin version of the Ila¯hiyya¯t), this doctrine is expressed through an extremely precise and technical phrasing: to take something cum condicione non rei alterius, ‘on the condition of no other thing’, is quite different from taking the same thing non cum condicione rei alterius, ‘not on the condition of any other thing’. The first formula (to take something ‘on the condition that no other thing is added to it’) leads to a separate existence; the second formula (to take something ‘not on the condition of a possible addition’) leads only to the possibility of considering something apart from the different and even innumerable conditions which might accompany it in reality.25 This is precisely what we find in the explication of the first reason for the error of the Platonists: ‘In short, if it is considered without the condition of [its] conjunction with another, then it is believed that it is considered with the condition that there is no conjunction [with another], so that [according to this view] it was only suitable to be examined because it was not conjoined, but separate.’ (VII, 2; transl. Marmura, p. 247). 3. Finally – and this is the only case in which, according to Avicenna, it is legitimate to appeal to negation by equipollence, or ontological separation – the same distinction is employed in book VIII to point out the difference between God (the Necessary Existent) and common being: The First, hence, has no quiddity. Those things possessing quiddity have existence emanate on them from Him. He is pure existence with the condition of negating privation and all other description of Him. … The meaning of my statement, ‘He is 25 Avicenna Latinus, Liber de philosophia prima sive Scientia divina, V, 1, p. 236, line 92–p. 7, line 10: ‘Hic est autem quiddam quod debet intelligi, scilicet quia verum est dicere quod de animali, ex hoc quod est animal, non debet praedicari proprietas nec communitas, nec est verum dicere quod de animali, ex hoc quod est animal, debet non praedicari proprietas vel communitas; scilicet nam si animalitas faceret debere non praedicari de eo proprietatem vel communitatem, tunc nec esset animal proprium nec esset animal commune; et secundum hoc debes intelligere magnam esse distantiam inter illa, et ob hoc etiam interest an dicatur quod animal, ex hoc quod est animal per se, sine condicione alterius, et an dicatur quod animal, ex hoc quod est animal per se, cum condicione non rei alterius. Si enim concederetur quod animal, ex hoc quod est animal per se, esset cum condicione quod non haberet esse in sensibilibus istis, non tamen concederetur quod platonitas esset in sensibilibus istis; esse enim animalis cum condicione non rei alterius in intellectu tantum est; animal vero per se, non cum condicione rei alterius, habet esse in sensibilibus. Ipsum vero in se in veritate sua est sine condicione alterius rei, quamvis sit cum mille condicionibus quae adiunguntur ei extrinsecus. Animal ergo per se ex sua animalitate habet esse in istis sensibilibus; hoc autem non facit debere ipsum esse separatum per se.’ [The italics are mine].

298 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Pasquale Porro

pure existence with the condition of negating all other additional [attributes] of Him,’ is not that this is the absolute existence in which there is participation [by others]. If there is an existent with this description, it would not be the pure existent with the condition of negation, but the existent without the condition of positive affirmation. I mean, regarding the First, that He is the existent with the condition that there is no additional composition, whereas this other is the existent without the condition of [this] addition (Avicenna, The Metaphysics of ‘The Healing’, VIII, 4, transl. Marmura, pp. 276 – 7).

The passage concerns how we should conceive divine simplicity in comparison with that of common being, yet it also implies, very clearly and precisely, the difference between a ‘positive’ and a ‘negative’ meaning of immateriality (or separation). The Avicenna Latinus has a very interesting formulation here: God is ‘naked being not on the condition of affirming [something of it]’ – esse exspoliatum non condicione affirmandi – in other words, God is pure or indeterminate being ‘on the condition of not adding a composition’ – cum condicione non addendi compositionem. Common being is indeterminate being ‘on the condition of denying [something of it]’ – condicione negandi – in other words, ‘not on the condition of being considered together with those dispositions that accompany it’ – non condicione additionis. 26 This text has many parallel passages in Thomas Aquinas.27 Just to mention one example: dicendum quod aliquid cui non fit additio potest intelligi dupliciter. Uno modo, ut de ratione eius sit quod non fiat ei additio; sicut de ratione animalis irrationalis est, ut sit sine ratione. Alio modo intelligitur aliquid cui non fit additio, quia non est de ratione eius quod sibi fiat additio: sicut animal commune est sine ratione, quia non est de ratione animalis communis ut habeat rationem; sed nec de ratione eius est ut careat ratione. Primo igitur modo, esse sine additione, est esse divinum: secundo modo, esse sine additione, est esse commune (Thomas de Aquino, Summa theol., I, q. 3, a. 4, ad 1).

To sum up, Avicenna seems to employ plain or simple negation to indicate the immateriality of being (example 1) and the indifference of the essences (example 26 Avicenna Latinus, Liber de philosophia prima sive Scientia divina, VIII, 4, p. 402, lines 48 – 60: ‘Primus igitur non habet quidditatem, sed super habentia quidditates fluit esse ab eo; ipse igitur est esse exspoliatum, condicione negandi privationes et ceteras proprietates ab eo. Deinde cetera alia quae habent quidditates sunt possibilia, quia habent esse per ipsum. Intentio autem de hoc quod dicimus quod ipse est esse exspoliatum condicione negandi ceteras additiones ab eo, non est quod ipse sit esse exspoliatum in quo communicet aliquid aliud esse, si fuerit esse cuius haec sit proprietas: ipse enim non est illud ens exspoliatum condicione negandi, sed est ens non condicione affirmandi, scilicet de primo, quod est ens cum condicione non addendi compositionem, sed hoc aliud est ens non condicione additionis, et, quia illud fuit universale quod praedicatur de omni re, istud vero non praedicatur de eo in quo est additio, ideo in omni quod est praeter illud est additio.’ 27 See for instance Thomas de Aquino, De potentia, q. 7, a. 2, ad 4 and ad 6; C. gent. I, c. 26.

Immateriality and Separation in Avicenna and Thomas Aquinas 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

299

2), which represent two cases of eidetic separation, and negation by equipollence or metathesis to indicate the immateriality of God and the separate substances (example 3), which represents a case of ‘positive’, ontological separation. The distinction adopted by the Latin Scholastic masters between abstraction and separation has thus conceptually (though not linguistically) a specific origin: the anti-Platonic polemic expounded by Avicenna in his theory of the indifference of the essences, especially in the seventh book of his Metaphysics. Now, that this is the actual doctrinal context and the actual function of the distinction is indirectly, and surprisingly, confirmed by Aquinas himself in the conclusive passage of the responsio of q. 5, a. 3 of his commentary on Boethius’ De trinitate – a passage which interpreters often neglect to consider: Et quia quidam non intellexerunt differentiam duarum ultimarum [scil. distinctionum] a prima, inciderunt in errorem ut ponerent mathematica et uniuersalia a sensibilibus separata, ut Pittagoras et Platonici (Thomas de Aquino, Super Boetium De Trinitate, q. 5, a. 3, ed. leon., p. 149, lines 287 – 90).

This remark is indeed crucial, and not only because it proves that Avicenna’s Metaphysics is the real source of the distinction between abstraction and separation (the three distinctions mentioned in the text). What scholars often fail to point out is that the context in which Aquinas proposes this distinction has to do with the epistemological status of mathematics, and not of metaphysics or theology. Therefore, the real and primary aim of the distinction is not that of affirming that some beings (God and the angels) are absolutely separate from matter (which is rather obvious), but, the exact opposite, that of denying that other beings (i. e. mathematical objects and universals) are effectively separate, as happens in all forms of Platonism or ontological realism.

5. Thomas Aquinas and the Rethinking of Avicenna’s Legacy Thus, the first two of Aquinas’ theses mentioned above – a and b – are attested in Avicenna. Yet, we immediately detect a clear difference between Aquinas and Avicenna: the distinction between two classes of immaterial beings does not imply, in Avicenna, any kind of split within the field of divine science; on the contrary, in Avicenna’s Metaphysics we find a gradual transition from what is taken ‘not on the condition of being material’ to what can be taken only ‘on the condition of not being material’. Or, as we can rephrase it: we find a gradual transition from what is abstract from matter to what is effectively separate. This is possible, however, only because Avicenna assumes that the proper subject of metaphysics – being qua being – is prior to the division between cause and effect, or between God and creatures. For Avicenna, God is not the cause of the subject of first philosophy; rather, God is part of the subject of metaphysics

300 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Pasquale Porro

(being qua being), and He is the cause of another part of the same subject (created being). For Aquinas, on the contrary, God enters philosophical theology, or metaphysics, as principle and cause of the subject of this science, which in Aquinas’ view is common being qua created being. We can now return to our initial question: why did Aquinas employ separation in order to define metaphysics? The final matter to which we need to attend is the fact that in Aquinas’ later commentary on the Metaphysics the distinction between abstraction and separation seems to be weakened, and the adjective separata is also used to refer to the objects of mathematics: Sed tamen hoc est manifestum, quod scientia mathematica speculatur quaedam inquantum sunt immobilia et inquantum sunt separata a materia sensibili, licet secundum esse non sint immobilia vel separabilia. … In hoc ergo differt mathematica a physica, quia physica considerat ea quorum definitiones sunt cum materia sensibili. Et ideo considerat non separata, inquantum sunt non separata. Mathematica vero considerat ea, quorum definitiones sunt sine materia sensibili. Et ideo, etsi sunt non separata ea quae considerat, tamen considerat ea inquantum sunt separata. … Physica enim est circa inseparabilia et mobilia, et mathematica quaedam circa [Al. circa quaedam] immobilia, quae tamen non sunt separata a materia secundum esse, sed solum secundum rationem, secundum vero esse sunt in materia sensibili (Thomas de Aquino, In Metaph., VI, lect. 1, eds Cathala, Spiazzi, p. 297, §§ 1161 and 1163 [The italics are mine]).

Moreover, in the Prologue to his commentary, Aquinas refers implicitly to different degrees of separation (and not of abstraction): the expression maxime separata implies that there might be beings more or less separated (though in connection with physics and mathematics the verb abstrahhio is still used). How should we read this attenuation? I believe that an answer can be given on two different levels. On the one hand, we could assume that Aquinas no longer uses the term ‘separation’ in its narrow, or strict, sense precisely because he realises progressively that separation is inessential, or superfluous, to define a science which has to do with two different classes of immaterial being and which, moreover, finds its proper subject, strictly speaking, only in what is ‘negatively’ immaterial. On the other hand, while commenting on Aristotle, Aquinas has no need to distinguish between two theologies, but can confine himself to the level of mere natural, or philosophical, theology. The case of the commentary on Boethius’ De trinitate is quite different for at least two reasons: 1. the De trinitate is a theological, not a philosophical, treatise; 2. the text is not just in this case a literal commentary, but also a commentary per modum quaestionis, which gives Aquinas the opportunity to discuss the issues he prefers. In this context, Aquinas has a very precise aim: that of making room for a new science in the system of theoretical sciences – theologia nostra or theologia sacrae scripturae. If this project entails some ambiguities, it is because Aquinas avails himself of Avicennian doctrines in order

Immateriality and Separation in Avicenna and Thomas Aquinas 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

301

to reach a conclusion that has nothing to do with Avicenna. Actually, Aquinas is well aware of the real solution adopted by Avicenna – that of applying to metaphysics a procedure very similar to that used for mathematics: Advertendum est autem, quod licet ad considerationem primae philosophiae pertineant ea quae sunt separata secundum esse et rationem a materia et motu, non tamen solum ea; sed etiam de sensibilibus, inquantum sunt entia, Philosophus perscrutatur. Nisi forte dicamus, ut Avicenna dicit, quod huiusmodi communia de quibus haec scientia perscrutatur, dicuntur separata secundum esse, non quia semper sint sine materia; sed quia non de necessitate habent esse in materia, sicut mathematica (Thomas de Aquino, In Metaph., VI, lect. 1, eds Cathala and Spiazzi, p. 298, § 1165).

Yet the fact is that Aquinas needs separation, in its strict sense, not in order to define the subject of metaphysics as philosophical theology, but, outside the latter, in order to define the subject of a new science. Consequently, Aquinas draws from Avicenna (and from the Masters of the Arts) the notion of separation, but transposes it to a different place (moreover, to a place Aquinas himself recognises as improper): not the context of the confutation of Platonism, but that of the discussion of the subject of metaphysics. Aquinas thus applies to the case of metaphysics as a whole what was originally destined only for its theological part, and what he finally applies to a science other than philosophy. The result is somehow paradoxical: by postulating two different sciences which both concern what is separate, Aquinas divides what in Avicenna was united (metaphysics as a single science that, starting from the negatively immaterial, comes to demonstrate the positively immaterial), and unites what in Avicenna must remain distinct (separation and abstraction: the former is that which is ‘on the condition of no addition’; the latter, that which is ‘not on the condition of an addition’). The stake of this doctrinal game is, in my view, clear: Aquinas intends to broaden and force a traditional tripartition which was by then inadequate to contain Christian theology as a distinct speculative science. To summarise: by seeking to distinguish separation from abstraction, Aquinas upsets the delicate balance obtained by Avicenna (from al-Fa¯ra¯bı¯) in his rethinking of Aristotle’s Metaphysics; this conscious choice by Aquinas implies some ambiguities, as I have tried to point out, but it also offers some undeniable advantages: it allows theology to constitute itself as a science independent of philosophy, while at the same time it allows metaphysics to remain ‘mere’ natural theology, without any undue or immediate admixture with revealed doctrines. Thus, there would be no need to make room in metaphysics, as a philosophical science of being qua being, for an eschatology, a prophetology, nor for a theory of sexual ethics according to revealed laws, as we find in the last two books of Avicenna’s Metaphysics.

302

Yet in order to obtain these results, Aquinas loses along the way one of the most sophisticated tools elaborated by Avicenna in order to make sense of Aristotle’s Metaphysics: that is, the distinction between plain negation and negation by metathesis; in other words, the distinction between ontological separation and what we may call eidetic, or intentional, separation. On the other hand, the issue of separation also casts a different light on the position and purpose of book VII within Avicenna’s Metaphysics – a position which has always puzzled his interpreters and readers, for it interrupts the transition from the general consideration of causes to the consideration of the First Cause (the only real ontological cause). A possible solution has been suggested by Amos Bertolacci in his rethinking of the structure of Avicenna’s Metaphysics. According to Bertolacci, ontology in Avicenna’s Ila¯hiyya¯t always intersects with henology, i. e. with a discussion of species and the properties of the one and the many. In other words, the concrete structure of Avicenna’s Metaphysics is a result of the coexistence and interaction of two axes, one fundamental, the other complementary. The main axis concerns the existent, or being, from four different perspectives: its role as the subject of metaphysics, its species, its properties and its causes. The second axis regards the one and the many, whose species and properties are also taken into account; accordingly, it encompasses a henology of species and a henology of properties. It might be useful to quote Bertolacci himself here: ‘the core of metaphysics is the result of two vertical axes (dealing respectively with the existent, and the one and the many), intersected by four horizontal lines (subject-matter, species, properties, causes). In the actual structure of Avicenna’s Metaphysics these two axes coexist and partially overlap’.28 This overlapping concerns especially the treatment of the species of the existent and the treatment of the species of one and many in book III. Book VII, on the contrary, should not be considered as a case of overlapping, since it would correspond to the section devoted to the henology of properties. In this sense, chapters 2 – 3 of book VII represent, for Bertolacci, ‘a sort of complement to the proof of the accidentality of quantity provided in chapters III, 3 – 5’, and therefore ‘a sort of pars destruens of the treatment of “one” and “many”’.29 I think that this is fundamentally correct, and yet it does not fully explain why book VII starts with a discussion of the one and the many, and their properties, and then turns into a more radical criticism of the notion of separation (also independently of quantity). One might say that, when speaking about numbers, it makes perfect sense to show that they do not possess a 28 See Bertolacci, The Reception of Aristotle’s ‘Metaphysics’ in Avicenna’s ‘Kita¯b al-Sˇifa¯ ’, p. 209. 29 See Bertolacci, The Reception of Aristotle’s ‘Metaphysics’ in Avicenna’s ‘Kita¯b al-Sˇifa¯ ’, p. 178. See also Marmura, Avicenna’s Critique of Platonists in Book VII, Chapter 2 of the Metaphysics of his Healing, pp. 355 – 70. ˘ ˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Pasquale Porro

Immateriality and Separation in Avicenna and Thomas Aquinas 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

303

separate existence. This is true, but why then also discuss here Plato’s theory of ideas? We would have expected such a criticism in book V. Well, my hypothesis is that this discussion is neither casual nor accidental. On the contrary, what Avicenna needs in the transition from the general account of causality to the theological part of his Metaphysics is precisely a clear distinction between separation by (plain) negation and separation by metathesis (or equipollence), i.e. – as we may rephrase it – between eidetic separation and ontological separation. In other words, in order to approach the theological part, Avicenna has to take leave, once and for all, of Platonism: it is only by pointing out the ‘roots’ of the errors of the Platonists (and Pythagoreans) that Avicenna can finally move from what is separate by plain negation to what is separate by equipollence, thus establishing a consistent route from the subject of metaphysics to its end, or purpose. Book VII is therefore certainly a treatise on the one and the many, but at the same time it is also a treatise on separation. The latter is absolutely essential in order to clarify both that metaphysics is a science of that which is separate (as Aristotle states in E 1) and that ‘separate’ may be understood in two different ways: separate is both that which is taken on the condition of not being associated with matter, and that which is not taken on the condition of being associated with matter. Without this distinction, the unity of first philosophy as onto-theology remains uncertain and problematic. Instead of being out of place or marginal, book VII would thus serve as the essential link in keeping together that which in Aristotle (or, at least, in Aristotle’s Metaphysics as we know it) remained merely juxtaposed: metaphysics as a science of being qua being and metaphysics as a divine science dealing with immovable and separate realities.

Bibliography Primary Sources Alexander Aphrodisiensis, In Aristotelis Metaphysica commentaria, CAG I, ed. M. Hayduck, Berlin: Reimer, 1891. Ammonius, In Porphyrii Isagogen sive V voces [= In Isag.], CAG IV/3, ed. A. Busse, Berlin: Reimer, 1891. –– , In Aristotelis De interpretatione commentarius, CAG IV/5, ed. A. Busse, Berlin: Reimer, 1897. [Anon.], De communibus artium liberalium, § 204, eds C. Lafleur, J. Carrier, in: Un instrument de rvision destin aux candidats  la licence de la Facult des arts de Paris, le “De communibus artium liberalium” (vers 1250?), Documenti e studi sulla tradizione filosofica medievale, 5, 1994, pp. 129 – 203. Aristotle, Opera, ed. I. Bekker (Academia Regia Borussica), Berlin: Reimer, 1831 – 70, 5 vols; repr. Berlin: De Gruyter, 1960 – 61.

304

˘

˘

˘

–– , Die Metaphysik [= Metaph.], ed. A. Schwegler, Tbingen: Fues, 1847 – 48, repr. Frankfurt a. M.: Minerva, 1960 and 1968. –– , Metaphysics [= Metaph.], ed. W.D. Ross, Oxford: Clarendon Press, 1924. –– , Metaphysica [= Metaph.], ed. W. Jaeger, Oxford: Clarendon Press, 1957. –– , Metaphysics [= Metaph.], transl. W.D. Ross, in The Complete Works of Aristotle. The Revised Oxford Translation, ed. J. Barnes, Princeton: University Press, 1995 (6th printing, with corrections). Asclepius, In Aristotelis Metaphysicorum libros A–F commentaria [= In Metaph.], CAG VI/2, ed. M. Hayduck, Berlin: Reimer, 1888. Avicenna [Ibn Sı¯na¯], The Metaphysics of ‘The Healing’, transl. and ed. M.E. Marmura, Provo, Utah: Brigham Young University Press, 2005. –– , Kita¯b al-Nagˇat, ed. M. Fahrı¯, Beirut: Da¯r al-afa¯q al-gˇadı¯da, 1985. –– , Remarks and Admonitions.˘ Part One: Logic, transl. and ed. S.C. Inati, Toronto: Pontifical Institute of Mediaeval Studies, 1984. –– , Liber de philosophia prima sive Scientia divina, I–IV, ed. S. Van Riet, Louvain/ Leiden: Peeters/Brill, 1977. –– , Liber de philosophia prima sive Scientia divina, V–X, ed. S. Van Riet, Louvain/ Leiden: Peeters/Brill, 1980. Boethius, De trinitate, II, ed. C. Moreschini, in id., De consolatione philosophiae – Opuscula theologica, Mnchen/Leipzig: K.G. Saur, 2000. David, Prolegomena et in Porphyrii Isagogen commentarium [= In Isag.], CAG XVIII/2, ed. A. Busse, Berlin: Reimer 1904. Elias, In Porphyrii Isagogen et Aristotelis Categorias commentaria [= In Isag., In Categ.], CAG XVIII/1, ed. A. Busse, Berlin: Reimer 1900. al-Fa¯ra¯bı¯, Commentary and Short Treatise on Aristotle’s ‘De interpretatione’, transl. and ed. F.W. Zimmermann, London: Oxford University Press, 1981. –– , Maqa¯la […] fı¯ ag˙ra¯d al-hakı¯m fı¯ kull maqa¯la min al-kita¯b al-mawsu¯m bi-l-huru¯f, in ˙ ˙ Abhandlungen, ed. F. Dieterici, Leiden: Brill,˙ 1890, Alfarabi’s philosophische pp. 34 – 8. –– , Maqa¯la fı¯ ag˙ra¯d ma¯ ba da l-tabı¯ a, anonymous edition, Hyderabad: Da¯ irat al˙ ˙ Ma a¯rif, 1349H/1930. Ioannes Philoponus, In Aristotelis Categorias commentarium, CAG XIII/1, ed. A. Busse, Berlin: Reimer 1898. Olympiodorus, Prolegomena et in Categorias commentarium [= In Categ.], CAG XII/1, ed. A. Busse, Berlin: Reimer 1902. Simplicius, In Aristotelis Physicorum libros quattuor priores commentaria [= In Phys.], CAG IX, ed. H. Diels, Berlin: Reimer 1882. Stephanus, In librum Aristotelis De interpretatione commentarium, ed. M. Hayduck, Berlin: Reimer, 1885. Syrianus, In Metaphysica commentaria [= In Metaph.], CAG VI/1, ed. W. Kroll, Berlin: Reimer, 1902. Thomas de Aquino, Commentary on Aristotle’s ‘Physics’, transls and eds R. Blackwell, R. Spath, W. Thirlkel, Notre Dame, Indiana: Dumb Ox Books: 1999. –– , Commentary on Aristotle’s ‘Metaphysics’, transl. and ed. J. Rowan, Notre Dame, Indiana: Dumb Ox Books, 1995. –– , In duodecim libros Metaphysicorum Aristotelis Expositio [= In Metaph.], eds M.–R. Cathala, R. Spiazzi, Turin/Rome: Marietti, 1950. –– , In octo libros Physicorum Aristotelis Expositio [= In Phys.], ed. P. Maggiolo, Turin/ Rome: Marietti, 1965. ˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Pasquale Porro

Immateriality and Separation in Avicenna and Thomas Aquinas

–– , Liber de veritate catholicae Fidei contra errores infidelium seu Summa contra gentiles, eds P. Marc, C. Pera, P. Caramello, t. I: introductio; t. II–III: textus, Turin/Rome: Marietti, 1961 – 67. –– , Quaestiones disputatae de potentia, ed. P.M. Pession, in Quaestiones disputatae, Turin/ Rome: Marietti, 196510, t. 2, pp. 1 – 276. –– , Summa contra gentiles, eds Fratres Praedicatorum [= ed. leon.], Sancti Thomae Aquinatis opera omnia, vols 13 – 15, Rome: Commissio Leonina, 1918/1926/1930. –– , Summa theologiae [= Summa theol.], ed. P. Caramello [cum textu ex recensione leonina], Turin/Rome: Marietti, 1948. –– , Super Boetium De trinitate, eds Fratres Praedicatorum [= ed. leon.], Sancti Thomae Aquinatis Opera Omnia, vol. 50, Rome/Paris: Commissio Leonina/ditions du Cerf, 1992.

Secondary Sources J.A. Aertsen, Medieval Philosophy and the Transcendentals. The Case of Thomas Aquinas, Studien und Texte zur Geistesgeschichte des Mittelalters 52, Leiden/New York/ Cologne: Brill, 1996. –– , Was heißt Metaphysik bei Thomas von Aquin?, in Scientia und ars in Hoch- und Sptmittelalter, eds I. Craemer-Ruegenberg, A. Speer, Miscellanea Mediaevalia 22/1, Berlin/New York: De Gruyter, 1994, pp. 217 – 39. O. Apelt, Beitrge zur Geschichte der griechischen Philosophie, Leipzig: Teubner, 1891, reprinted Aalen: Scientia, 1975. H.H. Berger, ‘Wegnemen’ en ‘scheiden’: Aristoteles’ verdeling van den theoretische wetenschappen, Tijdschrift voor Filosofie, 25, 1963, pp. 207 – 78. A. Bertolacci, The Reception of Aristotle’s ‘Metaphysics’ in Avicenna’s ‘Kita¯b al-Sˇifa¯ ’. A Milestone of Western Metaphysical Thought, Islamic Philosophy, Theology and Science 63, Leiden/Boston: Brill, 2006. J.J. Cleary, Aristotle & Mathematics. Aporetic Method in Cosmology & Metaphysics, Philosophia antiqua 67, Leiden/New York/Cologne: Brill, 1995. –– , Emending Aristotle’s Division of the Theoretical Sciences, Review of Metaphysics, 48, 1994, pp. 33 – 70. D.R. Cousin, A Note on the Text of Metaphysics 1026a14, Mind, 49, 1940, pp. 495 – 6. C.D’Ancona, Syrianus dans la tradition exgtique de la Mtaphysique d’Aristote. II. Antcdents et postrit, in Le commentaire entre tradition et innovation, ed. M.-O. Goulet-Caz, Paris: Vrin, 2000, pp. 311 – 27. V. Dcarie, La physique porte-t-elle sur des ‘non-spars’ ( !w¾qista)?, Revue des Sciences philosophiques et thologiques, 38, 1954, pp. 466 – 8. G. d’Onofrio, La scala ricamata. La philosophiae divisio di Severino Boezio, tra essere e conoscere, in G. d’Onofrio, ed., La divisione della filosofia e le sue ragioni. Atti del VII Convegno della SISPM (Assisi, 14 – 15 novembre 1997), Quaderni di Schola Salernitana, Cava de’ Tirreni: Avagliano 2001, pp. 11 – 63. G. Fine, Separation, Oxford Studies in Ancient Philosophy, 2, 1984, pp. 31 – 87. –– , Separation: A Reply to Morrison, Oxford Studies in Ancient Philosophy, 3, 1985, pp. 159 – 65. L.-B. Geiger, Abstraction et sparation d’aprs S. Thomas in De Trinitate q. 5, a. 3, Revue des Sciences philosophiques et thologiques, 31, 1947, pp. 3 – 40; reprinted in id., Philosophie et spiritualit, Paris: ditions du Cerf, 1963, I, pp. 87 – 124, and more recently in id., Penser avec Thomas d’Aquin. tudes thomistes prsentes par ˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

305

306 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Pasquale Porro

Ruedi Imbach, Vestigia, 26, Paris/Fribourg: ditions du Cerf/ditions Universitaires de Fribourg, 2000, pp. 139 – 83. D. Gutas, Avicenna and the Aristotelian Tradition. Introduction to Reading Avicenna’s Philosophical Works, Leiden/New York/Copenhagen/Cologne: Brill, 1988. E. Halper, One and Many in Aristotle’s Metaphysics, Columbus: Ohio State University Press, 1989. S.C. Inati, Ibn Sina on Single Expressions, in M. E. Marmura, ed., Islamic Theology and Philosophy. Studies in Honor of George F. Hourani, Albany: State University of New York Press, 1984, pp. 148 – 59. W. Jaeger, Aristoteles. Grundlegung einer Geschichte seiner Entwicklung, 2nd revised ed., Berlin: Weidmann, 1955. K. Kremer, Der Metaphysikbegriff in den Aristoteles-Kommentaren der Ammonius-Schule, Beitrge zur Geschichte der Philosophie und Theologie des Mittelalters 39/1, Mnster: Aschendorff, 1961. C. Lafleur and J. Carrier, Abstraction, sparation et tripartition de la philosophie thortique: quelques lments de l’arrire-fond farabien et artien de Thomas d’Aquin, Super Boetium “De Trinitate”, question 5, article 3, Recherches de Thologie et Philosophie mdivales, 67, 2000, pp. 248 – 71. C. Luna, Syrianus dans la tradition exgtique de la Mtaphysique d’Aristote. I. Syrianus entre Alexandre d’Aphrodise et Asclpius, in M.-O. Goulet-Caz, ed., Le commentaire entre tradition et innovation, Paris: Vrin, 2000, pp. 301 – 309. –– , Trois tudes sur la tradition des commentaires anciens  la ‘Mtaphysique’ d’Aristote, Philosophia antiqua 88, Leiden/Boston/Cologne: E.J. Brill, 2001. M.E. Marmura, Avicenna’s Critique of Platonists in Book VII, Chapter 2 of the Metaphysics of his Healing, in J.E. Montgomery, ed., Arabic Theology, Arabic Philosophy. From the Many to the One: Essays in Celebration of Richard M. Frank, Leuven: Peeters, 2006, pp. 355 – 70. P. Merlan, Abstraction and Metaphysics in St. Thomas’ Summa, Journal of the History of Ideas, 14, 1953, pp. 284 – 91. –– , From Platonism to Neoplatonism, The Hague: Martinus Njihoff, 1954, 2nd ed. 1960, 3rd ed. 1975. A. Moreno, The Nature of Metaphysics, The Thomist, 30, 1966, pp. 109 – 35. D. Morrison, Separation in Aristotle’s Metaphysics, Oxford Studies in Ancient Philosophy, 3, 1985, pp. 125 – 57. –– , Separation: A Reply to Fine, Oxford Studies in Ancient Philosophy, 3, 1985, pp. 167 – 73. –– , Wyqistºr in Aristotle, Harvard Studies in Classical Philology, 89, 1985, pp. 89 – 105. P. Natorp, Thema und Disposition der aristotelischen Metaphysik, Philosophische Monatshefte, 24, 1988, pp. 37 – 65 and 540 – 74. K. Nordberg, Abstraction and Separation in the Light of the Historical Roots of Thomas’ Tripartition of the Theoretical Sciences, in R. Tyçrinoja, A. Inkeri Lehtinen, D. Føllesdall, eds, Knowledge and the Sciences in Medieval Philosophy. Proceedings of the Eighth International Congress of Medieval Philosophy, vol. III, Helsinki: The Finnish Society for Missiology and Ecumenics, 1990, pp. 144 – 53. D.J. O’Meara, Le problme de la mtaphysique dans l’antiquit tardive, Freiburger Zeitschrift fr Philosophie und Theologie, 33, 1986, pp. 3 – 22. J. Owens, Metaphysical Separation in Aquinas, Mediaeval Studies, 34, 1972, pp. 287 – 306. –– , The Doctrine of Being in Aristotelian Metaphysics, Toronto: Pontifical Institute of Mediaeval Studies, 1951.

Immateriality and Separation in Avicenna and Thomas Aquinas 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

307

M.-D. Philippe, )va_qesir, pq|shesir, wyq_feim dans la philosophie d’Aristote, Revue thomiste, 48, 1948, pp. 461 – 79. C. Pietsch, Prinzipienfindung bei Aristoteles. Methoden und erkenntnistheoretische Grundlagen, Beitrge zur Altertumskunde, 22, Stuttgart: Teubner, 1992. P. Porro, Astrazione e separazione: Tommaso d’Aquino e la tradizione greco-araba, in Tommaso d’Aquino, Commenti a Boezio, ed. P. Porro, Milano: Bompiani, 2007, pp. 527 – 80. –– , Metafisica e teologia nella divisione delle scienze speculative del Super Boetium De Trinitate, in Tommaso d’Aquino, Commenti a Boezio, ed. P. Porro, Milano: Bompiani, 2007, pp. 467 – 526. –– , ed., Metaphysica–sapientia–scientia divina. Soggetto e statuto della filosofia prima nel Medioevo, Turnhout/Bari: Brepols/Pagina, 2006 [= Quaestio, 5, 2005]. K. Praechter, Review of Michaelis Ephesii In libros De partibus animalium, De animalium motione, De animalium incessu, ed. M. Hayduck, CAG XXIII/2, Gçttinger Gelehrte Anzeigen, 168, 1906, pp. 861 – 907. J.-D. Robert, La mtaphysique, science distincte de toute autre discipline philosophique, selon saint Thomas d’Aquin, Divus Thomas (Piacenza), 3rd ser., 24, 1947, pp. 206 – 22. A. Speer, The Hidden Heritage: Boethian Metaphysics and its Medieval Tradition, in Metaphysica–sapientia–scientia divina. Soggetto e statuto della filosofia prima nel Medioevo, ed. P. Porro, Turnhout/Bari: Brepols/Pagina, 2006 [= Quaestio, 5, 2005], pp. 163 – 81. L. Spellmann, Substance and Separation in Aristotle, Cambridge: Cambridge University Press, 1995. C. Steel, Theology as First Philosophy. The Neoplatonic Concept of Metaphysics, in Metaphysica–sapientia–scientia divina. Soggetto e statuto della filosofia prima nel Medioevo, ed. P. Porro, Turnhout/Bari: Brepols/Pagina, 2006 [= Quaestio, 5, 2005], pp. 3 – 21. M. Tavuzzi, Aquinas on Resolution in Metaphysics, The Thomist, 55, 1991, pp. 199 – 227. E. Trpanier, La philosophie de la nature porte-t-elle sur des spars ou des non-spars?, Laval thologique et philosophique, 2, 1946, pp. 206 – 209. J.F. Wippel, Metaphysics and separatio according to Thomas Aquinas, The Review of Metaphysics, 31, 1978, pp. 431 – 70, reprinted in id., Metaphysical Themes in Thomas Aquinas, Studies in Philosophy and the History of Philosophy, 10, Washington, D.C.: The Catholic University of America Press, 1984, pp. 69 – 104. C. Witt, Substance and Essence in Aristotle: An Interpretation of ,Metaphysics‘ VII–IX, Ithaca, New York: Cornell University Press, 1989. M. Wundt, Untersuchungen zur Metaphysik des Aristoteles, Stuttgart: Kohlhammer, 1953. H.G. Zekl, Topos: Die aristotelische Lehre von Raum, Hamburg: Meiner, 1990. A. Zimmermann, Ontologie oder Metaphysik. Die Diskussion ber den Gegenstand der Metaphysik im 13. und 14. Jahrhundert. Texte und Untersuchungen, 2nd expanded ed., Leuven: Peeters, 1998.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Two Senses of ‘Common’. Avicenna’s Doctrine of Essence and Aquinas’s View on Individuation* Gabriele Galluzzo Introduction The question I shall try to answer in this paper can be phrased as follows: Is Aquinas’s endorsement of Avicenna’s doctrine of essence compatible with his views on individuation?1 In Sections 1 and 2 I shall try to explain why some tension might be perceived between these two aspects of the Dominican Master’s thought. Roughly speaking, the potential conflict I see is the following. According to Avicenna’s doctrine an essence in itself is neither universal nor particular but rather indifferent to both characteristics. Universality and particularity, in other words, are properties an essence acquires as a result of a certain mode of existence, as a result of its existing, respectively, in the intellect or in the extra-mental world, and not properties it possesses in itself. Thus, within Avicenna’s framework an essence is universal or common only when existing in the intellect and in no other sense. Aquinas’s view on individuation, by contrast, seems to require that an essence be in some sense extra-mentally common. For Aquinas standardly analyses a sensible substance into essence, which accounts for the properties a sensible substance shares with all co-specific substances, and principle of individuation, which is responsible for a sensible substance’s being the individual substance it is. This suggests that co-specific individuals are made of a common constituent, the essence, and of a further

*

1

I wish to warmly thank Amos Bertolacci for his valuable explanations concerning several aspects of Avicenna’s doctrine of essence and Dag Nikolaus Hasse for his comments and suggestions on a first draft of this paper. I am also grateful to all the other participants in the Villa Vigoni conference, and in particular to Giorgio Pini, whose comments helped me to sharpen my views about some crucial points of my argument. I have already dealt with some aspects of this question in: Galluzzo, Aquinas on Common Nature. Although nothing of what I have argued for in my previous paper is strictly incompatible with my considerations in the present one, the following pages should be seen as a substantial refinement of my argument for the conclusion that Avicenna’s doctrine of essence and Aquinas’s views on individuation yield two different notions of commonality or universality. In particular, the distinction in Section 3 between the level of abstraction and the level of modal considerations is something that was not so clear in my mind when I drafted the 2004 paper.

310 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Gabriele Galluzzo

principle which accounts for their individuality. From the point of view of the metaphysical analysis of material substances, therefore, the essence seems to enjoy some kind of extra-mental commonality. But how can it do so, given that Avicenna’s doctrine demands that an essence be common only in the intellect? In Section 3 – 4, I shall suggest one possible way out of the conflict. In brief, my solution consists in distinguishing two levels of analysis (Section 3) and, consequently, two senses of ‘common’ or ‘universal’ (Section 4). Avicenna’s doctrine of essence concerns the level of the actual existence of an essence. It explains, in other words, how an essence can actually exist in different ways (i. e. in the extra-mental world and in the mind) and hence take on incompatible properties. Thus, when actual existence is concerned, an essence exists as universal only in the intellect, which knows an essence by abstracting it, i. e. by separating it out from the individuating conditions along with which it exists in the extra-mental world. The metaphysical analysis of material substances, by contrast, does not concern the actual existence of an essence. This is shown by the fact that the commonality of an essence is revealed through a series of modal considerations concerning the way in which an essence would exist if it could exist without the principle of individuation: if we could, in other words, strip the individuating conditions away from all co-specific substances we would be left with only one essence for each species. Modal considerations show that, although an essence is individual in the extra-mental world in that it exists as an individual, its individuality is not primitive but rather derivative, i. e. due to some principle external to it. Thus, the extra-mental commonality of an essence is simply the idea that the actual, extra-mental individuality of an essence is not metaphysically primitive. And being non-primitively individual is compatible with being actually individual. Accordingly, we should distinguish two senses of ‘common’ and, consequently, two senses of the expression ‘common nature’. If ‘common’ is taken in the sense of ‘actually common’ an essence is actually common only in the intellect, whereas it is actually non-common in the extra-mental world, as Avicenna’s doctrine maintains. But ‘common’ can also be taken in another sense, which can be referred to as ‘modally common’, since it is mainly revealed through modal considerations about the different metaphysical constituents of a substance. According to this second sense of ‘common’, something is common if its individuality is not primitive but derivative, that is due to something else. And in this sense an essence is extra-mentally common, as the metaphysical analysis of material substances plainly shows, for it owes its individuality to some extrinsic principle, i. e. the principle of individuation. But being common in this second sense is compatible with being actually common. For the second sense of ‘common’ indicates a counterfactual (and, as a matter of fact, counterpossible) property of a thing and not an actual one. Therefore, Aquinas’s

Avicenna’s Doctrine of Essence and Aquinas’s View on Individuation 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

311

endorsement of Avicenna’s doctrine of essence is also compatible with his view on individuation.

1. Essence as Indifferent: Aquinas’s Understanding of Avicenna’s Doctrine of Essence Throughout his career Aquinas holds to Avicenna’s view that an essence, when considered in itself or absolutely, is indifferent.2 What Avicenna means is that, when taken absolutely, an essence possesses only those properties which are indicated by its definition. For instance: the human essence, humanity, possesses only those properties which are spelt out in its definition, that is animality and rationality. All the other characteristics, including some important ontological properties such as being one or many, being universal or particular, do not belong to an essence when considered in itself, but only as a result of its acquiring some mode of existence or other. As is well known, the modes of existence Avicenna takes into account are two, extra-mental and mental existence. Thus, an essence is not in itself universal, but acquires universality as a result of its existing in the intellect. Likewise, it is not in itself particular, but acquires particularity as a result of its existence in matter, i. e. as a result of its extra-mental existence. In the extra-mental world an essence exists as an individual object of a certain kind or species, whereas in the mind it is nothing but a universal concept which represents all the individuals of a certain kind in the same way. Of course, since an essence does in fact acquire particularity in its extra-mental existence and universality in its mental one, particularity and universality are neither included in the definition of an essence taken in itself nor excluded by it. In this sense too, therefore, an essence, when taken in itself or absolutely, is indifferent. Universality and particularity invariably accompany an essence in its mental and extra-mental existence, respectively – and so are in this sense disjunctively necessary properties of an essence. Therefore, they are not accidental if by ‘accidental’ we mean ‘purely accidental’ and we wish thereby to refer to properties that something can, unqualifiedly, have or not have.3 For, when the 2

3

Cf. Aquinas, De ent. et ess., c. 3, pp. 374 – 75; Q. de quo., VIII, q. 1, a. 1, vol. I, pp. 51 – 3; Sent. lib. de an., lib. II, c. 12, pp. 115 – 6, lines 95 – 151; Q. de pot., q. 5, a. 9, ad 16, vol. II, p. 155; S. th., Ia, q. 85, a. 2, ad 2, pp. 412 – 3; a. 3, ad 1, p. 414 and ad 4, p. 414; Exp. Metaph., lib. VII, lect. 13, nn. 1570 – 71. The doctrine of the indifference of essence is presented in: Avicenna, Phil. pr., tr. V, c. 1. On Avicenna’s doctrine of essence in general see: Marmura, Avicenna’s Chapters on Universals; id., Avicenna on Primary Concepts; id., Quiddity and Universality; De Libera, La querelle, pp. 185 – 91; id., L’art, pp. 576 – 607; Black, Mental Existence. Cf. Avicenna, Phil. pr., tr. V, c. 1, pp. 232, 7 – 10; 234, 42 – 57; 235, 72 – 236, 91.

312 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Gabriele Galluzzo

essence exists in the extra-mental world, it is necessarily particular; when it exists in the mind, by contrast, it is necessarily universal. In another sense, however, i. e. when ‘accidental’ is taken in the sense of ‘non-essential’, universality and particularity can still be regarded as accidental properties, in that they fall outside the definition of an essence when taken absolutely.4 The ambiguity of the notion of accidentality is also relevant in the context of the debate over the essence-existence distinction, on which the doctrine of the indifference of essence is substantially based. When Aquinas in his late writings criticises Avicenna for holding that the existence of a thing is accidental to its essence, he is clearly taking ‘accidental’ in the sense of ‘purely accidental’ and crediting Avicenna with the claim that existence is related to essence in a purely accidental manner.5 But Avicenna maintains that mental and extra-mental existence – as well as the different properties accompanying these two forms of existence – are disjunctively necessary.6 So Thomas cannot be right in his critical remark.7 What is more, from the argument in the De ente et essentia, c. 3, it clearly emerges that Thomas himself takes the different modes of existence of an essence – and hence the different properties that accompany such modes of existence – to be accidental only in the sense of ‘non-essential’, in that they fall outside the definition of the essence taken absolutely. Thus, once the notion of accidentality is kept straight, there seems to be no fundamental disagreement between Avicenna and Aquinas with regards to the relation between the essence in itself and the different modes of existence it happens to possess. Admittedly, in so far as extra-mental particulars are concerned, Aquinas further maintains that essence and existence should be understood as two principles or ontological constituents of created things – a piece of doctrine which is hardly to be found in Avicenna’s writings. However, the general model according to which the 4 5 6

7

Cf. Avicenna, Phil. pr., tr. V, c. 1, pp. 228, 24 – 32; 230, 58 – 232, 74; 234, 42 – 57; 235, 82 – 236, 91. Cf. Aquinas, Exp. Metaph., lib. IV, lect. 2, nn. 556 – 60. For Avicenna’s main texts on the distinction between essence and existence see: Avicenna, Phil. pr., tr. I, c. 5, pp. 33, 25 – 40, 54; tr. V, c. 1, passim. In the light of Averroes’s subsequent criticisms (see n. 7) the texts concerning the relation among being, one and quiddity should also be added to the list: cf. Avicenna, Phil. pr., tr. III, c. 1, p. 106, 45 ff.; tr. III, c. 2, p. 114, 17 – 20; tr. III, c. 3, p. 117, 80 – 86; pp. 119, 19 – 122, 71; tr. VII, c. 1, p. 349, 7 – 18. In all likelihood, Thomas was led astray by Averroes’s partly inaccurate remarks in his Commentary on the Metaphysics: Averroes, In Metaph., lib. IV, t.c. 3, p. 67B–E; lib. X, t.c. 8, p. 257E–G; 257K (a further passage belonging to Averroes’ commentary on Met., Book V, is omitted in the Latin translation). Averroes does not seem to distinguish very clearly between ‘accidental’ as ‘purely accidental’ and ‘accidental’ in the sense of ‘nonessential’. On the fortune of Avicenna’s views on being and unity see Stephen Menn’s contribution to the present volume (where the case of Averroes is especially taken into consideration).

Avicenna’s Doctrine of Essence and Aquinas’s View on Individuation 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

313

Dominican Master describes the relation between essence and existence seems to be in line with Avicenna’s main intuition: the existence of a thing falls outside its essence, i. e. is external to it without being accidental in the strict sense of the term.8 My previous considerations already suggest that, despite what some texts seem to suggest, when talking of an essence considered in itself Avicenna does not wish to introduce a mysterious third realm of essences, enjoying some kind of not further specified mind-independent existence.9 What Avicenna intends to do, on the contrary, is simply to separate out, in accordance with his distinction between essence and existence, the properties which belong to an essence in virtue of what it is in itself from those following upon its existence, be it mental or extra-mental existence. An essence always exists either mentally or extramentally, it is just that the properties which accompany the mental or extramental existence of an essence are not properties of an essence as such.10 This distinction also enables Avicenna – and Aquinas, who follows Avicenna very 8 The sense in which the existence of a thing falls outside its essence without being accidental to it in the strict sense of the term is clarified by Aquinas by having recourse to the potentiality-actuality model. It is not built into the essence of a thing that it exists and hence the existence of a thing falls outside its essence. The essence, however, fixes the way in which a thing exists, if it exists. Thus, existing for a thing means to actualise a certain essence, which is a possible way of existing. As Aquinas sometimes expresses himself (De sub. sep., c. 8, p. D 55, lines 205 – 12), a thing receives the act of being, i. e. existence, in accordance with the principles of its essence. Accordingly, even though the existence of a thing is extrinsic with respect to its essence, the potentiality-actuality model guarantees that the concrete existence of a thing does not exceed the boundaries fixed by its essence. In this sense, essence and existence are not related in a purely accidental way. 9 For an analysis of Avicenna’s different texts concerning the status of the essence in itself see Wisnovsky’s contribution to the present volume. Some texts in Avicenna (see for instance: Phil. pr., tr. V, c. 1, pp. 233, 36 – 234, 42) seem to imply that the essence in itself is prior to or precedes in being the different properties accompanying mental and extra-mental existence and, consequently, mental and extra-mental existence as well. However, I do not take these texts to entail, necessarily, that the essence in itself enjoys some kind of ontological priority over its different ways of existence and so over existence in general. Possibly, all that Avicenna means is that, since extra-mental individuals and mental concepts are modes of existence of a certain essence, such an essence must be logically presupposed by its different modes of existence. The claim that Avicenna’s essence in itself enjoys some kind of being over and above the being it acquires in its different modes of existence is outlined in Owens, Common Nature, p. 4 (even though Owens confines his remarks to the Latin rendering of Avicenna’s text). 10 Cf. Avicenna, Phil. pr., tr. I, c. 5, p. 36, 78 – 82; 83 – 93 (but the whole discussion up to p. 40, 54 of the absolutely non-existent is relevant); see also: tr. V, c. 1, pp. 234, 44 – 6; 235, 82 – 236, 91. Thus, even if an essence for Avicenna is indifferent to the properties following upon both mental and extra-mental existence, it is not also indifferent to existence taken generally (i. e. so as to include both mental and extra-mental existence). For an essence always exists in some way or other, whether mentally or extra-mentally.

314 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Gabriele Galluzzo

closely here – to explain how an essence can take on contradictory properties, such as universality and particularity. Such properties are not constitutive of an essence as such but are rather possessed by an essence only as a result of its existing in different fashions. If universality and particularity were constitutive properties of an essence they should accompany it, as it were, everywhere. But universality does not belong to an essence when it exists in the extra-mental world, just as particularity is not possessed by an essence when it exists in the intellect. It is likely that Avicenna’s doctrine was also prompted by some concerns about sceptical arguments: How do we bridge the gap between mind and reality? How do we make it sure that our concepts reflect the actual conditions of extra-mentally existing things?11 Avicenna’s doctrine of essence solves the problem by maintaining that our concepts reflect real things in that it is one and the same essence that exists as an individual in the extra-mental world and as a universal concept in the mind and so things and concepts are in some sense made of the same stuff. Be that as it may, the fact that Avicenna does not introduce any mysterious third realm of essences does not make his doctrine any less ambiguous and difficult to interpret. As Giorgio Pini, for instance, has recently pointed out, there are at least two different readings of Avicenna’s doctrine, i. e. an ontological and an epistemological one.12 On the ontological reading, the essence in itself is a real constituent of both extra-mental things and concepts, i. e. something which both extra-mental things and concepts are made of in addition to the other constituents which characterise more specifically extramental and mental beings. According to this view, extra-mental individuals are constituted by essence plus individual characteristics or conditions, whereas concepts have as their constituents essence plus universal characteristics or conditions. On the epistemological and more deflationary reading, by contrast, the essence in itself is nothing but what we are left with when we mentally separate out an essence from the characteristics it enjoys either in its mental or extra-mental existence. According to this second interpretation, we get the essence in itself when we consider extra-mental individuals without their individual characteristics or when we consider concepts without their universal characteristics. The essence in its absolute consideration, in other words, is just the residue of our considering either extra-mental individuals or concepts without the properties which characterise extra-mental and mental existence, respectively.13 Pini further argues that Aquinas’s understanding of Avicenna’s 11 For this see: McGinnis, Logic and Science. 12 Cf. Pini, Absoluta consideratio. 13 It seems to me that the ontological reading has some solid ground in Avicenna’s text, especially when the Arab philosopher employs the part-whole model and appeals to the language of composition in order to describe the relation between the essence in itself

Avicenna’s Doctrine of Essence and Aquinas’s View on Individuation 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

315

doctrine shifts from the ontological reading – which is characteristic of Aquinas’s earlier works such as the De ente et essentia and the Quod., VIII.1.1 – to a more epistemological reading, which marks out the Dominican Master’s later presentations of Avicenna’s view. It is not my intent in this paper to go into the details of Pini’s argument – which would require a detailed analysis of the texts he presents in support of his general reconstruction. However, my considerations in Sections 2 and 3 are meant to cast some doubts on the chronological hypothesis and defend the ontological reading as Aquinas’s prevailing interpretation of Avicenna’s doctrine of essence, at least in so far as the extra-mental existence of an essence is concerned. As my analysis of the essencesupposit problem will show, in fact, most of Aquinas’s late texts on essence seem to presuppose that the essence should be understood as an ontological constituent of extra-mental individuals. This suggests that it should be understood as a constituent of concepts as well. Before tackling Aquinas’s explanation of the individuation of sensible substances, let me go back to the problem of whether Avicenna’s doctrine of essence leaves room for any kind of extra-mental commonality of an essence. In the detailed presentation of Avicenna’s doctrine of essence he offers in c. 3 of the De ente et essentia, Aquinas explicitly defends the view that it is the essence in itself – and not the essence as existing in the intellect – that is predicated of the individuals existing in the extra-mental world.14 Since Aquinas standardly describes predication as a mental operation and so connects it with the intellect, it is likely that the passage in the De ente et essentia introduces a different sense of ‘predication’. Probably, Aquinas is echoing and re-elaborating here on some remarks in Avicenna’s Philosophia prima, where the Arab philosopher insists that the fact that the animal in itself exists in the extra-mental world as an individual animal is compatible with saying that it also exists in the individual animals.15 Avicenna’s worry must be that his talking about the essence in itself could mislead people into thinking that it is not the essence belonging to the extramental individuals that he is talking about, but something different. On the contrary, the essence in itself is not a thing other than the essence we say extramental individuals possess. The same thought, I surmise, is expressed by Aquinas’s observation that it is the essence in itself – and not the essence as existing in the intellect – that is predicated of extra-mental individuals. What Thomas means is that an essence is that which makes of the extra-mental and the essence taken together with the properties it acquires in its different modes of existence (cf. Avicenna, Phil. pr., tr. V, c. 1, pp. 233, 33 – 5; 233, 39 – 234, 42; 234, 50 – 57). I do not wish, however, to insist on this point here. 14 Cf. Aquinas, De ent. et ess., c. 3, p. 375, lines 120 – 22. 15 Avicenna, Phil. pr., tr. V, c. 1, p. 234, 50 – 57, but see also pp. 234, 58 – 237, 22, where Avicenna criticises the Platonist view that the abstracted animal exists in reality (see in particular p. 237, 16 – 21, where the idea of predication crops up as well).

316 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Gabriele Galluzzo

individuals the kind of thing they are and hence is something somehow existing in individuals as well as something individuals can be rightly said to possess. If this is Aquinas’s main thought, it becomes clear why he does not connect, in this instance, predication with mental existence: human beings are rational animals and so possess a certain kind of essence whether or not our intellect applies to them the general concept ‘human being’. Essentialism, in other words, is a view about how things are and not about how we understand or conceptualise things. In spite of his defence of the objectivity of our talk of essences, however, in the De ente et essentia Aquinas is quite explicit that the extra-mental world is populated only by individual entities. What is more, in each individual everything is individuated.16 This suggests that there is no room for something existing in the extra-mental world as a universal or a common entity, not even as a common or universal constituent of particular entities. Universality and commonality can only be found in the intellect, where an essence exists as a concept representing all the individuals of the same kind in the same way.17 So, Aquinas’s view seems to be that the essence that is predicated of extra-mental individuals and is in some sense present in them is not something common or universal, i. e. is not something all the individuals of the same kind share in any ontologically significant sense of the term. Of course, we are still allowed to say that individuals belonging to the same species possess the same essence, but this should not be taken in the literal sense that there is one common component identical in all the individuals of the same kind, but rather in the weaker sense that all the individuals of the same kind possess objective and essential characteristics which group them together and so allow the intellect to class them non-arbitrarily under the same concept. This seems also the meaning of Aquinas’s famous remark in his commentary on the first book of the Sentences to the effect that universals possess certainly some extra-mental ground, but receive their formal completeness, i. e. their full existence as universals, only in the mind.18 This seems to amount to saying that universals have some kind of potential or incomplete existence in the extra-mental world, which is made actual and complete by the intellect. Things, in other words, have objective, essential features which enable us to group them together in a non-arbitrary way. But this does not imply that they share some actual, common constituent.19 16 17 18 19

Cf. Aquinas, De ent. et ess., c. 3, p. 374, lines 80 – 82. Cf. Aquinas, De ent. et ess., c. 3, p. 375, lines 91 – 6. Cf. Aquinas, In I Sent., d. 19, q. 5, a. 1, vol. I, p. 486. In terms of the standard classification of Medieval views on universals, Aquinas’s understanding of Avicenna’s doctrine of essence seems to suggest a nominalist position concerning sameness in kind: things are not the same in kind because they somehow share some common constituent, but only because they are similar in some relevant and essential features. On this view, there is nothing common in reality, it is only our intellect that is able to consider things only with respect to the aspects in which they are similar,

Avicenna’s Doctrine of Essence and Aquinas’s View on Individuation 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

317

2. Essence as Common: Aquinas’s View on Individuation 1. When we turn to Aquinas’s view on individuation, we seem to get a rather different picture of essence. For Aquinas consistently maintains that sensible substances, i. e. material objects, should be analysed into two ontological constituents, the essence or common nature,20 which accounts for the objective characteristics all the individuals of a certain kind share, and the principle of individuation, which explains, by contrast, why an individual is the individual it is, i. e. why it is distinct from all the other individuals of the same kind. In the light of what I have been saying about Aquinas’s understanding of Avicenna’s doctrine of essence, my main task is explaining in which sense the essence which is present in all the individuals of the same kind and which is contrasted with the principle of individuation can be said to be common. For Avicenna’s doctrine of essence seems to imply that there is no ontologically significant sense in which an essence is common when it exists in extra-mental individuals. Before directly addressing this problem, however, I wish to say in this section some more words about Aquinas’s doctrine of individuation and common nature. I shall tackle this topic by focusing on the question of the relation between essence and supposit, which is often forgotten when discussing Aquinas’s doctrine of essence in spite of its pertinence to the question at stake. On Aquinas’s understanding, a supposit is a full-fledged individual of a certain kind: a man or an angel are standard examples of supposita. Thus, a supposit includes the principles or characteristics that make of a certain individual the kind of individual it is as well as those that make of it the particular individual it is. This means that in the case of sensible substances the supposit includes the essence and the principle of inviduation (individual matter), whilst in the case of angelic substances it includes only the essence, since angels are self-subsisting essences and so are individuals in themselves or of

leaving aside those in which they are, instead, dissimilar. Aquinas’s view on individuation, by contrast, seems to point toward a realist solution to the problem of universals: things are similar in kind because they share some common constituent, even though it may be difficult to specify what the commonality involved exactly amounts to. Although my opinion is that Aquinas’s general position is realist in spirit, I wish to present Thomas’s discussion of the problem of universals in abstraction from the standard labels we are used to employing when putting some order in the Medieval debate. 20 For Aquinas’s use of the expression ‘common nature’ to refer generally to the essence of things see for instance: Aquinas, In I Sent., d. 8, q. 4, a. 1, ad 2, vol. I, p. 219; d. 19, q. 4, a. 2, vol. I, p. 483; d. 34, q. 1, a. 1, vol. I, p. 788; C. gent., II, c. 52, n. 1275; IV, 40, n. 3780; S. th., Ia, q. 13, a. 9, pp. 68 – 9; q. 29, a. 2, pp. 151 – 2; q. 76, a. 3, pp. 354 – 6; Q. de pot., q. 9, a. 1, p. 226; a. 5, arg. 13, pp. 234 – 5; ad 13, p. 237.

318 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Gabriele Galluzzo

themselves.21 Usually, therefore, Aquinas does not include in the notion of supposit the act of being, i. e. the ontological principle that unites with a certain essence and enables it to exist.22 Occasionally, however, and in particular in Quod., II.2.2, he seems to think that properly speaking a supposit should comprise not only the essential properties and, when necessary, the individuating principles, but also the act of being which enters into composition with the essence to make up the structure of actual, existing individuals.23 Aquinas’s oscillation seems to conceal a deeper uncertainty regarding what counts as an individual (and hence as a supposit). Not including the act of being in the notion of supposit means to concede that something can be taken to be an individual regardless of whether it exists or not. In other words, when we lay down the conditions for something to be an individual, a full-fledged member of a certain kind, we are not concerned with the individual’s actually existing: some individuals exist, some others do not, but are still thinkable as individuals in that, if they existed, they would be complete, individual objects. Including the act of being in the notion of supposit amounts to saying that it makes sense to talk about a complete individual only in the presence of an actually existing individual. The things, in other words, about which we evaluate whether they are identical with their essence or not are the actually existing individuals and such individuals, in so far as they exist, evidently include the act of being. Aquinas’s waverings concerning the notion of supposit do not affect the question of the relation between essence and supposit in the case of material substances. For on both understandings of supposit, i. e. whether we take the act of being to be part of the supposit or not, supposit and essence turn out to be non-identical in the case of material substances. How we understand the notion of supposit is, by contrast, relevant to the case of angels. For, when Aquinas does not include the act of being in the supposit, he consistently maintains that supposit and essence are identical in the case of angels.24 Angels, in fact, are selfsubsisting essences, i. e. essences that do not need to be received in something else in order to subsist. Accordingly, angels turn out to be individual essences, i. e. essences that do not need something external to themselves to be individuated. Thus, if the act of being is excluded from what it is to be a 21 When I say that angels are individual in themselves or of themselves I simply mean that, since their essence does not need any additional, extrinsic constituent to subsist, it does not need any additional, extrinsic constituent to be an individual, either. Angels, in other words, are created as individuals, i. e. as individual essences. Thus, the individuality of the essences of angelic substances is in this sense primitive. Of course, from a different perspective the individuality of angels (or of their essences) is not absolutely primitive, in that it depends on God’s creative act that brings them into being. 22 Cf. for instance: Aquinas, C. gent., IV, c. 40, n. 3779; Q. de pot., q. 9, a. 1, p. 226 23 Cf. Aquinas, Q. de quo., II, q. 2, a. 2, vol. II, pp. 215 – 8 (esp. p. 217, lines 85 – 102). 24 Cf. Aquinas, Q. de pot., q. 9, a. 1, p. 226.

Avicenna’s Doctrine of Essence and Aquinas’s View on Individuation 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

319

supposit, angels will clearly be identical with their essence, because the essence of each angel is precisely an individual self-subsisting essence. By contrast, when he takes the act of being to be part of the supposit, as he does for instance in Quod., II.2.2, Thomas denies that the identity between supposit and essence holds for the case of separate substances, too.25 For – as Aquinas himself makes it clear – essence and supposit are distinct if there is something in the supposit that falls outside the essence. Thus, since for Aquinas the act of being falls outside the essence and is external to it, if the supposit includes the act of being essence and supposit will not be identical. All things considered, it is not difficult to see why Thomas usually prefers to leave the act of being out of the notion of supposit. For one thing, in fact, this understanding of supposit allows one to talk of actual as well as possible individuals. And talking about possible individuals presupposes that something is conceived of as a full-fledged individual before its being brought into actual existence. For another thing, Aquinas’s general view on the nature of angels fits in better with the claim that a supposit does not include the act of being. For an angel is an individual on account of its essence-principle alone and not also on account of its act of being. An angel is individual because its essence is a self-subsisting essence and selfsubsistence implies individuality. Be that as it may, it is clear that Avicenna’s doctrine of essence is mainly concerned with material substances, where the identity between supposit and essence does not hold regardless of what account of supposit one chooses to endorse. Therefore, in what follows I shall mainly focus on the case of material substances. 2. Aquinas discusses the question of the relation between supposit and essence in a number of texts, within both philosophical and theological contexts. In philosophical contexts, the question about essence and supposit is part of a more general problem Aristotle raises in Book VII (Z), c. 6, of the Metaphysics, namely the question as to whether a thing is identical with its own essence. Admittedly, Aristotle’s notion of ‘thing’ includes more than Aquinas’s notion of ‘supposit’. For a supposit is a full-fledged individual substance, whereas Aristotle’s investigation into the identity claim concerns also things other than substances, such as accidents and accidental composites (i. e. things like ‘white man’, which are the result of the composition of a substance with an accidental property). However, it is clear that the discussion about whether or not sensible substances are identical with their essences is also one of Aristotle’s main concerns. As a matter of fact, Aquinas’s commentary on Met., Book VII will be of much help in Section 3 in determining the extra-mental status of essence. Suffice it to say for the time being that Aquinas credits Aristotle with the view 25 Cf. Aquinas, Q. De quod., q. 2, a. 2, vol. II, p. 217, lines 85 – 102.

320 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Gabriele Galluzzo

that a particular material substance is not identical with its essence26 – a view which, as we have seen, is in agreement with Aquinas’s consistent doctrine throughout his writings. As to theological contexts, there are mainly three areas where Aquinas discusses the problem of the identity between essence and supposit: (i) texts concerning the nature of God, for instance His simplicity (S. th., Ia, q. 3, a. 3; C. gent., I, c. 21) (ii) Trinitarian contexts (S. th., Ia, q. 29, a. 2; Q. de pot., q. 9, a. 1); (iii) finally, texts concerned with the metaphysics of incarnation, with particular reference to the problem as to whether the union of the divine word with the human nature is made in persona or not (Q. verb. inc., a. 1; S. th., IIIa, q. 2, a. 2). I have analysed Aquinas’s use of the supposit-essence problem in theological contexts in another paper of mine27 and hence I shall not dwell on the point in the present context. The only thing which I would like to recall here and which is particularly relevant to our concern is that from all such texts it clearly emerges that, when Aquinas claims that in the case of material substances supposit and essence are not identical, what he means is that they are really, i. e. mind-independently, distinct.28 This is easily realised if we turn our attention for instance to the question of the simplicity of God: What would the point be of insisting that in God supposit and essence are identical, if the non-identity between supposit and essence introduced no mind-independent distinction but a merely conceptual one? After all, supposit and essence can be distinguished conceptually even in the case of God. Thus, to say that they are identical in the case of God is supposed to exclude some kind of mind-independent distinction and, consequently, composition in God’s nature. Similar considerations could be advanced also for the case of incarnation.29 What Aquinas wishes to conclude, therefore, is that in the case of material substances supposit and essence are distinct independently of any activity of our mind. When saying that essence and supposit are mind-independently distinct, I want to exclude even the possibility that the distinction between essence and supposit might be correctly described as a particular kind of conceptual 26 Cf. Aquinas, Exp. Metaph., lib. VII, lect. 11, nn. 1535 – 6. 27 See Galluzzo, Aquinas on Common Nature. 28 Cf. Aquinas In I Sent., d. 5, q. 1, a. 1, vol. I, p. 151; d. 34, q. 1, a. 1, vol. I, pp. 787 – 90; Q. de pot., q. 9, a. 1, p. 226; Q. verb. inc., a. 1, p. 442; a. 2, ad 7, p. 428; S. th., IIIa, q. 2, a. 2, pp. 1872 – 4. For texts where Aquinas employs more nuanced expressions (which do not alter in any event his general point), see: C. gent., lib. IV, c. 40, n. 3781; S. th., Ia, q. 3, a. 3, pp. 16 – 7; S. th., Ia, q. 29, a. 2, ad 3, p. 151; Q. de quo., II, q. 2, a. 2, vol. II, p. 217, lines 107 – 8. 29 For an analysis of the metaphysical machinery behind Aquinas’s doctrine of incarnation, see: Cross, The Metaphysics, pp. 51 – 64. Cross touches upon several issues that are relevant to the understanding of the notions of essence and supposit which I am trying to defend in this section (see in particular his explanation of the part-whole model).

Avicenna’s Doctrine of Essence and Aquinas’s View on Individuation 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

321

distinction (distinction of reason), i. e. the kind of conceptual distinction which depends not only on the characteristics of our mind but also on the objective features of extra-mental things. In book I of his Commentary on the Sentences, in fact, Aquinas implicitly distinguishes two different kinds of distinction of reason, i. e. of conceptual distinction.30 According to one kind of distinction, two things are distinct in the intellect only on account of an operation of the intellect itself, only because, in other words, our intellect is capable of introducing distinctions even if they have no objective ground in the extramental world. According to another kind of distinction, by contrast, two things are distinct in the intellect not only on account of an operation of the intellect but also on account of some objective features of the world. Since also this second kind of distinction is a conceptual distinction, Aquinas’s point is not that the things that are distinct in the intellect are also distinct in reality. On the contrary, such things are distinct only in the intellect. However, in the case of the second kind of conceptual distinction our intellect’s conceiving things as distinct or separate is somehow grounded on reality and is not just the product of our mental activity. As an example of the second kind of conceptual distinction Aquinas presents the case of the multiplicity of divine attributes. The different divine attributes are in reality, i. e. in God, only one thing in that they are all really identical with the divine essence. However, our attribution to God of a multiplicity of attributes is grounded on the nature of God Himself, i. e. on the plenitude or richness of God’s nature. In God there is no real multiplicity and hence our conception of the divine attributes as multiple and distinct is due only to the limitedness of our intellect, which is unable to capture the infinity of God’s nature. Still, the conceptual distinction of different divine attributes is not unrelated to God’s nature, in that God’s nature contains in its simplicity all the perfections signified by the attributes we think of as distinct. Now, when I say that, in the case of material things, the distinction between essence and supposit is mind-independent, what I am suggesting is that such a distinction should not be understood as an instance of the second kind of conceptual distinction Aquinas mentions in his Commentary on the Sentences, i. e. the conceptual distinction which is grounded on the nature of the things we think of. For the distinction between essence and supposit does not depend on the activity of our intellect and hence is there whether we think about things or not. For if the distinction between essence and supposit in material creatures were only a conceptual distinction, material objects would not be really composed of essence and principle of individuation, just as the divine essence is not really composed of the different perfections which the divine attributes signify. Texts show, however, that Aquinas takes material objects to be really composed of essence and principle of individuation. Therefore, the distinction 30 Cf. Aquinas, In I Sent., d. 2, q. 1, a. 3, vol. I, pp. 63 – 72 (esp. pp. 69 – 71).

322 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Gabriele Galluzzo

between essence and supposit must be stronger than any kind of conceptual distinction, including the one which is grounded on how things are in reality. The sense in which supposit and essence are mind-independently distinct in the case of material substances can be further refined. As I see things, Aquinas conceives of the relation between supposit and essence according to a part-whole model. In other words, in material substances the essence is a part, i. e. an ontological constituent, of a supposit, i. e. of an individual. Thus, it is not incorrect to describe an individual as ‘essence plus something else’, the something else being the principle of individuation which marks off one particular individual from all the other co-specific individuals. To my knowledge, only in one text does Aquinas cast some doubts on the partwhole model as an adequate explanation of the relation between essence and supposit, namely Quod., II.2.2, which I have already mentioned in connection with the problem of the identity between a thing and its essence. In this text, Aquinas suggests that it is not unqualifiedly correct to describe the relation between essence and supposit as one between a part and its whole, but it is better to say that supposit and essence are signified, respectively, as whole and part. In all likelihood, Aquinas’s occasional scepticism about the part-whole model is due to a peculiar feature of his doctrine of essence. As is known, according to Aquinas the essence of material substances includes not only the form of such substances, but also their so-called common matter, i. e. the type of matter all individuals of a certain species possess.31 The principle of individuation for material substances, by contrast, is the so-called materia signata or individual matter, i. e. the material characteristics which are proper to one individual alone and so distinguish it from the other co-specific individuals. Upon reflection, the couple common matter-individual matter is a real obstacle to conceiving of the relation between essence and supposit in terms of part and whole. For individual matter seems to be nothing but a further specification and determination of common matter: if the common matter of human beings is flesh and bones, for instance, their individual matter will be individual flesh and bones, i. e. flesh and bones of certain, specified dimensions. Thus, it seems difficult to view individual matter as a further, distinct component which is added on to the essence in order to make up a whole, in that individual matter relates to a part of the essence, i. e. common matter, not as something extrinsic and distinct, but rather as the more determinate relates to the less determinate. And in fact Aquinas quite understandably observes in the body of Quod., II.2.2 that the part-whole model would be more properly applicable if the essence of material substances contained form alone and not also common matter. For in such a case the essence would be purely formal and the principle of 31 Cf. for instance: Aquinas, De ent. et ess., c. 2, pp. 67 – 84, pp. 370 – 71; Exp. Metaph., lib. VII, lect. 9, nn. 1467 – 9.

Avicenna’s Doctrine of Essence and Aquinas’s View on Individuation 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

323

individuation purely material. Matter and form, on the other hand, are totally distinct components and hence also essence and principle of individuation would be, in the hypothetical situation, two different entities which make up a third, distinct entity, i. e. the supposit. In the situation considered, in other words, both essence and principle of individuation would be easily conceived of as parts of the supposit they jointly make up. In spite of these difficulties, my impression is that Aquinas’s uneasiness in the text at issue with the part-whole model does not undermine his general intuition that the essence is an ontological constituent of material objects. For it is known that Aquinas devises no intermediate distinction between real distinction and distinction of reason, i. e. conceptual distinction. What is more, while Aquinas recognises, as we have seen, that there are different kinds of distinction of reason, i. e. conceptual distinction, he never explicitly admits of different kinds of real distinction. His real distinction, however, seems to include different degrees of mind-independent distinctness. Two things, in other words, can be distinct from one another to a higher or lesser degree than two others. For instance, Aquinas’s reasonable point in Quod., II.2.2 is that essence and principle of individuation (and consequently essence and supposit) are distinct to a lesser degree than matter and form. It is not difficult to see why Aquinas thinks so. In the standard Aristotelian picture, as long as a material substance exists, its matter and its form are only one object and not two. The matter and form of a material substance are, as we say, coinciding entities. Still, matter and form are distinct, i. e. are two distinct constituents of a material substance. So, how can we set them apart? One line of thought is to appeal to the idea that matter and form have or, at least, may have different life histories. The one can exist when the other does not. A piece of matter, for instance, can take on different forms over time. Presumably, matter and form will also have different modal properties: there are things matter can do that form cannot do or changes matter can undergo that form cannot. However, appealing to the different life histories of matter and form seems to be enough to set them apart. So, in so far as they may have different life histories, matter and form are in some sense separable. The case of the relation between principle of individuation and essence seems to be very different. The existence of individual matter is always bound up with the essence it renders individual. There is no time at which individual matter exists and the essence it makes individual does not. Conversely, the essence of a sensible substance, in so far as it is not a self-subsisting essence but always needs something else to subsist, cannot exist without the principle of individuation which enables it to subsist. Thus, there is no sense in which principle of individuation and essence can exist separately. Therefore, Aquinas seems to be right in maintaining that matter and form are more distinct than essence and principle of individuation. On the other hand, Thomas seems to think that this difference in degree between the

324 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Gabriele Galluzzo

matter-form and the essence-principle of individuation relations should not prevent us from regarding essence and principle of individuation as mindindependently distinct and from conceiving of them as two ontological constituents of an individual thing. Therefore, I shall assume that supposit and essence are mind-independently distinct in the case of material substances and that the part-whole model works after all as the best explanation of their relation. According to Aquinas an individual substance can be correctly described as made of two ontological constituents, namely the essence and the principle of individuation. More important for our purposes, this understanding of the structure of material beings strongly suggests that the essence-component is in some sense common to all the individuals of the same species. For if one individual of a certain natural kind is distinct from another on account of the principle of individuation, i. e. of its different individual material characteristics, there follows that co-specific individuals are not distinct on account of their essence and so the essence-component of an individual must be somehow common or shared. To see that this is actually the way Aquinas views things, I would like to call the attention to his De potentia, q. 9, a. 1, a text devoted to introducing the main terms involved in the discussion of the trinity, ‘person’, ‘essence’, ‘subsistence’ and ‘hypostasis’. The thesis defended in the article is that the distinction between essence and individual is real, i. e. mind-independent, in the case of material substances, while it is only a distinction of reason in the case of immaterial ones (i. e. God and angels). In so far as material substances are concerned, Thomas justifies his claim as follows: The reason for this distinction [i.e. the distinction between individual and essence] is that several subjects may have a common nature; thus several men have in common the nature of man. Hence the need of distinguishing what is one from what is multiplied … For if whatsoever is in the individual were to belong to the common nature, there would be no possible distinction between individual substances of the same nature. Now what is in the individual substance besides the common nature is individual matter (which is the principle of singularity) …32 32 Cf. Aquinas, Q. de pot., q. 9, a. 1, p. 226: ‘Huius autem distinctionis ratio est, quia inveniuntur plura subiecta in una natura convenire, sicut plures homines in una natura hominis. Unde oportuit distingui quod est unum, ab eo quod multiplicatur: natura enim communis est quam significat definitio indicans quid est res; unde ipsa natura communis, essentia vel quidditas dicitur. Quidquid ergo est in re ad naturam communem pertinens, sub significatione essentiae continetur, non autem quidquid est in substantia particulari, est huiusmodi. Si enim quidquid est in substantia particulari ad naturam communem pertineret, non posset esse distinctio inter substantias particulares eiusdem naturae. Hoc autem quod est in substantia particulari praeter naturam communem, est materia individualis quae est singularitatis principium’. English translation: Aquinas, On the Power of God, vol. III, pp. 98 – 9 (slightly modified).

Avicenna’s Doctrine of Essence and Aquinas’s View on Individuation 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

325

What is striking about this argument is its distinctively realist flavour. Aquinas seems to assume right from the start that in order to explain how there can be many individuals of the same species we need to posit in each of the co-specific individuals a mind-independent distinction between two ontological constituents. More particularly, we should distinguish between something which is in itself common and one, i. e. the essence which gets individualised and multiplied, and something which is responsible for the individualisation and multiplication of the common component. It is crucial to Aquinas’s argument that the Dominican Master speaks of ‘individuals of the same species’. For it is not unconceivable to imagine a situation in which individual material substances would possess individual essences and so would be distinct from one another not on account of some extra component which particularises their common nature, but on account of their individual essences themselves. However, the problem with this suggestion is that, according to Aquinas, in the situation imagined individual substances would not belong to the same species. For they would differ on account of their essences, and each essential difference entails a difference in species.33 The situation imagined is not dissimilar from the actual case of angels. Angels are self-subsisting essences and hence individual substances.34 However, since the difference between one angel and another is a difference in essence, in that each angel is a different self-subsisting essence, one angel differs from another in species and not in number. For difference in number occurs only when two things have the same essence and differ on account of characteristics external to the essence. This is why co-specific material substances differ in number: their common essence must be received in matter in order to subsist and so it gets numerically multiplied on being received in different pieces of matter. In conclusion, for Aquinas there cannot be more than one essence for each species of material substance and the plurification of such an essence must be accounted for by a principle of individuation which is external to the essence itself. In other words, if we could strip the principle of individuation away from all co-specific material substances we would be left with only one essence for each species.

33 Cf. Aquinas, In II Sent., d. 3, q. 1, a. 4, vol. II, p. 97; d. 17, q. 2, a. 1, ad 5, vol. II, p. 430; C. gent., II, c. 81, n. 1621; c. 93, n. 1797; Q. de pot., q. 3, a. 10, p. 71. 34 Cf. for instance: Aquinas, De unit. intell., c. 5, p. 311, lines 63 – 7; 71 – 4; S. th., Ia, q. 76, a. 2, ad 3, p. 354; Q. de an., q. 2, ad 5, pp. 19 – 20, lines 361 – 92; Q. de spir., a. 8, ad 8, p. 63, lines 312 – 24; Exp. Peryer., lib. I, lect. 10, p. 50, lines 95 – 103.

326 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Gabriele Galluzzo

3. A Possible Way out: Actual vs. Modal Properties35 If my analysis so far is correct, there seems to be some tension in Aquinas’s conception of essence. On the one hand, according to Aquinas’s interpretation of Avicenna’s doctrine of essence there is nothing common or universal in the extra-mental world. In the extra-mental world everything is individual. Moreover, the essence, when considered in itself or absolutely, is neither particular nor universal. Commonality and unity, which are the distinguishing marks of universality, accompany only the mental existence of an essence. It is only in the intellect, which separates out the essence from the individuating principles, that an essence can be regarded as common and one. On the other hand, Aquinas’s views on the individuation of material substances suggest that each material substance should be analysed into two ontological constituents, the common essence and the principle of individuation which multiplies and particularises such an essence. It might seem, therefore, that there is after all something common existing in the extra-mental world, i. e. the common nature which all the individuals of a certain species share. Is there any genuine conflict between these two aspects of Aquinas’s thought? If not, is there a way of dispelling the appearance of conflict? Finally: Is there any connection between the sense in which an essence is common from the point of view of individuation and the Avicennian claim that it is common only in the intellect? These are the questions I shall try to give an answer to in this section. As a step towards a possible solution, let us see first of all what Aquinas means by ‘essence’ or ‘common nature’ when talking about individuation. As I have already hinted, his point seems to be that if we could strip the individuating principles away from each individual of a certain kind we would be left with only one essence. However, since Aquinas believes that we cannot in fact strip the individuating principles away from a sensible substance, it is clear that in whatever sense an essence is common in the individuals possessing it, it is not actually common. In actuality, an essence exists in the extra-mental world as individuated and multiplied. Probably, the text where Aquinas best explains the sense in which an essence can be said to be common, even though not actually common, is his commentary on Met. Z 11.36 In the text, Aquinas tries to make sense of Aristotle’s claim that an individual sensible substance, i. e. a composite of matter 35 In this section I have taken inspiration from the remarks on coinciding entities contained in Lowe, A Survey, pp. 59 – 76, where it is argued that coinciding entities (and ontological constituents can be certainly thought of as coinciding entities) can be distinguished on the basis of their different historical and/or modal properties. The application of such conceptual tools to the case of Aquinas’s ontology is, of course, mine. 36 Cf. Aquinas, Exp. Metaph., lib. VII, lect. 11, nn. 1535 – 6.

Avicenna’s Doctrine of Essence and Aquinas’s View on Individuation 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

327

and form, is not identical with its own essence.37 Aquinas’s argument in favour of Aristotle’s claim seems to have two crucial premisses. (P1) First, an essence is what is expressed by a definition. Therefore, since a definition refers only to the species an individual belongs to and not to the individual as such, an essence pertains only to the species and not to the individual. (P2) Second, the essences of material substances, of necessity, exist only in individuals. Such essences, in other words, are not self-subsisting essences, but rather need something else, i. e. some kind of material substratum, to subsist. From these two premisses, Aquinas seems to draw two conclusions. (C1) First, since the essence belongs to the species and the essences of material substances only exist in individuals, then we must distinguish within each individual the essence from the principle of individuation. The principle of individuation, besides being what individualises and multiplies the essence, is also the principle enabling it to subsist. For the essence of material substances cannot subsist by itself. (C2) The second conclusion is that, since the essence of material substances contains matter, we must also distinguish, within each material substance, between common matter, which falls within the essence, and individual matter (materia signata), which falls outside the essence. Individual matter is precisely the principle of individuation. As can be seen, (C2) is Aquinas’s standard distinction between common and individual matter, while (C1) perfectly reflects his view on the relation between essence and supposit I have tried to illustrate in Section 2, i. e. the claim that in material substances the essence is distinct from the supposit and is an ontological constituent of it. Some new element comes from the further remark which Aquinas appends to his main argument.38 He observes that, even though in the extra-mental world an essence cannot exist apart from individuals, i. e. apart from the individuating principles enabling it to subsist, it can do so in the intellect, which knows essences by separating them from the material and individual conditions along with which they exist in the extra-mental world. 37 Aristotle discusses the question of the identity (or non-identity) between a thing and its essence in Met. Z 6 and then goes back to it in Met. Z 11 (1037a33-b7), in the course of his final summary of the results achieved by the enquiry into the notion of essence. It is when commenting on the Z 11 passage (Exp. Metaph., lib. VII, lect. 11, nn. 1535 – 6) that Aquinas presents his final interpretation of the identity claim. For, as he himself observes (Exp. Metaph., lib. VII, lect. 11, n. 1535), Aristotle’s discussion in Z 6 leaves it unclear whether sensible substances are in fact identical with their essence or not. In his final reconsideration of the question in Z 11, however, Aristotle makes it clear that things which can be analysed into a formal and a material constituent are not identical with their own essence (1037b3 – 4). This clearly includes also the case of material substances. For more on the identity thesis and for a general reconstruction of Aquinas’s intepretation of Met., Book VII, I take the liberty of referring to Galluzzo, Aquinas’s interpretation. See also: id., Averroes and Aquinas. 38 Cf. Aquinas, Exp. Metaph., lib. VII, lect. 11, nn. 1535 – 6.

328 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Gabriele Galluzzo

Therefore, even if an individual material substance is not identical with its own essence in the extra-mental world, it is in some sense identical with its essence in the intellect. For in the intellect an individual exists without material and individuating conditions and so in its mental form of existence does not contain anything in addition to its essence. Aquinas’s remark in the commentary on Met. Z 11 might be taken to be offering an easy way out of the contrast between Avicenna’s doctrine of essence and the Dominican Master’s view on individuation. According to Avicenna’s doctrine or at least to Aquinas’s understanding thereof, an essence is not universal or common in the extra-mental world, for there it exists as an individual of a certain kind. Neither is it common in itself, for in itself it is indifferent. An essence is common, by contrast, in the intellect, where it exists as a universal concept. Now, when Aquinas says that a material substance is analysable into a common essence and a principle of individuation, what he means, at least to stick to his own words, is that if we strip the principle of individuation away from all the individuals of a certain kind we are left with only one essence. However, it is only the intellect that can set apart essence and principle of individuation and so it is only in the intellect that an essence exists as one and common. Thus, the suggestion is that Aquinas’s talking of a material substance’s being analysable into two different constituents is just another way of saying that our intellect is capable of separating essence and principle of individuation.39 Such an intellectual operation is not arbitrary, in that it is based on the objective features of extra-mental things and on their essential similarities. However, it does not force us to introduce any extra-mental commonality or universality with regards to the essence of material substances. Although this suggestion is very neat and economical, I find it, all things considered, slightly misleading. Its main fault is that of not distinguishing between the natural operation of abstraction and the philosophical considerations that lead us to analyse sensible substances into essence and principle of individuation. In other words, we arrive at the conclusion that sensible substances are composed of essence and principle of individuation not as a result of our natural cognitive processes, but rather through a series of philosophical considerations concerning the metaphysical structure of sensible substances. Therefore, the claim that essence is the common constituent of sensible substances can hardly be reduced, as the suggestion just presented holds, to the 39 For texts that might be taken to support this suggestion see: Aquinas, In I Sent., d. 19, q. 4, a. 2, vol. I, p. 483; q. 5, a. 1, vol. I, p. 486; C. gent., I, c. 26, n. 241. In all these passages Aquinas puts emphasis on the idea that commonality and unity belong to an essence only when it exists in the mind. In the second text in particular, he makes also reference to the already mentioned claim that universals have only a potential or incomplete existence in the extra-mental world, which is made actual and complete by the intellect. This claim too goes in the direction of the suggestion I have just presented.

Avicenna’s Doctrine of Essence and Aquinas’s View on Individuation 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

329

fact that the natural process of abstraction presents to the intellect the essence without individuating conditions. Thus, I shall offer another way out of the difficulty, which, I think, is closer to Aquinas’s main intuition. My view is that Avicenna’s doctrine of essence and Aquinas’s view on individuation introduce two different levels of analysis and hence appeal to different kinds of consideration. Avicenna’s doctrine concerns what can be called the actual existence of an essence. It explains in other words how it is possible for an essence to actually exist in different manners and to enjoy incompatible properties. The solution consists in distinguishing what an essence is in itself from the properties it enjoys in its different modes of actual existence, be it mental or extra-mental existence. Aquinas’s doctrine of individuation, by contrast, does not have to do with the actual existence of an essence, but rather with the metaphysical structure of extra-mental things, which can be unravelled only by appealing to modal considerations, i. e. to situations which depart from the actual existence of an essence. In other words, just as we can distinguish two different constituents of a thing, such as for instance matter and form, on the basis of their different life histories, so we can set apart essence and principle of individuation on the basis of their different modal properties. The implicit assumption in the argument is that two things which have different modal properties are not identical. The very language Aquinas employs to characterise the common constituent of a sensible substance reveals his appealing to the modal properties of an essence. If we could – Aquinas seems to imply – strip the principle of individuation away from the individuals of a certain kind we would be left with only one essence; or alternatively: if the common nature of a certain species could exist separately, i. e. in separation from the conditions along with which it actually exists, it would be just one.40 More specifically, Aquinas is here appealing to some sort of counterpossible considerations (per impossibile) – as is signalled by the ‘could’ in my English rendering of his argument – since it is in fact metaphysically impossible to separate essence and principle of individuation. However, the fact that such considerations are based on metaphysical 40 Cf. Aquinas, Exp. Metaph., lib. VII, lect. 11, nn. 1535 – 6; C. gent., II, c. 42, n. 1275; Q. de spir., a. 8, p. 80, lines 189 – 204. See also Exp. Peryer., lib. I, lect. 10, p. 50, lines 95 – 103, where Aquinas proves (as in Q. de spir., a. 8) that there does not exist more than one separate substance for each species by appealing to the thought-experiment that, if whiteness could exist separate from matter, it would be just one. In Q. de spir., a. 1, p. 13, lines 363 – 76 the example of whiteness crops up again, this time around to support the claim that there can be only one thing that is its own being or existence: just as it is impossible that there would be more than one whiteness, if whiteness could exist in separation from any subject, so there cannot be more than one self-subsisting act of being. The connection with the case of God and His unicity may suggest one more reason why Aquinas is willing to put ontological weight on the modal status of an essence, but I do not wish to pursue this issue at present.

330 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Gabriele Galluzzo

counterpossibility does not exclude it that they may still reveal genuine, metaphysical properties of an essence. Quite the contrary, modal considerations – i. e. considerations about how an essence would exist if we could get rid of the principle of individuation – enable us to separate the properties an essence possesses as a result of its actual existence from its fundamental metaphysical properties. The fundamental metaphysical property of an essence is the fact of being sharable. The essence of sensible substances is not actually and literally shared because it always exists together with the principle of individuation. However, from the point of view of the metaphysical structure of extra-mental things an essence is always sharable. It is important to stress once again that the modal considerations which underlie the metaphysical analysis of material substances should not be confused with the process of abstraction by which we obtain universal concepts. Abstraction is in fact an undeliberate and natural process, whereas the metaphysical analysis of material objects is clearly based on voluntary and deliberate considerations of the intellect, which reflects upon the modal properties of things.41 Of course, the fact that our intellect is capable of separating, through abstraction, an essence from the principle of individuation may help us to carry through the metaphysical analysis of material objects.42 It 41 It might be observed that, in some cases, abstraction does not turn out to be such an easy and natural process, which can be carried through unaided by philosophical considerations. We may experience difficulties – it might be remarked – in forming the concepts corresponding to certain peculiar common natures, such as for instance those natures that have unique instantiations, like the sun and the moon. My response to this observation would be that the difficulties we experience with the case of the sun and the moon do not concern our concepts of what the sun or the moon is (which are obtained rather naturally and spontaneously) but have rather to do with whether the concept of the sun or that of the moon should be taken to be concepts of individuals or of species. Unique instantiation might be wrongly taken to exclude a distinction between individual and species. To see the distinction, of course, philosophical considerations are required. 42 Cf. for instance: Aquinas, Q. de spir., q. 8, p. 80, lines 189 – 204; Exp. Peryer., lib. I, lect. 10, p. 50, lines 95 – 103, where Aquinas always associates the thought that an essence would be just one if it could exist separately with the claim that it does exist as one in the intellect. I take it, however, that the meaning of these texts is not that the extra-mental commonality of an essence reduces itself to its potentiality to exist as one and common in the intellect, but rather that the kind of existence an essence enjoys in the intellect can help us to push forward our investigation into the metaphysical structure of material things. Amos Bertolacci rightly pointed out to me that also the modal considerations by means of which we distinguish between common nature and principle of individuation are mental in character and hence the extra-mental commonality of an essence can be fully grasped only in the mind. I agree with him. I would like only to insist that modal considerations reveal the existence of extra-mental natures, while abstraction by itself does not tell us anything about the metaphysical structure of the

Avicenna’s Doctrine of Essence and Aquinas’s View on Individuation 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

331

may help us to see, for instance, that conceiving of the essence without the principle of individuation is at least logically, even though not metaphysically, possible – a claim, incidentally, which seems to be implied by Aquinas’s remark, in his commentary on Met. Z 11, to the effect that logic can treat essences as if they existed in separation from individuating conditions.43 However, our discovery of the modal properties of essences cannot be reduced to the ability of our intellect to get universal concepts through abstraction or, to put it better, to the things’ potentiality to be known by the intellect in a universal form.44 things the concepts of which we come to acquire (for further development of this idea, see below, n. 44). 43 By ‘logically possible’ I simply mean that there is nothing contradictory in supposing that an essence exists without individuating conditions, as is testified to by its so existing in the intellect. Given the metaphysical structure of the extra-mental world, however, the essence of a sensible substance cannot exist without individuating conditions and its so existing, therefore, is metaphysically, although not logically, impossible. Alternatively, one could distinguish between two kinds of metaphysical possibility, i. e. absolute metaphysical possibility (which amounts to the same thing as logical possibility) and conditional metaphysical possibility. Absolutely speaking, there is no metaphysical impossibility in supposing the essence of a sensible substance to exist without individuating conditions. Conditionally, however, i. e. given the general metaphysical presuppositions that govern God’s creation of the world (such as, for instance, the need of introducing a hierarchy of beings with different degrees of perfection and metaphysical composition) it is metaphysically impossible for the essence of a sensible substance to exist without individuating conditions. Since I find this second, Leibnizstyle solution slightly anachronistic, I prefer to distinguish straightaway between logical and metaphysical possibility. 44 One might try to soften the sharp contrast I have introduced between abstraction and metaphysical or modal considerations by talking, as an alternative, of two different interpretations of the process of abstraction, which we can label for simplicity ‘realist’ and ‘nominalist’ interpretation. According to the realist interpretation, abstraction is a process of separation by which the intellect is able to isolate the common constituents of things and separate them out from the individuating conditions along with which they exist in the extra-mental world. One crucial assumption of this way of construing abstraction is that there are common constituents in the extra-mental world, which only need an intellect to be isolated and so brought into light. According to the nominalist interpretation, by contrast, there are no common constituents in the extra-mental world, but only similarities among things. Abstraction, therefore, is simply a process of selective attention by means of which the intellect is capable of spotting the respects in which things are similar and leave out those in which they are dissimilar. On my reconstruction, Aquinas is clearly a supporter of the realist interpretation of abstraction. I wish to pursue this way of making my point in a future paper. For now, I shall confine myself to observing that both the realist and the nominalist interpretation of abstraction are philosophical reconstructions of the natural process by which we acquire general concepts. Thus, some distinction between the natural process of abstraction and its metaphysical interpretation seems to be in order even if one prefers to speak of two ways of conceiving the process rather than distinguishing between abstraction and metaphysical considerations.

332 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Gabriele Galluzzo

My general point about the compatibility between Avicenna’s doctrine of essence and Aquinas’s metaphysical views can be formulated also in a slightly different way. According to Aquinas’s doctrine in the De ente et essentia, the extra-mental sensible world is populated by individuals. All the individuals belonging to a certain species are instances of a certain essence: all human beings are instances of the human nature. Therefore, in the extra-mental world essences are individual, i. e. exist as individuals. However, the individuality of an essence can be primitive or derivative. It is primitive if an essence needs nothing else to be individual, whereas it is derivative if an essence needs something else, i. e. something extrinsic, to become individual. And it is here that the metaphysical analysis enters the picture. Such an analysis does not call into question the claim that an essence is individual in its actual existence, but rather reveals that the individuality of an essence is not primitive, in that it is due to a principle external to the essence itself. Aquinas’s argument for the claim that in material substances supposit and essence are mind-independently distinct contributes precisely to establish the point that the individuality of an essence is not primitive, but rather due to a principle distinct from it. To the same effect is Aquinas’s observation that, if an essence could exist separately from the principle of individuation, it would be just one and not multiplied. Reflecting upon the way in which an essence would exist, if it could exist separately, shows once again that its actual individuality is not primitive but rather derivative. And, as I see things, being non-primitively individual is compatible with being actually individual. For being primitively individual and being derivatively individual are precisely properties we attribute to essences which are in fact actually individual, and so attributing such properties in some sense presupposes that an essence be individual in its actual, extra-mental existence. Thus, there can be no possible conflict between the two claims.

4. Two Senses of ‘Common’ It seems to me, therefore, that the notion of ‘common nature’, by which Aquinas often refers to the essence of sensible things, is in some sense ambiguous and that disambiguation in this case requires distinguishing between two different senses of ‘common’. If by ‘common’ we mean ‘actually common’ it is clear that, according to Avicenna’s doctrine of essence, an essence is actually common only in the intellect, where it exists as a universal concept. This notion is clear enough, but a couple of things should be clarified. The first is that, when Aquinas says that a concept is common or universal, he means to refer to the representational content of a concept and not to its ontological status.45 A 45 See for this distinction: Aquinas, Exp. Peryer, lib. I, lect. 3, p. 15, lines 51 – 78.

Avicenna’s Doctrine of Essence and Aquinas’s View on Individuation 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

333

concept, in other words, is common or universal because for each species there is in the intellect just one concept representing all the individuals belonging to the species in the same way, i. e. with respect to their common aspects. From the ontological point of view, however, i. e. from the point of view of the kind of thing a concept is, a concept is just a quality of the mind and hence is in this sense individual. In any event, it is clearly the representational content that Aquinas has in mind when he describes a concept as common or universal. The second thing to be clarified is that the fact that a concept is common or universal because it represents all the individuals of a certain kind is not equivalent to and so should not be confused with the intellect’s conscious attribution of universality to a concept.46 Like many other medieval philosophers, Aquinas distinguishes in fact between first and second intentions. First intentions are concepts such as ‘man’, ‘horse’ and the like, i. e. first-order concepts that directly represent extra-mental things. Second intentions, by contrast, are concepts such as ‘species’, ‘genus’, ‘particular’ and ‘universal’. Such concepts do not directly represent extra-mental things but rather reflect the way in which things are known by our intellect and so exist therein. Second intentions, in other words, are second-order concepts, i. e. attributes of firstorder concepts. It is the concept ‘man’ – and not the particular men represented by the concept ‘man’ – that can be called both a species and a universal. On Aquinas’s standard account, the intellect obtains second-order concepts by reflecting upon first-order concepts and in particular upon the way they represent the extra-mental things falling under them. The intellect, for instance, attributes the property of being universal to the concept ‘man’ after realising that such a concept is a unique concept representing all men in the same way. Thus, when I say that, according to Aquinas, an essence is universal in the mind in that it exists therein as a universal concept, I do not mean to refer to universality taken as a second intention, but rather to the objective fact that first-order concepts – such as ‘man’, ‘horse’ and the like – represent all the individuals falling under them regardless of whether the intellect in fact recognises the representational function of these concepts or not. It is, in other words, the universality that goes along with first intentions, and not that connected with second intentions, that I am thinking of. In conclusion, our first sense of ‘common’ in the expression ‘common nature’ can be summarised as follows: according to Avicenna’s doctrine of essence or at least to Aquinas’s interpretation thereof, an essence is actually universal only in the mind where it exists as a firstorder concept representing all individuals falling under it in the same way. 46 For texts where Aquinas introduces, explicitly or implicitly, this distinction see: Aquinas, In I Sent., d. 2, q. 1, a. 3, vol. I, p. 67; Q. de pot., q. 7, a. 9, p. 201; S. th., Ia, q. 85, a. 2, ad 3, pp. 412 – 3; a. 3, ad 4, p. 414; Sent. lib. de an., lib. II, c. 12, p. 116, lines 139 – 43.

334 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Gabriele Galluzzo

If my reconstruction is correct, however, Aquinas presents us with a second sense of ‘common’, i. e. the sense emerging from the analysis of the metaphysical structure of material objects. This second sense of ‘common’ concerns the extramental existence of an essence. Since an essence, however, exists in the extramental world as an individual, the second sense of ‘common’ cannot be understood as ‘actually common’. Given the importance that modal considerations play in introducing the second sense of ‘common’, I shall refer to it simply as ‘modally common’. As we have seen, the modal commonality of an essence seems to amount to the basic idea that the individuality of an essence is not primitive but rather derivative, in that it depends on a principle falling outside the essence, the principle of individuation. In other words, although an essence is individual in the extra-mental world, it is not individual by itself or primitively but is rather individuated by something else. Thus, commonality or sharability in the sense specified turns out to be a fundamental modal property of an essence, i. e. a property which we discover by reflecting upon how an essence would exist if it could exist separately, i. e. in separation from the principle of individuation. Considerations about the modal property of an essence, in so far as they concern the (metaphysically) counterpossible existence of an essence, do not bear on the way an essence exists in actuality. Thus, being actually individual and being modally common seem to be compatible properties. In this sense, Aquinas’s endorsement of Avicenna’s doctrine of essence does not clash with his general analysis of the individuation of material substances.

Conclusions I suppose that my general reading of Aquinas’s doctrine of essence could be classified as realist as opposed to a more nominalist and deflationary one, in that I insist a good deal on the mind-independentness of the distinction between essence and supposit as well as on the intrinsic commonality and sharability of essence. In the history of philosophy, however, general labels are sometimes insidious and may even turn out to be not sufficiently powerful to capture the differences between one philosophical position and another. Thus, what is important is not so much whether Aquinas is a realist or not as which conceptual tools he employs in defending his particular version of realism. With regard to this, my general view is that the reconciliation between Avicenna’s doctrine of essence and Thomas’s view on individuation can be achieved by means of a set of three related distinctions: (i) The distinction between the level of the actual existence of an essence and the analysis of the metaphysical structure of a sensible substance.

Avicenna’s Doctrine of Essence and Aquinas’s View on Individuation 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

335

(ii) The consequent distinction between two senses of ‘common’: ‘actually common’ and ‘modally common’, i. e. common in the sense of not being primitively, but rather derivatively individual. Avicenna’s doctrine of essence establishes only that an essence is actually common only in the intellect, but does not prevent an essence from being extra-mentally common when ‘common’ is taken in the sense of ‘modally common’ – the sense which concerns the metaphysical analysis of material substances. (iii) Finally, distinctions (i) and (ii) presuppose a further distinction between abstraction and the modal considerations that guide us in the analysis of the metaphysical structure of material things. Abstraction is an automatic and natural process leading to the formation of universal concepts. Modal considerations, by contrast, are part of a deliberate philosophical investigation into the structure of things. The fact that abstraction and modal considerations are mental processes of so different a kind suggests that there is more to the extra-mental commonness of an essence than our intellect’s natural ability to understand material things without their individuating conditions. Reflection upon the process of abstraction and the way in which an essence exists in the mind may certainly be of some help in carrying through the metaphysical analysis of extra-mental objects. However, the modal considerations that shape our metaphysical analysis of extra-mental objects should not be confused with or reduced to the natural process by which we acquire concepts and so sort the world into different kinds of things.

Bibliography Primary Sources Aristotle, Metaphysics. A Revised Text with Introduction and Commentary, 2 vols, ed. W.D. Ross, Oxford: Clarendon Press, 1924. Averroes, Averrois Cordubensis in Aristotelis Metaphysicorum libros commentarium, in Aristotelis opera cum Averrois commentariis, vol. VIII, Venice: apud Junctas, 1562 – 74, reprinted Frankfurt a.M.: Minerva, 1962. Avicenna, Liber de philosophia prima sive scientia divina, I–IV, ed. S. Van Riet, Louvain/ Leiden: Peeters/Brill, 1977. –– , Liber de philosophia prima sive scientia divina, V–X, ed. S. Van Riet, Louvain/ Leiden: Peeters/Brill, 1980 Thomas Aquinas, Scriptum super librum Sententiarum, 2 vols, ed. R.P. Mandonnet, Paris: Lethellieux, 1929. –– , Liber de veritate catholicae fidei seu Summa contra gentiles, 3 vols, ed. C. Pera, Taurini-Romae: Marietti, 1961. –– , Summa theologiae, Rome: Editiones Paulinae, 1962 –– , In duodecim libros Metaphysicorum expositio, eds M.-R. Cathala and R. Spiazzi, Taurini-Romae: Marietti, 1964.

336 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Gabriele Galluzzo

–– , Quaestiones disputate, 2 vols, eds P. Bazzi, M. Calcaterra, T.S. Centi, E. Odetto, P.M. Pessiom and R. Spiazzi, Taurini-Romae: Marietti, 1965. –– , De substantiis separatis, cura et studio fratrum Praedicatorum, Opera omnia, t. XL, Opuscula I, Rome: Editori di S. Tommaso, 1969. –– , De ente et essentia, c. 3, cura et studio fratrum Praedicatorum, Opera omnia, t. XLIII, Opuscula IV, Rome: Editori di S. Tommaso, 1976. –– , De unitate intellectus contra Averroistas, cura et studio fratrum Praedicatorum, Opera omnia, t. XLIII, Opuscula IV, Rome: Editori di S. Tommaso, 1976. –– , Sententia libri de anima, cura et studio fratrum Praedicatorum, Opera omnia, t. XLV, Rome/Paris: Commissio Leonina/Vrin, 1984. –– , Expositio libri Peryermenias, cura et studio fratrum Praedicatorum, Opera omnia, t. I* 1, Rome/Paris: Commissio Leonina/Vrin, 1989. –– , Quaestiones disputate de anima, ed. B.-C. Bazn, Opera omnia, t. XXIV.1, Rome/ Paris: Commissio Leonina/Les ditions du Cerf, 1996. –– , Quaestiones de quolibet, 2 vols, cura et studio fratrum Praedicatorum, Opera omnia, t. XXV, Rome/Paris: Commissio Leonina/Les ditions du Cerf, 1996. –– , Quaestio de spiritualibus creaturis, ed. J. Cos, Opera omnia, t. XXIV 2, Rome/Paris: Commissio Leonina/Les ditions du Cerf, 2000. –– , On the Power of God, transl. and ed. English Dominican Fathers, 3 vols., London: Burns, Oates & Washbourne 1932–4 (repr. in one vol., Westminster, Maryland: Newman 1952).

Secondary Sources D. Black, Mental Existence in Thomas Aquinas and Avicenna, Documenti e studi sulla tradizione filosofica medievale, 10, 1999, pp. 45 – 79. R. Cross, The Metaphysics of the Incarnation, Oxford: Oxford University Press, 2002. A. De Libera, La querelle des universaux de Platon  la fin du Moyen Age, Paris: ditions du Seuil,1996. –– , L’art des gnralit. Thories de l’abstraction, Paris: Aubier, 1999. G. Galluzzo, Aquinas on Common Nature and Universals, Recherches de Thologie et Philosophie mdivales, 70, 2004, pp. 131 – 71. –– , Aquinas’s Interpretation of Aristotle’s Metaphysics, Book Zeta, Recherches de Thologie et Philosophie mdivales, 74(2), 2007, pp. 423 – 81. –– , Averroes and Aquinas on Aristotle’s Criterion of Substantiality, Arabic Sciences and Philosophy, 19, 2009, pp. 157 – 87. E.J. Lowe, A Survey of Metaphysics, Oxford: Oxford University Press, 2002. M.E. Marmura, Avicenna’s Chapters on Universals in the Isagoge of his Shifa’, in Islam: Past Influence and Present Challenge, eds. E.T. Welsh and P. Cacia, Albany, New York: State University of New York Press, 1979, pp. 34 – 6. –– , Avicenna on Primary Concepts in the Metaphysics of his al-Shifa’, in Logos Islamikos: Studia Islamica in honorem Georgii Michaelis Wickens, eds. R.M. Savory and D.A. Agius, Toronto: Pontifical Institute of Mediaeval Studies, 1984, pp. 219 – 39. –– , Quiddity and Universality in Avicenna, in Neoplatonism and Islamic Thought, ed. P. Morewedge, Albany, New York: SUNY Press, 1992, pp. 77 – 87. J. McGinnis, Logic and Science: The Role of Genus and Difference in Avicenna’s Logic, Science and Natural Philosophy, Documenti e studi sulla tradizione filosofica medievale, 18, 2007, pp. 165 – 86.

Avicenna’s Doctrine of Essence and Aquinas’s View on Individuation 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

337

J. Owens, Common Nature: A Point of Comparison Between Thomistic and Scotistic Metaphysics, Mediaeval Studies, 19, 1957, pp. 1 – 14. G. Pini, Absoluta consideratio naturae: Tommaso d’Aquino e la dottrina avicenniana dell’essenza, Documenti e studi sulla tradizione filosofica medievale, 15, 2004, pp. 387 – 438.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

On the Latin Reception of Avicenna’s Theory of Individuation Martin Pickav I It is well known that medieval philosophers and theologians in the Latin West debated vigorously over the principle of individuation, i. e. over the principle that accounts for the individuality of individuals and for the multiplication of things belonging to one species. The debate was especially lively at the end of the thirteenth and the beginning of the fourteenth centuries, a development no doubt fueled by the condemnation of certain aspects of Thomas Aquinas’s theory of individuation in 1277. During the controversy many items were proposed as candidates for the principle of individuation. These candidates include matter, form, matter and form together, quantity, relations, a negation, existence, haecceity, etc.1 One view that was often mentioned but never seriously defended during the height of the debate is the view that objects are individuated by their accidents or by the aggregation of their accidents. This theory, to which later medieval authors sometimes refer as the ‘old opinion’ (antiqua opinio)2, can be found prominently in Porphyry and Boethius. In his short treatise On the Trinity, Boethius, for instance, explains: The variety of accidents brings about numerical difference. For three men differ neither by genus nor by species but by their accidents. For even if we were to separate by the mind all their accidents, the locations of each of them are still different, which we cannot in any way make into one, because two bodies will not occupy one location, which is an accident. And therefore they are numerically many, because they become many through their accidents.3

And in Boethius’s translation of Porphyry’s Isagoge we can read the following:

1 2 3

See my The Controversy over the Principle of Individuation. See, e. g., text I.6 in the appendix. Boethius, De trinitate, c. 1, p. 168: ‘Sed numero differentiam accidentium varietas facit. Nam tres homines neque genere neque specie, sed suis accidentibus distant; nam vel si animo cuncta ab his accidentia separemus, tamen locus cunctis diversus est, quem unum fingere nullo modo possumus: duo enim corpora unum locum non obtinebunt, qui est accidens. Atque ideo sunt numero plures, quoniam accidentibus plures fiunt.’

340 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Martin Pickav

Things like this are called individuals because each consists of properties, the combination of which will never be the same in something else. For the properties of Socrates will never be found in any other particular thing. But the properties of a human being (I mean those that belong to a human being as something common) will be the same in many, or rather, they will be the same in all particular human beings insofar as they are human beings.4

The motivation for such a theory seems to derive from the fact that we indeed identify individuals by their different sets of accidents. I identify two human beings as two different individuals because each of them is at a different location, has a different size, accent, clothing, etc. But it is less clear that we are entitled to move from this fact about how we discriminate between individuals to a deeper ontological claim concerning their individuation. However, on the other hand it is plausible to say that each individual object possesses some properties that it shares with other objects of the same kind and some properties that belong solely to itself; surely an object is not individuated by any of the common properties. The Porphyrio-Boethian account of individuation was very influential in Late Antiquity and in the Early Middle Ages.5 But whether accidental individuation is a promising solution to the problem of individuation depends largely on the ontological framework in which it is presented. It also depends, of course, on whether accidental individuation can be formulated in an unambiguous way. In the Boethius passage quoted above, for instance, it remains open whether individuation is caused by one accident alone (spatial location) or by a combination or bundle of all of its accidents. Moreover, are the individuating accidents particular properties or are they common properties which themselves need to be individuated by something else? However this may be, the ‘old opinion’ looks extremely vulnerable if it is understood on the background of a strictly Aristotelian understanding of accidents as entities that are ontologically dependent on the substances in which they inhere. With the Aristotelian background in mind medieval philosophers frequently advance one or more of the following six arguments against the idea of accidental individuation: 4

5

Porphyry, Isagoge II § 15 (translatio Boethii, pp. 13 f.; cf. versio Graeca, p. 7): ‘Indiuidua ergo dicuntur huiusmodi quoniam ex proprietatibus consistit unumquodque eorum quorum collectio numquam in alio eadem erit; Socratis enim proprietates numquam in alio quolibet erunt particularium; hae uero quae sunt hominis (dico autem eius qui est communis) proprietates erunt eaedem in pluribus, magis autem et in omnibus particularibus hominibus in eo quod homines sunt.’ See also Boethius’s commentary on this passage in In Isagogen Porphyrii (ed. sec.) III, c. 11, pp. 253 f. and IV, c. 1, pp. 241 f. See Chiaradonna, La teoria dell’individuo in Porfirio; Erismann, Collectio proprietatum; id., L’individualit explique par les accidents; Gracia, The Problem of Individuation in the Early Middle Ages; King, The Problem of Individuation, pp. 163 – 7; Sorabji, Self, pp. 137 – 53.

On the Latin Reception of Avicenna’s Theory of Individuation 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

341

(1) Each individual substance is a per se being (ens per se). But if a substance owes its individuality to the reception of an accident or an aggregation of accidents, then the individual would essentially be a composite of a substance and an accident and thus merely a being per accidens. In other words: accidental individuation makes all individuals into composites and takes away the essential unity individual substances seem to enjoy.6 (2) Substances are ontologically prior to their accidents. For under normal circumstances accidents exist only insofar as – and because – they inhere in substances. But if substances are ontologically prior to accidents, then it also follows that an individual substance is ontologically prior to its particular accident. And this rules out that accidents are capable of accounting for the individuality of that on which they depend.7 (3) Individual (or numerical) unity is a property that belongs to every existing thing by itself. For if numerical unity were to come about by something added, say, by the addition of an accident, then one could ask about that added thing whether or not it has numerical unity by itself or from something else. If the latter is the case, we are at risk of an infinite regress, for then we have to ask with regard to this other thing whether it has numerical unity by itself or from something else. If, however, the former is true then one might wonder why the accident is numerically one, but the object to which it is added is not.8 (4) In each category (or class of being) there are genera, species, and individuals. Now it would destroy the division between the categories if a species (say, of the category of substance) requires, in order to be a species (instead of being a genus), some element from a different category. The same reasoning applies to individuals belonging to each of the categories.9 (5) Since accidents are in themselves common, they are incapable of turning something into a particular this. An aggregation of many different accidents will 6

7

8 9

See, e. g., Peter John Olivi, Quodlibet III, q. 4, pp. 177 f.; John Duns Scotus, Quaestiones super libros Metaphysicorum Aristotelis, lib. VII, q. 13, n. 21, pp. 224 f.; id., Lectura II, dist. 3, p. 1, q. 4, n. 65, p. 247; id., Ordinatio II, dist. 3, p. 1, q. 4, n. 70, pp. 422 f. Here and in the following footnotes I only mention a few examples of authors presenting the various arguments. Basically all the authors in part I of the appendix (and many of part II) bring forward one or more of the following reasons against accidental individuation. See, e. g., Godfrey of Fontaines, Quodlibet VII, q. 5, p. 320; Peter John Olivi, Quodlibet III, q. 4, p. 177; John Duns Scotus, Quaestiones super libros Metaphysicorum Aristotelis, lib. VII, q. 13, nn. 24 – 6, pp. 225 – 7; id., Lectura II, dist. 3, p. 1, q. 4, n. 79, pp. 252 f.; id., Ordinatio II, dist. 3, p. 1, q. 4, nn. 82 – 8, pp. 429 – 33. See, e. g., John Duns Scotus, Quaestiones super libros Metaphysicorum Aristotelis, lib. VII, q. 13, n. 27, p. 227; id., Lectura II, dist. 3, p. 1, q. 4, nn. 73 – 8, pp. 250 – 52; id., Ordinatio II, dist. 3, p. 1, q. 4, nn. 76 – 81, pp. 426 – 9. See, e. g., John Duns Scotus, Quaestiones super libros Metaphysicorum Aristotelis, lib. VII, q. 13, nn. 28 – 30, pp. 227 f.; id., Lectura II, dist. 3, p. 1, q. 4, nn. 91 – 4, pp. 258 f.; id., Ordinatio II, dist. 3, p. 1, q. 3, n. 63, pp. 419 f. and ibid., q. 4, nn. 89 – 98, pp. 433 – 8.

342 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Martin Pickav

of course restrict the commonality of the set of properties, but it will never be able to rule out that there can, at least in principle, be another item instantiating exactly the same set of properties. Individuals, on the other hand, can by definition never be instantiated more than once.10 (6) Two or more individuals differ from each other as substances; in other words: they differ substantially. Plato is substantially different from Socrates. But accidents can only account for an accidental difference and therefore fail to account for the robust difference between individuals.11 Some of these arguments against accidental individuation were first formulated by Peter Abelard in the first half of the twelfth century.12 And Abelard’s criticism seemed to have had the effect that ‘no medieval philosopher argued seriously for accidental individuation after Abelard proposed his objection.’13 All this is to say that in the thirteenth and the fourteenth centuries the theory of accidental individuation was generally considered a bad idea and something with which most people preferred not to be associated. The latter is in particular true for defenders of the view according to which material objects are individuated by quantity, i. e. the quantitative dimensions of the underlying matter. Although proponents of quantitative individuation are often attacked by their opponents using one or more of the arguments mentioned above, these proponents are eager to disassociate themselves from the accidental individuation view.14 The quantity involved in individuating material objects, so they argue, is not an accident in the common sense.

10 See, e. g., Peter of Auvergne, Quodlibet II, q. 5, edited in Hocedez, Une Question indite de Pierre d’Auvergne, p. 371; John Duns Scotus, Quaestiones super libros Metaphysicorum Aristotelis, lib. VII, q. 13, n. 32, p. 229. 11 See, e. g., Bonaventure, In II Sententiarum, dist. 3, p. 1, a. 2, q. 2, p. 106a; Godfrey of Fontaines, Quodlibet VII, q. 5, p. 321; Peter John Olivi, Quodlibet III, q. 4, p. 177. 12 See King, Metaphysics, pp. 72 – 5. 13 King, The Problem of Individuation, p. 167. The fact that no one positively endorsed accidental individuation after Abelard is perhaps the true reason why later (13th- and 14th-century) authors refer to this position as the ‘old opinion.’ 14 Thomas Aquinas is somewhat an exception. In his Super Boetium de trinitate he explicitly argues that one accident, i. e. quantity, does indeed account for individuation (q. 4, a. 2, pp. 122 – 6). But he also explains that the individuating quantity is not an accident in the complete sense. For more detail on this see Wippel, The Metaphysical Thought of Thomas Aquinas, pp. 351 – 75. Another exception is Albert the Great (at least in his earlier works). For Albert, individuals are ‘individuated’ by their accidents, but since each individual is also a substance and a subject, it is also true that the singularity of the substance or subject is due to a more profound principle, which is not an accident nor an aggregation of accidents. In other words, by using the notion ‘individual’ in a slightly peculiar way (namely for something that is already composed of a substance and accidents), Albert arrives at a harmonizing interpretation according to which both the proponents and the opponents of accidental individuation come out to be correct. For

On the Latin Reception of Avicenna’s Theory of Individuation 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

343

II At the beginning of the fourteenth century it is common to find references to Avicenna as one of the proponents of accidental individuation. Good examples are the various texts in which John Duns Scotus deals with the problem of individuation. In many of them Avicenna appears next to Porphyry and Boethius (see texts I.7, I.8, and I.9 in the appendix). However, there is no reason to believe that Duns Scotus was the first to consider Avicenna’s theory of individuation as basically the same as the Porphyrio-Boethian theory. The editors of Duns Scotus’s Quaestions on the Metaphysics notice that Scotus’s arrangement of philosophical authorities is very likely based on the arrangement Scotus found in the discussion of the principle of individuation in q. 5 of Godfrey of Fontaines’s seventh Quodlibet. There too we find Avicenna next to Porphyry and Boethius (see text I.5 in the appendix). But it is easy to find other medieval authors who present Avicenna in the same way: Peter of Falco (I.1), Henry of Ghent (I.2), Richard Middleton (I.3), Roger Marston (I.4), John of Paris (I.10), and Peter of Auvergne (I.6). I have no reason to believe that this list is complete. But are these later medieval philosophers correct in considering Avicenna as a proponent of accidental individuation? The Latin Avicenna no doubt uses the language of accidents when he talks about individuation. One passage that is often quoted by the authors mentioned above is from chapter 4 of the fifth treatise of Avicenna’s Liber de philosophia prima: And then the nature [of a thing] will be such that necessary concomitants befall it, necessary concomitants consisting of properties and accidents through which the nature is individuated and becomes designated.15

Another passage popular among Latin authors can be found a little bit earlier in chapter 2: A nature that requires matter does not possess existence unless the matter has been prepared [for its reception]; accidents and dispositions thus befall its existence from the outside, and through these accidents and dispositions the nature is individuated.16 Albert’s account of individuation see Roland-Gosselin, Le ‘De ente et essentia’ de S. Thomas d’Aquin, pp. 89 – 103. 15 Avicenna, Liber de philosophia prima, tr. 5, c. 5, p. 264: ‘Et erit tunc natura sic quod accident ei concomitantia ex proprietatibus et accidentibus, per quae natura individuatur et fit designata.’ For Latin authors who quote this passage see texts I.2, I.5, I.6 and I.10 in the appendix. 16 Ibid., c. 2, p. 240: ‘Quae vero ex istis naturis eget materia, non habet esse nisi cum materia fuerit praeparata; unde ad eius esse adveniunt accidentia et dispositiones extrinsecus per quae individuatur.’ Scotus refers to this passage in texts I.7, I.8 and I.9 in the appendix. See also Liber de philosophia prima, tr. 3, c. 3, p. 117: ‘Dico igitur quod

344 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Martin Pickav

Both texts are taken from Avicenna’s famous account of essences in which he develops his idea that quiddities or essences can be considered in three ways: (1) absolutely and in themselves, (2) insofar as they are in the mind, and (3) insofar as they exist outside the mind.17 Insofar as essences exist in the mind they have the property to be universal; insofar as they exist in things outside the mind, an essence is singular. To be universal or to be singular are thus accidental with respect to the essence insofar as the essence is considered in itself. But it is one thing to say that to be universal and to be singular is accidental for an essence and it is another thing to say that essences and the objects possessing such essences are individuated by accidents. There is yet another, albeit closely related, reason suggesting that Avicenna endorses accidental individuation. This reason has to do with Averroes’s criticism of Avicenna’s account of unity. In his Long commentary on Aristotle’s Metaphysics, Averroes famously accuses Avicenna of teaching that unity is an accidental feature. Or in medieval terms: Averroes reprimands Avicenna for confusing transcendental unity with the unity which belongs to the category of quantity, i. e. the unity which is the principle of number.18 But if unity is supposed to be an accidental feature, so is individuality, because individuality is apparently nothing else than a specific kind of unity. Latin authors discuss Averroes’s criticism frequently, and although they usually defend Avicenna from the accusations, they are by the same route familiar with this particular reading of Avicenna’s account of unity and individuation.19 However, there are also serious reasons to be doubtful as to whether Avicenna really thought that objects are individuated by accidents or by an aggregation or bundle of accidents. In the following text from the Liber de philosophia prima Avicenna anticipates one of the above-mentioned standard objections to accidental individuation (objection 5): unitas vel dicitur de accidentibus vel dicitur de substantia; cum autem dicitur de accidentibus, non est substantia, et hoc est dubium; cum vero dicitur de substantiis, non dicitur de eis sicut genus nec sicut differentia ullo modo: non enim recipitur in certificatione quidditatis alicuius substantiarum, sed est quiddam comitans substantiam, sicut iam nosti. Non ergo dicitur de eis sicut genus vel sicut differentia, sed sicut accidens. Unde unum est substantia, unitas vero est intentio quae est accidens.’ 17 For a detailed analysis of the argument of Liber de philosophia prima, tr. 5, chapters 1 – 4 see de Libera, L’art des gnralits, 499 – 607. 18 Averroes, Long Commentary on Aristotle’s Metaphysics, lib. IV, com. 3, in Aristotle, Opera cum Averrois commentariis, vol. 8, fol. 67rB, D–E: ‘Avicenna autem peccavit multum in hoc, quod existimavit, quod unum et ens significant dispositiones additas essentiae rei … Et etiam, quia existimavit, quod unum dictum de omnibus praedicamentis, est illud unum quod est principium numerorum. Numerus autem est accidens. Unde opinatus fuit iste, quod hoc nomen unum significat accidens in entibus.’ 19 For some aspects of the medieval discussion triggered by Averroes’s criticism see de Libera, D’Avicenne  Averros, et retour.

On the Latin Reception of Avicenna’s Theory of Individuation 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

345

If A is a universal intention and B, another universal intention, is added to it, then there results from this some restriction. But because this is the restriction of one universal by some other universal, there remains a more restricted universal that can nevertheless be shared by others. If, for example, this person Socrates is defined and it is said that he is a philosopher, then there is sharing with others. And if it is said that he is a chaste philosopher, there is still sharing with others, and if it is said that he is a chaste philosopher who was unjustly killed, even this is still shared with others. … And if it is added that he was killed in this city at this day, then this whole enumeration with its specific uniqueness is still completely universal; for it is possible to predicate this enumeration of many things, unless it is restricted to one individual.20

Strictly speaking, Avicenna is here only concerned with the definability of individuals. His point is that it is impossible to define an individual by simply enumerating all of his properties, for although such an enumeration becomes less and less universal with each property added, it never yields a description that necessarily fits only one individual. But the point made can easily be translated into a metaphysical claim: an aggregation of (common) accidents never renders a substance individual. And some medieval readers read the above-mentioned passage exactly in this metaphysical sense.21 In fact, when Avicenna explicitly addresses his own account of individuation he clearly presents himself as a defender of the traditional Aristotelian view according to which objects are essentially individuated through matter.22 This is 20 Avicenna, Liber de philosophia prima, tr. 5, c. 5, pp. 276 f.: ‘Cum enim a fuerit intentio universalis et ei adiungitur b, alia scilicet similiter intentio universalis, poterit tunc hic esse restrictio aliqua. Sed quia est restrictio unius universalis per aliud universale, remanet postea propinquius universale in quo potest esse communio. Verbi gratia, cum definitur hic Socrates et dicitur esse “philosophus”, in hoc etiam communio est; cum vero dicitur “philosophus castus”, adhuc etiam communio est; si vero dicitur “philosophus castus qui occisus fuit iniuste”, adhuc etiam hoc commune est. … Si vero additur “quem occiderunt in civitate illa et in die illo”, haec etiam tota enumeratio cum sua individualitate omnino universalis est; possibile est enim illam praedicari de multis, nisi restrinxeris circa individuum.’ A similar argument can also be found in the Isagoge part of Avicenna’s Shifa’, chapter I, 12; see Marmura, Avicenna’s Chapter on Universals in the Isagoge of his Shifa’, pp. 50 f. 21 See Peter of Auvergne, Quaestiones in Metaphysicam, lib. VIII, q. 25, edited in Hocedez, Une Question indite de Pierre d’Auvergne, p. 381: ‘… et quotcumque accidentia accedant ad hoc, totum aggregatum commune est, sicut dicit Avicenna, quia si componentia sunt communia, et compositum. Manifestum ergo quod accidens non est causa individuationis.’ Peter of Auvergne is a strange case. Why does he consider Avicenna as a proponent of accidental individuation in his later quodlibetal questions (see text I.6 in the appendix) when he is aware in his earlier Metaphysics commentary that this cannot be Avicenna’s position? The puzzle can maybe explained with the peculiar character of the quodlibetal question, which is basically a paraphrasis of Godfrey of Fontaines’s view on individuation. 22 For Avicenna’s theory of individuation see Bck, The Islamic Background; Black, Individuation, Self-Awareness, and God’s Knowledge of Particulars.

346 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Martin Pickav

true even for immaterial objects like souls. In the following passage from his Liber de anima he writes: The multiplicity of things either derives from the essence and the form; or it derives from the relation to matter and to the multiple origin from places which surround every matter with respect to something else; or the multiplicity derives from the times proper to the each of the things that befall it with its accidents; or it derives from the causes which divide matter. But souls are not different in essence and form, for they have the same form. Their difference is therefore due to the recipient of their essence to which the essence is related as its own, and this is the body. But if the soul were to exist without the body, then it would be impossible that one soul differs from another in number. And this applies generally to all things. For in the case of things which are many and whose essences are mere intentions, their species have been multiplied in their individuals, and their multiplicity exists only on account of their subjects and their recipients and the things that are affected by them or it exists on account of some relation to these or to their times.23

Although Avicenna here strictly speaking only says that souls are individuated and multiplied by their relationships to bodies, it is obvious that bodies account for multiplication qua being material and not qua possessing yet another form, a forma corporeitatis. For then the question of individuation would just reappear at a different level. Note also that Avicenna explicitly says that this account of individuation applies ‘generally to all things’. All this calls for a conciliatory interpretation of the respective roles matter and accidents play in Avicenna’s theory of individuation. The picture supported by the text seems to be the following : Matter is the primary principle of individuation, because matter individuates forms and thus also form-matter composites. But once the individual forms or the individual form-matter composites exist they will immediately receive a variety of accidents, which can in a broader sense also to be said to contribute to their individuation.24 For it is through such accidents that we identify (and discriminate among) different 23 Avicenna, Liber de anima, tr. 5, c. 3, pp. 105 f.: ‘Multitudo enim rerum aut est ex essentia et forma, aut est ex comparatione quae est ad materiam et originem multiplicatam ex locis quae circumdant unamquamque materiam secundum aliquid aut ex temporibus propriis uniuscuiusque illarum quae accidunt illis accidentibus, aut ex causis dividentibus illam. Inter animas autem non est alteritas in essentia et forma: forma enim earum una est. Ergo non est alteritas nisi secundum receptibile suae essentiae cui comparatur essentia eius proprie, et hoc est corpus. Si autem anima esset tantum absque corpore, una anima non posset esse alia ab alia numero. Et hoc generaliter est in omnibus; ea enim quorum essentiae sunt intentiones tantum et sunt multa, quorum multiplicatae sunt species in suis singularibus, non est eorum multitudo nisi ex sustinentibus tantum et receptibilibus et patientibus ex eis, aut ex aliqua comparatione ad illa aut ad tempora eorum.’ On this text see Druart, The Human Soul’s Individuation; Sebti, Avicenne, pp. 25–36. For a list of medieval authors drawing on Avicenna’s theory in Liber de anima, tr. 5, c. 3 see Hasse, Avicenna’s ‘De anima’ in the Latin West, p. 297. 24 This seems to be exactly what is said in the text in n. 16 above.

On the Latin Reception of Avicenna’s Theory of Individuation 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

347

individuals in the first place. Yet strictly speaking, accidents – whether they come alone or in bundles – are for Avicenna not responsible for individuation.25

III If, however, Avicenna himself never held the view that objects are primarily individuated by accidents or bundles of accidents, does this mean that all his Latin readers in the thirteenth and fourteenth centuries got him completely wrong? By no means. There are actually many authors who are quite aware that Avicenna never defended accidental individuation. William of Auvergne and John of La Rochelle, for example, correctly ascribe to Avicenna the doctrine that objects are primarily individuated by matter (see texts II.1 and II.2 in the appendix)26. Moreover, despite the impression created by the texts in the first part of the appendix, most authors never mention Avicenna when they refer to accidental individuation. It might be enough here to point at some examples: Bonaventure (II.4), Roger Bacon (II.5), Robert Kilwardby (II.6), William de la Mare (II.7), Giles of Rome (II.8), and Peter John Olivi (II.9). Bacon and Giles are especially interesting, because they seem to be keen on adding names to the list of the traditional bundle theorists Porphyry and Boethius; Bacon adds al˙ aza-lı¯’s name to his list, Giles mentions Simplicius. Why don’t they also add G Avicenna? Yet the main question with which we are confronted is this: What moves authors such as Henry of Ghent, Roger Marston, Godfrey of Fontaines, and John Duns Scotus to regard Avicenna as an accidental individuation theorist? And why do they seem to adopt an interpretation of Avicenna that is obviously wrong? A possible solution to this puzzle may appear once we have discovered the origin of the strange association of Avicenna, Porphyry and Boethius. Avicenna, however, is not the only one whom the aforementioned authors associate with Porphyry and Boethius; almost all the passages in question also mention John Damascene. Now it is well known that both Avicenna and John Damascene play a very prominent role in question 8 of Henry of Ghent’s second Quodlibet (disputed in the Advent of 1277). In this quodlibetal question Henry reacts to Aquinas’s recently condemned view according to which matter is the sole principle of individuation.27 The text has a peculiar structure: In his 25 John of La Rochelle (text II.2 in the appendix) is a good medieval example for this interpretation of Avicenna’s theory of individuation. 26 See also text II.3 from Peter of Tarantaise’s Sentences commentary. 27 Henry mentions the condemned articles explicitly. See Quodlibet II, q. 8, p. 45. On this text see also Brown, Henry of Ghent; Aertsen, Die Thesen zur Individuation in der Verurteilung von 1277; Suarez-Nani, Les anges et la philosophie, pp. 75 – 85.

348 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Martin Pickav

reply Henry first turns to ‘the Philosopher’ (i. e., Aristotle) and rejects what he has to say (or not to say) about individuation of immaterial substances, then Henry moves on to ‘our philosophizers’ (nostri philosophantes), followers of Aristotle’s account of individuation in Henry’s time, who are similarly criticized, and finally he turns to ‘our saints’ (nostri sancti), who adhere, according to Henry, to the correct theory of individuation. Because they consider matter as the only principle of individuation, ‘the Philosopher’ and ‘the philosophizers’ hold that without matter there cannot exist multiple individuals belonging to one and the same species. ‘Our saints’, such as John Damascene, teach on the other hand that there can be different immaterial entities belonging to one species and that they are distinct from each other by means of ‘characteristic properties’ (characteristicae proprietates). To illustrate John’s point Henry quotes a longer passage from John’s Elementary Introduction to Dogma: Each thing through which a species differs from another species is called substantial and natural and constitutive difference and natural quality and natural property and property of nature. And so human being and ox differ from each other, because a human being is rational, but an ox is irrational; and ‘rational’ is the substantial and constitutive difference of human being, but ‘irrational’ is that of ox. The same applies to the other species, substances and natures and forms. However, each thing through which a hypostasis of the same species differs from a co-substantial hypostasis is called adventitious difference and quality and hypostatic and characteristic property. But this is an accident.28

To see John Damascene in this context is not surprising; authors of the thirteenth and fourteenth centuries quote him frequently as an authority when they deal with the nature of angels and immaterial substances in general.29 It is 28 Henry of Ghent, Quodlibet II, q. 8, pp. 47 f.: ‘Quomodo ergo et penes quid accipitur in immaterialibus et incorporalibus individuatio et specierum distinctio, Damascenus iam tetigit cum dixit: “characteristicis proprietatibus”. Quid autem appellet “characteristicam proprietatem”, exposuit prius ubi dixit in eodem libro: “Omnis res qua differt species ab altera specie, ‘substantialis’ et ‘naturalis’ et ‘constitutiva differentia’ dicitur et ‘qualitas naturalis’ et ‘naturalis proprietas’ et ‘proprium naturae’, veluti differunt homo et bos ad invicem, quoniam homo quidem rationalis est, bos autem irrationalis, et est rationale substantialis differentia et constitutiva hominis, irrationale autem bovis. Similiter et in reliquis speciebus, substantiis et naturis et formis. Omnis autem res, in qua differt hypostasis ab eiusdem speciei et consubstantiali hypostasi, dicitur ‘adventitia differentia’ et ‘qualitas’ et ‘hypostatica’ et ‘characteristica proprietas’. Haec autem est accidens.”’ For the Latin text of De institutione elementari (in Robert Grosseteste’s translation) see H. Gravius’s edition of John’s works. 29 John’s appearance may also help to explain why Henry contrasts nostri sancti with the philosophus and the philosophantes. In another text, his Dialectica, John refers to the teaching of the ‘holy Fathers’ (sancti patres), when he explains that there is an angelic species under which we can find a multitude of individuals. See Dialectica (translatio Roberti Grosseteste), c. 11, p. 13. For other references to sancti patres in this context see

On the Latin Reception of Avicenna’s Theory of Individuation 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

349

with the aim of providing an interpretation of John Damascene’s account of individuation that Avicenna enters the picture at this point. One of the first things Henry remarks is that John and Avicenna use the term ‘accident’ in the same broad sense as referring to something that affects essences and is outside the natures of the essences themselves.30 So to talk about accidental individuation in this sense of accident means not that something is individuated by an item belonging to some of the accidental categories (i. e., quantity, quality, relation, location etc.), but only that the individuating factor is not part of the essence of the individuandum. But what is then the ‘accident’ causing the individuation of an immaterial substance (and of substances in general), if it is not one or more items belonging to the accidental categories? For Henry the principle of individuation is nothing other than the subsistence (subsistentia) of the substance, i. e. the special mode of being that a substance enjoys when it exists in reality. At the end of his response he summarizes his view in the following way: And so it has to be said that (if the essence of an angelic nature or the essence of whatever creature does not oppose this) God can make that two individuals – also two that are separated from any accident (if it is possible at all for a substance to exist without any accident) – exist under one ultimate species, so that the two individuals, in what is substantial to them alone, are distinct through diverse subsistences in effect.31

Let me make two comments on this quotation and the picture emerging from it: (1) As the talk of ‘subsistence in effect’ (subsistentia in effectu) on the one hand and God on the other hand indicates, Henry distinguishes between remote and proximate (or precise) causes of individuation. Subsistence is the proximate cause, but the agent by whose action a substance exists in effect is also a cause or principle of individuation.32 And because material substances exist or subsist in matter, matter too can be called a principle of individuation. Aristotle and the ibid., c. 30, pp. 27 f. The passages from the Dialectica will show up prominently in William de la Mare’s Correctorium fratris Thomae (written between 1277 and 1282, i. e., at the same time or briefly after Henry’s quodlibetal question). See q. 11, edited in Glorieux, Les premires polmiques thomistes, pp. 60 f. William ends his attack on Aquinas with the line (ibid.): ‘Ecce quod angelus est species specialissima, tam secundum doctrinam philosophorum quam secundum doctrinam Sanctorum Patrum.’ Although Henry does not mention the Dialectia passage(s) in his quodlibetal question, this common emphasis on John Damascene and the views of the ‘saints’ can be no coincidence. 30 Henry of Ghent, Quodlibet II, q. 8, p. 48. 31 Ibid., p. 53: ‘Sic ergo dicendum quod, non repugnante essentia naturae angelicae vel cuiuscumque creature essentia, duo individua, etiam absoluta ab omni accidente (si tamen possibile sit substantiam aliquam creatam existere sine omni accidente), possunt fieri a Deo sub una specie specialissima, in solis sibi substantialibus distincta per diversas subsistentias in effectu.’ 32 In this sense, Henry calls God the ‘first principle of individuation’, see ibid., p. 51.

350 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Martin Pickav

‘philosophizers’ therefore were wrong not in that they regarded matter as a principle of individuation, but that they regarded matter as the only principle of individuation.33 (2) Although Henry refers to subsistence as an ‘accident’ in the broad sense (for to subsist as this or that individual is accidental to the essence), he clearly distinguishes between different meanings of ‘accident’. Accidents in the strict sense, i. e. really distinct properties inhering in substances, accidents to which Henry refers as ‘posterior accidents’ (accidentia posteriora), are not capable of individuating substances; they allow us only to identify individuals and discriminate among them. Finally, Henry quotes Porphyry as someone who really only talks about accidents and bundles of accidents as principles for identifying individuals.34 In summary, according to Henry, the principle of individuation is something outside of the essence or form of the individuated object. In this sense, and in this sense alone, the principle of individuation is an accident. But because immaterial substances consist of more than just their essences, a multitude of individuals under the same species is possible. With respect to John Damascene and Avicenna, Henry intends to create the impression that his account of individuation and of the quasi-accidental nature of the individuating principle is shared by the two authorities. Yet insofar as he completely suppresses any indication of the crucial role matter plays in Avicenna’s account of individuation, Henry somewhat ‘misinterprets’ Avicenna. Yet I think there is an obvious explanation for this ‘misinterpretation’. As it is not surprising to see John Damascene quoted in a text about angelic individuation, so it is not surprising to see Avicenna used in the same context. But usually the two authorities appear on opposing sides; Avicenna is normally considered as one of those philosophers who deny that without matter there can be a multiplicity of individuals belonging to one species, whereas for John such a multitude is possible. In other words, Avicenna is commonly regarded as a proponent of the very theory Henry rejects.35 One of Henry’s implicit goals in 33 See also ibid., pp. 47 and 51. 34 Ibid., ad arg., pp. 54 – 5 and 57. For more on Henry’s theory of individuation see my Henry of Ghent on Individuation. 35 See, e. g., Peter of Tarentaise (Pope Innocent V), In II Sententiarum, dist. 3, q. 2, a. 1, p. 34b: ‘Non est simile, quia animae ut dicit Avicenna individuantur per materias corporum, quamvis ab eis separatae retineant individuationem, sicut cera impressionem sigilli; vel saltem per relationem inclinationis naturalis ad corpus determinatum; angeli vero non sic habent materiam per quam individuentur’; Thomas Aquinas, De ente et essentia, c. 3, p. 376b: ‘Sed cum essentia simplicis non sit recepta in materia, non potest ibi esse talis multiplicatio; et ideo oportet ut non inueniantur in illis substantiis plura indiuidua eiusdem speciei, sed quot sunt ibi indiuidua, tot sunt ibi species, ut Avicenna expresse dicit’; Thomas Sutton, Quaestiones ordinariae, q. 27, pp. 752 f.: ‘Ergo natura specifica angeli nullo modo multiplicatur. Istam rationem ponit Avicenna sub his verbis: “Ex his autem naturis, quae non eget materia ad permanendum vel ad incipiendum, si

On the Latin Reception of Avicenna’s Theory of Individuation 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

351

Quodlibet II, q. 8 is to give what he considers the correct interpretation of Avicenna’s account of individuation. According to Henry of Ghent, ‘the Philosopher’ and ‘the philosophizers’ commit a simple mistake in claiming that the absence of matter precludes multiplication of members under one and the same species. For their conclusion would only follow if the essence of an immaterial angelic species were capable of existing by itself. Then there would indeed be no room at all for something that could cause their individuation. If on the other hand such an essence does not exist by itself, but requires something further to exist actually, then there is no reason to believe that the mere absence of matter renders such a multiplication impossible. In Henry’s words, the proponents of the traditional Aristotelian account of individuation are mistaken, for they consider immaterial substances as identical with their essences and thus as necessary beings, ‘many gods’. But as Avicenna has noticed correctly, the essences of immaterial substances are indifferent with regard to their actual existence and they are therefore also indifferent with respect to their existence as individuals. For this reason, Henry holds Avicenna to be committed to the view that immaterial substances do allow for multiple members of one and the same species.36 In his quodlibetal question Henry never explicitly addresses how to interpret Avicenna. But this is what he no doubt would have said if he had been asked about the correct way of understanding Avicenna’s teaching on individuation: Although Avicenna sometimes mentions the role matter plays in individuation, matter is not the exact principle of individuation, because it doesn’t apply to all beings. Immaterial beings are also capable of individuation because, as in all creatures, their essences are not identical with their actual existence, and it’s the latter that is responsible for their individuation.

IV Whatever we may think about Henry of Ghent’s reading of Avicenna, we still seem far away from the theory of accidental individuation. For one thing is clear: Henry does not consider Avicenna as someone for whom objects are est, impossibile est eam multiplicari, et species huiusmodi est una numero”, quia scilicet non multiplicatur in pluribus suppositis secundum numerum …’. 36 Henry of Ghent, Quodlibet II, q. 8, p. 39: ‘Quod autem non ex se sed solum ab alio agente singulare est in supposito subsistens, quia ex se nulli appropriatur et est essentia tantum, quantum est ex se, indifferenter natum est esse singulare, subsistendo in unico supposito, vel universale, subsistendo in pluribus. Quod etiam bene dicit Avicenna et determinat in Vo Metaphysicae suae. Ex quo sequitur apertissime quod necesse est ut non sit essentia creaturae, in quantum creatura est, quin possit, quantum est ex se, in plura individua multiplicari, quantumcumque sit abstracta a materia.’ For the ‘many gods’ see ibid., p. 41.

352 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Martin Pickav

individuated by accidents in the proper sense. But a look at another of Henry’s quodlibetal question shows us that we are not too far away. A couple of years later Henry comes back to the topic of individuation in question 8 of his fifth Quodlibet (disputed either Advent 1280 or Lent 1281). This time the setup of his response is more traditional; after developing what the principle of individuation is supposed to explain, he begins by presenting accidental individuation as one attempt to account for individuation. From the criticism of accidental individuation he moves on to the Aristotelian view and its deficiencies before he finally presents his own account. The way in which Henry presents accidental individuation merits our attention because he adds John Damascene and Avicenna to the traditional authorities Porphyry and Boethius (see text I.2 in the appendix). However, there is no indication that Henry changed his mind regarding Avicenna. Since he still explains that Avicenna’s use of the term ‘accident’ in the context of individuation should not be understood in the narrow sense, there is no reason to believe that he forgot his earlier interpretation.37 It must thus be with full awareness that Henry places Avicenna next to Porphyry and Boethius in the setup of his response. I can only think of three possible reasons why Henry suddenly associates Avicenna with the defenders of accidental individuation. (1) Maybe he simply wanted to create a certain rhetorical effect. Referring not only to Porphyry and Boethius but also to John Damascene and Avicenna lends accidental individuation more credence and makes it more reasonable to discuss this view first. (2) Associating Avicenna with accidental 37 In Quodlibet V, q. 8, Henry describes his own candidate for the principle of individuation as an accident. And he mentions the specific meaning the term ‘accident’ can have in this context. See ibid., fol. 166rM: ‘Et quia talis negatio sive determinatio, licet nihil rei apponit in separatis supra essentiam qua determinetur, inquantum tamen est extra intentionem formae, ut forma est, large sumendo accidens potest dici accidens, ut secundum hoc huiusmodi individuatio dicatur fieri per aliquod accidens formae superveniens. … Unde Avicenna assignans differentiam inter accidens quo individuantur formae separatae et formae materiales generabiles et corruptibiles, dicit quinto Metaphysicae: Postquam determinata est species specialissima accidunt ei concomitantia ex accidentibus et proprietatibus per quae natura individuatur, et fit designata. Ipsae vero proprietates et accidentia aut erunt adventitia tantum, ita ut non sunt de essentia ullo modo, et haec sunt accidentia quae accidunt individuis rerum simplicium, aut erunt dispositiones superadditae …’ The same is true for his treatment of individuation in Quodlibet XI, q. 1, fol. 438vQ: ‘Essentia sive natura sive forma specifica ex se non est nisi ipsa … cetera autem omnia sunt ei aliquo modo accidentalia, cuiusmodi sunt esse in anima sive in intellectu ut in cognoscente et esse in supposito existente extra animam. Sed etiam accidit ei quodammodo esse in ipso supposito, secundum quod hoc alias saepius exposuimus secundum Avicennam … Quod quidem esse determinatum non est vere accidens, licet sit extra rationem speciei. Determinatio necessario fit per aliquid. Si enim illud esset vere accidens, tunc suppositum sub specie non esset unum nisi per accidens et nullum per se quod falsum est.’

On the Latin Reception of Avicenna’s Theory of Individuation 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

353

individuation creates almost the strongest possible contrast with the traditional interpretation of Avicenna as a defender of material individuation. (3) Henry’s response begins with a scrutiny of two positions he will ultimately criticize. But he does not reject them entirely. This is particularly obvious from the way he treats the so-called Aristotelian account. Instead of simply dismissing it, he points out that it does not explain the ‘precise and proximate cause of individuation’. Moreover, once the precise cause of individuation has been established it will become clear to what extent the previous positions are in a sense correct.38 For someone with this pedagogical strategy it makes sense to start by running together positions such as Porphyry’s, Boethius’s and Avicenna’s on the sole ground that they all speak about accidents and their role in individuation. For as it will turn out they are all, in some measure, expressing the truth, albeit in different respects. Henry’s treatment of individuation in Quodlibet V, q. 8 is notorious. Its notoriety is due to the fact that he in the end identifies the ‘precise and proximate’ cause of individuation with a ‘double negation’. Each individual is characterized by two negations, for an individual is essentially what is not further divided in itself and not identical with something else. This leads Henry to identify the principium individuationis with the negations themselves.39 Although the text is best understood in the context of Henry’s other discussions of individuation, uncharitable readers have often focused on it separately.40 Yet it should not come as a surprise that Henry’s critics found not only a seemingly strange theory in Quodlibet V, q. 8, but also other materials, for example, Henry’s presentation of accidental individuation. This, I believe, explains why we can find after Henry’s fifth Quodlibet a whole series of texts depicting Avicenna as a proponent of accidental individuation. There is no reason to believe that, for instance, Duns Scotus’s listing in texts I.7 – 9 in the appendix is original. Scotus knew both Henry’s Quodlibet as well as Quodlibet VII, q. 5 of Godfrey of Fontaines (who himself engages critically and extensively with Henry’s views). Since all the other authors listed in part I of the appendix engage in their respective questions with key aspects of Henry of Ghent’s account of

38 Quodlibet V, q. 8, fol. 165vL: ‘Haec ergo est opinio sua de causa individuationis et determinationis formae ad suppositum in formis materialibus plurificabilibus per supposita quantum est ex parte formae, et est vera in talibus formis ut secundum cursum naturae, sed non explicat praecisam et proximam causam individuationis … et per ipsam verificantur opiniones praedictae inquantum talium formarum individuatio fit per materiam ut est sub quantitate quae est accidentalis.’ 39 Ibid., fol. 166rM. 40 On this point see my Henry of Ghent on Individuation.

354 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Martin Pickav

individuation, it is more than plausible that they owe their depiction of Avicenna’s view also to him.41 So at the end there is good news. Medieval Latin authors of the thirteenth and fourteenth centuries did not systematically misunderstand Avicenna’s account of individuation. In the wake of Henry of Ghent, Avicenna’s view got somewhat caricatured by association with Porphyry and Boethius. But this was due to the heat of the debate and not to a serious misreading. Moreover, almost all the authors depicting Avicenna in the way described above would later in their texts exempt him from the misgivings of the accidental individuation view. Still it is somewhat ironic that the one medieval author who spilled more ink than anyone else over Avicenna’s account of individuation should be responsible for the caricature.

Appendix I Avicenna as a Proponent of Accidental Individuation (Selected Passages) I.1 Peter of Falco, Quaestiones disputatae, q. 5, pp. 198 f. [ca. 1279 – 81]: ‘Nam essentia ut essentia non dicit quid in actu nec quid ut unum nec ut multa, sicut docet Avicenna, V Metaphysicae; nam humanitas ut humanitas dicit essentiam vel quidditatem hominis, humanitas ut humanitas non est haec humanitas, quae est in Sorte, tunc omnis humanitas esset haec humanitas, immo quod sit haec humanitas et in Sorte, hoc habet per aliquid additum ultra essentiam, scilicet per respectum ad suppositum et per accidentia, unde individuatio fit per accidentia, ut arguit Avicenna, V Metaphysicae.’ I.2 Henry of Ghent, Quodlibet V, q. 8, fol. 165rI [Advent 1280 or Lent 1281]: ‘Nulla ergo restat difficultas in videndo quomodo in creaturis fiat per individuationem suppositum absolutum, nisi in modo individuationis quem Porphyrius posuit per accidentales proprietates, quae ut dicit numquam simul in eodem contingit reperiri. Et similiter Boethius dicit in commento super Porphyrium de individuis sub eadem specie: Quantum ad substantiam unum sunt non habentia substantialem differentiam, sed accidentibus efficiuntur, ut accidentibus saltem distare videantur. Plures sunt de quibus species praedicatur, non substantiae diversitate, sed accidentium multitudine. Et in principio de Trinitate: In numero differentiam accidentium varietas facit. Et Ioannes Damascenus in libro de duplici natura et una hypostasi in Christo dicit circa 41 Only Peter of Falco’s text (I.9) shows no traces of Henry’s fifth Quodlibet. But his text does also not list Avicenna next to John Damascene, Porphyry and Boethius. I still believe Peter is influenced by Henry, but by Quodlibet II, q. 8 rather than by Quodlibet V, q. 8.

On the Latin Reception of Avicenna’s Theory of Individuation 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

355

principium: Omnis res qua differt species ab altera specie substantialis et naturalis, constitutiva differentia dicitur; omnis autem res in qua differt hypostasis ab eiusdem speciei consubstantiali hypostasi dicitur adventitia differentia et qualitas et hypostatica et characteristica proprietas; hoc autem est accidens; velut differt homo ab homine altero, quoniam hic quidem est longus, hic autem brevis, hic quidem albus, hic vero niger. Et Avicenna in quinto Metaphysicae suae dicit: postquam determinata est species specialissima, erit natura, sic quod accidunt ei concomitantia ex accidentibus et proprietatibus per quae natura individuatur et fit designatum. Quod non potest stare, quoniam accidentia omnia huiusmodi sunt posterioris et adventitiae naturae, sicut et isti confitentur. Ergo naturaliter praesupponunt, etsi non tempore, substantiam ut eis subiectam. Subiecta autem substantia non est nisi individua …’ I.3 Richard of Middleton, In II Sententiarum, dist. 3, a. 4, q. 1, p. 58 [after 1281]: ‘Primo ostendo, quod Angelus sit substantia una numero formaliter per aliquid superadditum essentiae suae. Damascenus lib. 3, cap. 11 vult, quod indiuiduatio sit per accidens. Item Avicenna 3 Meta., cap. 3 dicit, quod unitas substantiae non recipitur in certificatione quidditatis alicuius substantiarum, sed est quiddam communicans substantiam, dicitur ergo de eis sicut accidens. Item Porphyrius ponit rationem indiuiduationis per accidentia. … Dixerunt autem quidam, quod substantia creata est una numero formaliter, per aliquid superadditum essentiae suae. De quorum opinione videtur fuisse Avicenna. Dicunt enim, quod denominatiua significant accidens, unum autem est denominatiuum. Praeterea, quia unitas est principium numeri et numerus est quantitas discreta, quae est accidens, et ita videtur eis, quod omnis unitas creata sit accidens. Praeterea, quia aliter esset nugatio, dicendo substantia una, nisi unitas significet aliquid superadditum substantiae, quae est una. Sed hoc opinio non videtur rationabilis. Unde et commentator super 4 Metaph. reprehendit Avicenna de hoc et dicit sic:42 Avicenna autem multum peccavit in hoc quod existimavit quod unum et ens significant dispositiones additas essentiae rei et mirum est de isto homine quomodo erravit tali errore. Et arguit sic contra ipsum: Si res esset una per aliquam rem additam suae naturae, sicut credidit Avicenna tunc nihil esset unum per se et per suam substantiam, sed per rem additam suae substantiae, et illa res, quae est una si dicitur, quod est 42 See Averroes, Long Commentary on Aristotle’s Metaphysics, lib. IV, com. 3, in Aristotle, Opera cum Averrois commentariis, vol. 8, fol. 67rB–vG.

356 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Martin Pickav

una per intentionem additam suae essentiae, quaeretur etiam de illa re per quam sit una. Si igitur sit una per intentionem additam illi iterum quaeretur, et sic procedit in infinitum. … Ad primum in oppositum dicendum, quod illa auctoritas Damasceni intelligenda est quantum ad manifestationem indiuiduationis, manifestatur enim Petrus esse aliud indiuiduum a Ioanne per diuersa accidentia, quae in eis sunt. Ad secundum dicendum, quod auctoritas Avicennae neganda est in proposito. Ad tertium dicendum est sicut ad primum.’ I.4 Roger Marston, Quodlibet II, q. 30, p. 294 [1282 – 4]: ‘Sapientes mundi plures, ut Boethius, Porphyrius et Avicenna, videntur velle quod individua solo accidente differant, quibusdam proprietatibus accidentalibus in uno repertis quae non sunt in alio possibiles inveniri, ut sunt patria, parentela, locus, tempus, et propria nominatio. Sed cum accidentia habeant esse a subiecto, magis accipiunt individuationem ab ipso quam subiectum ab eis.’43 I.5 Godfrey of Fontaines, Quodlibet VII, q. 5, pp. 319 f. [1290/91 or 1291/92]: ‘Videtur ergo quod individuatio fiat per accidentia, et hoc videntur dicere philosophi et sancti doctores. Nam Porphyrius dicit quod individua differunt per accidentales proprietates quas nunquam contingit simul in pluribus reperiri. Et Boetius in commento dicit quod individua sunt eadem specie; quantum ad substantiam unum sunt non habentia substantialem differentiam, sed accidentibus videtur effici ut in accidentibus saltim differre videantur plures substantiae de quibus species praedicatur et non substantiae diversitate, sed accidentium multitudine; et in principio libri de Trinitate: in numero differentiam accidentium varietas facit. Et Damascenus de Duabus naturis et una persona Christi, circa principium: omnis res qua differt species ab altera specie substantialis et naturalis et constitutiva differentia dicitur; omnis autem res in qua differt hypostasis ab eiusdem speciei cum substantiali hypostasi dicitur adventitia differentia et qualitas et hypostatica et caracteristica proprietas; hoc autem est accidens; vel ut differt homo ab homine altero, quoniam hic quidem est longus, hic est autem brevis, hic autem albus, hic vero niger. Et Avicenna, quinto Metaphysicae: postquam determinata fuerit natura facta species specialissima, erit natura determinata sic quod accidunt ei concomitantia ex accidentibus et proprietatibus per quae natura individuatur et sic designata videtur.’ 43 For the examples of individualizing accidents see also Auctoritates Aristotelis, auctoritates Porphyrii, ed. Hamesse, p. 300: ‘Et sunt septem proprietates, ut dicit Boethius, scilicet forma, figura, locus, stirps, nomen, patria, tempus. Haec septem propria continet omnis homo.’ However, such a list of seven properties cannot be found in Boethius’s works.

On the Latin Reception of Avicenna’s Theory of Individuation 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

357

I.6 Peter of Auvergne, Quodlibet II, q. 5, edited in Hocedez, Une Question indite de Pierre d’Auvergne, p. 371 [1297]: ‘De hoc autem fuit antiqua opinio: quod individuationis principium est aggregatio proprietatum accidentalium; quarum collectionem impossibile est in alio reperire secundum Porphirium. Unde Avicenna dicit V Metaphysicae capite 3: nature speciei specialissime accidunt concomitantia ex proprietatibus et accidentibus per que individuatur et sic designata est.’ I.7 John Duns Scotus, Quaestiones super libros Metaphysicorum Aristotelis, lib. VII, q. 13, pp. 219 f.: ‘Circa istam quaestionem quidam dicunt naturam individuari per aliquod positivum aliquo modo aliud a natura; quidam non. Prima pars habet quinque vias. Prima est de multis accidentibus aggregatis. Porphyrius: individuum constat ex septem proprietatibus. Et Boethius ibi in commento; idem De Trinitate. Et Damascenus; et Avicenna V Metaphysicae 2.’ I.8 John Duns Scotus, Lectura II, dist. 3, p. 1, q. 4, nn. 61 – 4, pp. 246 f.: ‘Quaeritur an quantitas sit illud positivum quo substantia materialis est haec et singularis et indivisibilis in multas partes subiectivas. Quod sic, videtur: Per Boethium De Trinitate, in principio: “Differentiam in numero accidentia faciunt, quae si omnia auferas, saltem locum auferre non potes; quare diversa numero sunt in diversis locis”; sed quod sint in locis diversis, non est nisi per quantitatem; ergo etc. Praeterea, Damascenus in Elementario suo, cap. 5, hoc etiam dicit, quod “omne per quod aliqua different hypostatice, accidens est”. Praeterea, Avicenna V Metaphysicae cap. 2: “Naturae quae eget materia ad esse adveniunt accidentia, per quae individuatur”.’ I.9 John Duns Scotus, Ordinatio II, dist. 3, p. 1, q. 4, n. 70, pp. 421 f.: ‘Quarto quaeritur utrum substantia materialis per quantitatem sit individua vel singularis. Quod sic: Boethius De Trinitate: “In numero differentiam, accidentium varietas facit, nam tres homines neque specie neque genere sed suis accidentibus differunt; nam si vel animo cuncta accidentia separamus, locum tamen cunctis diversus est, quem unum fingere duobus nullo modo possumus: duo enim corpora unum locum non obtinebunt, qui est accidens, et ideo sunt numero plures quanto accidentia plura sunt”. Et inter omnia accidentia, primum accidens est quantitas, quam etiam specialiter videtur exprimere in “loco” (dicendo quod “eundem locum fingere non possumus”), qui competit eis in quantum quanta.

358 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Martin Pickav

Praeterea, Damascenus in Elementario cap. 5 (non computando prooemium): “Omnis res in qua differt hypostasis ab eiusdem speciei hypostasi, dicitur adventicia differentia, et characteristica proprietas et qualitas hypostatica; hoc autem accidens est, – velut ‘differt homo ab altero homine, quoniam hic quidem longus, ille autem brevis’”. Praeterea, Avicenna in V Metaphysicae 2: “Natura quae materia eget, – ad eius esse adveniunt accidentia et dispositiones, extrinsece, per quae individuatur”.’ I.10 John of Paris, Quaestio de causa individuationis, edited in Mller, Eine Qustion ber das Individuationsprinzip, p. 343 [1304 – 5?]: ‘Quidam enim dicunt accidentia esse causam individuantionis, per quae species communis ad determinatum contrahitur individuum. Et huic opinioni alludit plurium auctoritas. Porphyrius enim dicit quod individuatio fit per accidentales proprietates, quarum collectio eadem numero in alio reperiri non potest. Boethius etiam in Commento dicit, ibidem, quod individua sunt sub una specie unum quantum ad substantiam, non habentia substantiae differentiam, sed accidentibus efficiuntur, ut accidentibus saltem distare videtur. Plures enim sunt, de quibus species praedicantur, non substantiae diversitate, sed accidentium diversitate vel multitudine. Item, Boethius in principio De Trinitate … Item, Ioannes Damascenus in libro De duplici natura et una hypostasi in Christo, dicit circa principium: “Omnis res …” Item, Avicenna in V libro Metaphysicae suae dicit: “postquam natura determinata est species specialissima erit natura sic quod accidunt ei concomitantia ex accidentibus et proprietatibus, per quae natura individuatur et fit designata.”’ II Accidental Individuation Without Avicenna (Selected Passages) II.1 William of Auvergne, De universo, II, c. 9, pp. 852bf.: ‘Et de prima quidem quaestionum praenominatarum audivisti sermonem Aristotelis, et sermones Alpharabii, et Avicennae, et aliorum, qui in parte ista Aristoteli consenserunt. Qui omnes in hanc unam sententiam concordaverunt, quod omnis diversitas, omnis numerus a materia est corporali.’ II.2 John of La Rochelle, Summa de anima, II, c. 113, pp. 272 f.: ‘Notandum tamen est secundum Avicennam, quod formis materialibus propter materiam accidunt disposiciones multe quae non habent ex sua essencia, scilicet ex hoc quod sunt forme. … Relinquitur igitur quod accidit humanitati ex materia iste modus multiplicationis et diuisionis que est singularibus.

On the Latin Reception of Avicenna’s Theory of Individuation 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

359

Accidunt ei eciam alia preter hoc, quia cum fuerit in materia, acquiritur ei aliquis modus quantitatis, qualitatis, situs et ubi, quae omnia abstracta sunt a natura ipsius. Si enim humanitas, ex hoc quod humanitas determinaret sibi accidencia huiusmodi, tunc oporteret unumquemque hominem conuenire cum altero et in huiusmodi accidentibus, cum conueniant in ipsa humanitate in quantum huiusmodi: essent igitur eedem qualitates et quantitates, etc. Relinquitur ergo quod forma humana non habet ex sua essencia huiusmodi accidencia; sed ista ei accidunt propter comparacionem ipsius ad naturam.’ II.3 Peter of Tarentaise (Pope Innocent V), In II Sententiarum, dist. 17, q. 1, a. 2, p. 142a: ‘Hoc aliquid dicitur individuum per se subsistens, hoc autem dupliciter: aut habet principium individuationis intra essentiam propriam, et omne tale compositum est ex materia et forma, quia principium individuationis est materia, aut habens extra se, sic non omne, quod est individuum per se subsistere valens, est compositum ex materia et forma, et tale est anima rationalis quae secundum Avicennam individuatur per corpus.’ II.4 Bonaventure, In II Sententiarum, dist. 3, p. 1, a. 2, q. 2, p. 105a: ‘Secundo quaeritur, utrum personalis proprietas sit in Angelis accidentalis, vel substantialis. Et quod accidentalis, videtur sic. 1. Boethius dicit, quod “omne principium manat de genere accidentium, non solum proprium individui, sed etiam speciei”; sed individuum magis approximat accidentibus quam species: ergo multo fortius proprietas individualis de genere accidentium est. Sed discretio personalis est proprietas individualis: ergo etc. 2. Item, Richardus de sancto Victore ait, “quod in divinis est personalis discretio per originem, in Angelis per qualitatem, in hominibus utroque modo.” … 3. Item, Porphyrius dicit, “quod individuum constat ex proprietatibus, quarum collectionem impossibile est in altero reperire”; sed tales sunt proprietates accidentales: ergo individuatio est per accidentia. Sed per eadem est personalis discretio, per quae est individuatio: ergo etc.’ II.5 Roger Bacon, Questiones supra libros prime philosophie Aristotelis, pp. 226 f.: ‘Queritur … de causa sue individuationis scilicet, et quid facit hujusmodi individuum esse individuum. Et videtur quod non accidentia: quia nullum accidens est perfectio substantie … Contrarium a multis dicitur, quia Porphyrius, “individua solo numero differunt”; set numerus est accidens, ergo cum accidens ex accidentibus fiat, ergo unitas numeralis fit ex accidentibus, quare et individuatio. Item, Boethius in libro De Trinitate, “tres homines nec genere nec specie discrepant set solo accidente”, quare accidens facit

360 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Martin Pickav

individuationem. Istud idem dicit in Commento supra Porphyrium, ipse Boethius loquitur ibi de illis 7 proprietatibus. Item, in Metaphysica Argazelis, in causa prima, quia ibi non est compositio differentie vel accidentis, ideo non est in eo dualitas ullo modo, quare omnis compositio, vel compositio differentie cum alio, vel accidentis; ergo individuum cum sit compositum, aut vere erit per additionem differentie, quod non est verum quia differentia antecedit speciem, individuum sequitur, fiet ergo per compositionem accidentium, ergo etc.’44 II.6 Robert Kilwardby, Quaestiones in librum secundum Sententiarum, q. 17, p. 63: ‘Quod sit forma accidentalis, videtur per Porphyrium. Qui dicit quod individuum dicitur, quoniam “consistit ex proprietatibus quarum collectio numquam in alio eadem erit.” Item Boethius De Trinitate cap. 2 …’ II.7 William de la Mare, Scriptum in secundum librum Sententiarum, dist. 3, q. 6, p. 62: ‘Alii dicunt quod accidentia sunt causa individuationis et hoc videtur sensisse Porphyrius qui dicit quod individuum constat ex proprietatibus quarum collectio numquam potest in aliquo reperiri. Hoc idem videtur per Boethium, De trinitate: “In numero”, inquit, “differentiam accidentium varietas facit.”’ II.8 Giles of Rome, In II Sententiarum, dist. 3, pars 1, q. 2, a. 1, p. 184: ‘Tertiam difficultatem faciunt dicta Doctorum. Nam Boethius in libro de Trinitate circa principium ait, quod in numero differentiam accidentium varietas facit. Et subdit, quod tres homines, neque genere, neque specie, sed suis accidentibus differunt. Et Porphyrius ait, quod individua dicimus huiusmodi, quoniam ex proprietatibus consistit unumquodque eorum, quarum collectio nunquam in alio eadem erit. Et Simplicius in libro super praedicamentis ait: Numero autem differunt, quaecumque cum cursu accidentium proprietatem suae substantiae determinaverunt …’ II.9 Peter John Olivi, Quaestiones in secundum librum Sententiarum, q. 12, p. 212: ‘Quidam enim voluerunt quod individuatio dicat collectionem accidentium, de quorum numero Porphyrius et Boethius fuisse videntur.’

˙ aza-lı¯, Metaphysica, p. 26. 44 See al-G

On the Latin Reception of Avicenna’s Theory of Individuation 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

361

Bibliography J.A. Aertsen, Die Thesen zur Individuation in der Verurteilung von 1277, Heinrich von Gent und Thomas von Aquin, in Individuum und Individualitt im Mittelalter (Miscellanea Mediaevalia 24), eds J.A. Aertsen and A. Speer, Berlin/New York: Walter de Gruyter, 1996, pp. 249 – 65. ˙ aza-lı¯, Metaphysica, ed. J.T. Muckle, Toronto: St. Michael’s College, 1933. al-G Aristotle, Opera cum Averrois commentariis, vol. 8, Venice, 1562, reprinted Frankfurt a. M.: Minerva, 1962. Avicenna, Liber de philosophia prima sive Scientia divina, ed. S. van Riet, 3 vols, Louvain: Peeters, 1977 – 83. A. Bck, The Islamic Background: Avicenna (b. 980; d. 1037) and Averroes (b. 1126; d. 1198), in Individuation in Scholasticism: The Later Middle Ages and the CounterReformation, 1150 – 1650, ed. J.J.E. Gracia, Albany, New York: SUNY Press, 1994, pp. 39 – 67. D. Black, Individuation, Self-Awareness, and God’s Knowledge of Particulars: The Intersection of Three Avicennian Problems (forthcoming). A.M.S. Boethius, De consolatione philosophiae et opuscula sacra, ed. C. Moreschini, Munich/Leipzig: K.G. Saur, 2005. –– , In Isagogen Porphyrii commenta (Corpus Scriptorum Ecclesiasticorum Latinorum 48), ed. S. Brandt, Vienna: Tempsky, 1906. Bonaventure, Opera omnia, vol. 2: Commentarius in II Sententiarum, eds PP Collegii S. Bonaventurae, Quarachi: Typographia Collegii S. Bonaventurae, 1885. S.F. Brown, Henry of Ghent, in Individuation in Scholasticism: The Later Middle Ages and the Counter-Reformation, 1150 – 1650, ed. J.J.E. Gracia, Albany, New York: SUNY Press, 1994, pp. 195 – 219. R. Chiaradonna, La teoria dell’individuo in Porfirio e l’IDIYS POIOM stoico, Elenchos 21, 2000, pp. 303 – 31. T.-A. Druart, The Human Soul’s Individuation and Its Survival After the Body’s Death: Avicenna on the Causal Relation Between Body and Soul, Arabic Sciences and Philosophy 10, 2000, pp. 259 – 73. C. Erismann, Collectio proprietatum. Anselme de Canterbury et le problme de l’individuation, Mediaevalia. Textos e estudos 22, 2003, pp. 55 – 71. –– , L’individualit explique par les accidents: Remarques sur la destine ‘chrtienne’ de Porphyre, in Complments de substance: tudes sur les proprits accidentelles offertes  Alain de Libera, eds C. Erismann and A. Schniewind, Paris: Vrin, 2008, pp. 51 – 66. Giles of Rome, In secundum librum Sententiarum, Venice, 1581, reprinted Frankfurt a.M.: Minerva, 1968. P. Glorieux, Les premires polmiques thomistes: 1. Le Correctorium corruptorii ‘Quare’, Kain: Le Saulchoir, 1927. Godfrey of Fontaines, Les Quodlibet cinq, six et sept (Les Philosophes Belges, vol. 3), eds M. De Wulf and J. Hoffmans, Louvain: Institut suprieur de philosophie, 1914. J.J.E. Gracia, Introduction to the Problem of Individuation in the Early Middle Ages, Munich/Vienna: Philosophia Verlag, 1984. J. Hamesse, ed., Les Auctoritates Aristotelis, Louvain: Publications universitaires, 1974. D.N. Hasse, Avicenna’s ‘De anima’ in the Latin West: The Formation of a Peripatetic Philosophy of the Soul, 1160 – 1300, London/Turin: The Warburg Institute/Nino Aragno Editore, 2000.

362 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Martin Pickav

Henry of Ghent, Quodlibeta, 2 vols, ed. J. Badius Ascensius, Paris, 1518; reprinted Leuven: Bibliothque S. J., 1961. –– , Opera omnia, vol. 6: Quodlibet II, ed. R. Wielockx, Leuven: Leuven University Press, 1983. E. Hocedez, Une Question indite de Pierre d’Auvergne sur l’individuation, Revue noscolastique de philosophie 39, 1934, pp. 355 – 86. John Damascene, Dialectica, version of Robert Grosseteste, ed. O.A. Colligan, St. Bonaventure, New York: Franciscan Institute, 1953. –– , De institutione elementari, version of Robert Grosseteste, ed. H. Gravius, Cologne, 1546. John Duns Scotus, Opera omnia, vol. 7: Ordinatio II, dist. 1 – 3, ed. Commissio Scotistica, Vatican City: Typis Polyglottis Vaticanis, 1973. –– , Opera omnia, vol. 18: Lectura II, dist. 1 – 6, ed. Commissio Scotistica, Vatican City: Typis Polyglottis Vaticanis, 1982. –– , Opera philosophica, vols 3 – 4: Quaestiones super libros Metaphysicorum Aristotelis, eds G. Etzkorn et al., St. Bonaventure, New York: Franciscan Institute, 1997. John of La Rochelle, Summa de anima, ed. J.G. Bougerol, Paris: Vrin, 1995. P. King, The Problem of Individuation in the Middle Ages, Theoria 66, 2000, pp. 159 – 84. –– , Metaphysics, in The Cambridge Companion to Abelard, eds J.E. Brower and K. Guilfoy, Cambridge: Cambridge University Press, 2004, pp. 65 – 125. A. de Libera, L’art des gnralits. Thories de l’abstraction, Paris: Aubier, 1999. –– , D’Avicenne  Averros, et retour. Sur les sources arabes de la thorie scolastique de l’un transcendantal, Arabic Sciences and Philosophy 4, 1994, pp. 141 – 79. M.E. Marmura, Avicenna’s Chapter on Universals in the Isagoge of his Shifa’, in Islam: Past Influence and Present Challenge, eds A.T. Welch and P. Cachia, Edinburgh: Edinburgh University Press, 1979, pp. 34 – 56. J.P. Mller, Eine Qustion ber das Individuationsprinzip des Johannes von Paris O.P. (Quidort), in Virtus politica. Festgabe zum 70. Geburtstag von Alfons Hufnagel, eds J. Mçller and H. Kohlenberger, Stuttgart-Bad Cannstatt: Fromann, 1974, pp. 335 – 56. Peter of Falco, Quaestiones disputatae, 3 vols, ed. A.-J. Gondras, Louvain: ditions Nauwelaerts, 1968. Peter John Olivi, Quaestiones in secundum librum Sententiarum, 3 vols, ed. B. Jansen, Quaracchi: Typographia Collegii S. Bonaventurae, 1922 – 6. –– , Quodlibeta quinque, ed. S. Defraia, Grottaferrata: Editiones Collegii S. Bonaventuae, 2002. Peter of Tarentaise (Pope Innocent V), In IV libros Sententiarum, Toulouse, 1652; reprinted Ridgewood, New Jersey: Gregg Press, 1964. M. Pickav, The Controversy of the Principle of Individuation in Quodlibeta (1277–ca. 1320): A Forest Map, in Theological Quodlibeta in the Middle Ages: The Fourteenth Century, ed. C. Schabel, Leiden: Brill, 2007, pp. 17 – 79. –– , Henry of Ghent on Individuation, Proceedings of the Society for Medieval Logic and Metaphysics 5, 2005, pp. 38 – 49 [http://www.fordham.edu/gsas/phil/klima/SMLM/ PSMLM5/PSMLM5.pdf ]. Porphyry, Isagoge, translatio Boethii, ed. B.G. Dod, in Aristoteles Latinus, vol. I.6 – 7, Bruges/Paris: Descle de Brower, 1966. –– , Isagoge, ed. A. Busse, in Commentaria in Aristotelem Graeca, vol. IV.1, Berlin: G. Reimer, 1887.

On the Latin Reception of Avicenna’s Theory of Individuation 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

363

Richard of Middleton, In IV libros Sententiarum, 4 vols, Brescia, 1591, reprinted Frankfurt a.M.: Minerva, 1963. Robert Kilwardby, Quaestiones in librum secundum Sententiarum, ed. G. Leibold, Munich: Bayerische Akademie der Wissenschaften, 1992. Roger Bacon, Opera hactenus inedita, vol. 10: Questiones supra libros prime philosophie Aristotelis, eds R.W. Steele and F.M. Delorme, Oxford: Clarendon Press, 1930. Roger Marston, Quodlibeta quatuor, eds G. Etzkorn and I. Brady, Quaracchi: Typographia Collegii S. Bonaventurae, 1968. M.-D. Roland-Gosselin, Le ‘De ente et essentia’ de S. Thomas d’Aquin, Kain: Le Saulchoir, 1926. M. Sebti, Avicenne: l’me humaine, Paris: Presses universitaires de France, 2000. R. Sorabji, Self: Ancient and Modern Insights about Individuality, Life, and Death, Chicago: Chicago University Press, 2006. T. Suarez-Nani, Les anges et la philosophie, Paris: Vrin, 2002. Thomas Aquinas, Super Boetium de trinitate, in id., Opera omnia, vol. 50, ed. Leonina, Rome/Paris, 1992. –– , De ente et essentia, in id., Opera omnia, vol. 43, ed. Leonina, Rome, 1976. Thomas Sutton, Quaestiones ordinariae, ed. J. Schneider, Munich: Bayerische Akademie der Wissenschaften, 1977. William of Auvergne, De universo, in id., Opera omnia, vol. 1, Paris, 1674, reprinted Frankfurt a.M.: Minerva, 1963. William de la Mare, Scriptum in secundum librum Sententiarum, ed. H. Kraml, Munich: Bayerische Akademie der Wissenschaften, 1995. J.F. Wippel, The Metaphysical Thought of Thomas Aquinas, Washington, DC: Catholic University of America Press, 2000.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Scotus and Avicenna on What it is to Be a Thing Giorgio Pini Scotus’s debt to Avicenna’s Metaphysics has been subjected to careful scrutiny in the last few years. After a landmark article by tienne Gilson of some eighty years ago, it became for some time customary to assume that Scotus had been strongly influenced by Avicenna in a number of key issues.1 This claim is now increasingly being regarded with some caution. This is mostly due to historical and textual considerations. It is now recognized that Scotus’s discussion and interpretation of Avicenna should be considered in the larger context of Scotus’s complex relationship with the later medieval theologian, Henry of Ghent. Scotus, it appears, focused on those passages by Avicenna that Henry of Ghent had already taken into account before him. Admittedly, this does not entail that Scotus had only second-hand knowledge of Avicenna’s text.2 But it is difficult to resist the conclusion that Scotus discussed and quoted Avicenna because he discussed and quoted Henry of Ghent. As to the supposedly exceptional influence that Avicenna’s thought had on Scotus, a simple count of Scotus’s explicit quotations from Avicenna’s Metaphysics suggests some caution. Scotus’s references turn out to be more or less as many as those present in Thomas Aquinas.3 Even though this rough comparison should not be taken as absolutely 1

2 3

Gilson, Avicenne et le point, p. 147: ‘Il est d’abord difficile de ne pas Þtre frapp de l’influence exerce par Avicenne sur la formation du scotisme …’ In that article, Gilson stressed Avicenna’s influence on Scotus with regard to the doctrine of the subject matter of metaphysics, the object of the intellect, and the abstraction mechanism. Some forty years later, Gilson repeated basically the same assessment and stressed Avicenna’s influence on Scotus’s teachings on the subject matter of metaphysics, the common nature, and the univocity of the concept of being. See Gilson, Avicenne en Occident, pp. 109 – 12. Specific aspects of the complex relationship between Avicenna and Scotus have recently been considered by Sondag, La rception; Porro, Duns Scot; and Druart, Ibn Sina (Avicenna). Sondag, La rception, p. 591. On the influence of Avicenna’s metaphysics on Henry of Ghent’s thought, see Teske, Henry of Ghent’s Debt. I base this remark on a comparison between the first three books of Scotus’s Ordinatio and of Thomas Aquinas’s commentary on the Sentences (I have left out the fourth book because only the first thirteen distinctions of the fourth book of Scotus’s Ordinatio have been critically edited at this date). In the first book of Aquinas’s commentary on the Sentences, there are some 50 explicit references to Avicenna’s Metaphysics against the 67 references present in the first book of Scotus’s Ordinatio. Things are reversed in the second book, where to 42 references to Avicenna’s Metaphysics in Aquinas’s commentary there correspond 21 references in Scotus’s Ordinatio. As to the third book, there are 9

366 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Giorgio Pini

conclusive evidence, it is a strong indication that Scotus’s debt towards Avicenna is considerable but not exceptional among late thirteenth-century authors. It is interesting to notice that things seemed to have changed after Scotus, at least as far as English authors are concerned. Both William of Ockham and Adam Wodeham virtually stopped referring to Avicenna’s Metaphysics, as attested by the indexes of quotations in the critical editions of their works. Setting aside these general concerns about Avicenna’s influence on Scotus, I will consider a specific case, namely Scotus’s interpretation of a famous passage taken from the fifth chapter of the first book of Avicenna’s Metaphysics. In that chapter, Avicenna makes some famous claims about thing and being as the first notions received in our minds.4 This paper has three parts. First, I consider Thomas Aquinas’s and Henry of Ghent’s views about how to interpret Avicenna’s claims about what it is to be a thing. Second, I turn to Scotus’s view, which may be seen as a development and criticism of Henry of Ghent’s position. Third, I conclude with some remarks on the role our concept of thing (or being or something) plays in cognition. Specifically, I argue that Scotus’s distinction between our situation before and after the fall – namely, the distinction between, on the one hand, what we were originally supposed to cognize and, on the other hand, our current limitations – is particularly relevant for his interpretation of the claim that being and thing are the objects of the intellect, and more in general for his interpretation of Avicenna’s Metaphysics. My conclusion is that Avicenna’s influence on Scotus is undeniable (just as is the case with Thomas Aquinas and Henry of Ghent). More importantly, however, even with regard to such a fundamental doctrine as the primacy of the concepts of thing and being over other concepts, it makes little sense to speak of Avicenna’s influence without specifying in which way Avicenna was interpreted. In turn, the way Avicenna was interpreted had often much to do with concerns that were extraneous to Avicenna’s original viewpoint. Behind the interpretation of Avicenna, the main interest of these authors was to understand how things really are in the world. Avicenna offered some important conceptual tools to find an answer to this most fundamental question, but the way Avicenna was interpreted in turn depended on the basic metaphysical assumptions that each of these authors made.

4

references in Aquinas against 4 references in Scotus. I have taken the information about Aquinas from Vansteenkiste, Avicenna-Citaten, pp. 457 – 507. The data about Scotus are taken from the indexes to volumes 1 – 10 of the Vatican edition of Scotus’s Opera omnia. On Avicenna’s influence on Aquinas, see Anawati, Saint Thomas d’Aquin, pp. 449 – 65; Wippel, The Latin Avicenna, pp. 51 – 90. In this paper, I follow the practice of putting a term between inverted commas (e. g., ‘being’) when referring to the term and of italicizing it (e. g., being) when referring to the corresponding concept.

Scotus and Avicenna on What it is to Be a Thing 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

367

I Avicenna’s claim in chapter I.5 that thing and being, as well as necessary, are the first notions impressed in our minds is one of the most often quoted passages from his Metaphysics. 5 After making that claim, Avicenna goes on to stress the difficulty of providing a clarification of such general notions in terms of other notions. It seems that every attempt to clarify what it is to be a thing and what it is to be a being must be done by referring to concepts that already presuppose the concepts of thing and being. Consequently, any description of thing and being turns out to be circular. However, this does not end Avicenna’s discussion. Specifically with regard to the concept of thing, Avicenna does actually suggests two possible descriptions of what it is to be a thing. The first description is extremely generic. A thing is described as something about which a true statement can be made. Avicenna remarks that this description does not escape the charge of circularity, because it uses notions that can only be explained by presupposing what a thing is, and it basically amounts to saying that a thing is a thing (again) about which a true statement is made. Avicenna, however, admits that even such a circular description may be useful, as it directs one’s attention to what we are trying to identify, namely things. In other terms, even though such a description fails to explain what a thing is, it can be used to discriminate if something is a thing or not. As long as I can make a true statement about something, that something is a thing: The case is similar with somebody’s statement: ‘The thing is that about which it is valid [to give] an informative statement,’ for ‘is valid’ is less known than ‘the thing’; and ‘informative statement’ is [likewise] less known than ‘the thing’. How, then, can this be the definition of the thing? Indeed, ‘is valid’ and ‘information’ are known only after one uses, in explaining what they are, [terms] indicating that each is either a ‘thing,’ a ‘matter,’ a ‘whatever,’ or a ‘that which [is]’ – all of these being like synonyms of the world ‘thing.’ How, then, can the thing be truly defined in terms of what is known only through it? Yes, in this and in similar things there may be some act of directing attention: but, in reality, if you say, ‘The thing is that about which it is valid [to give] an informative statement,’ it is as if you have said, ‘The thing is the thing about which it is valid [to give] an informative statement,’ because the meaning of ‘whatever,’ ‘that which,’ and ‘the thing’ is one and the same. You would have then included ‘the thing’ in the definition of ‘the thing.’ Still, we do not deny that through this [statement] and its like, despite its vitiating starting point, there occurs in some manner a directing of attention to the thing.6

5

6

Avicenna, The Metaphysics of ‘The Healing’, I, 5. See Aertsen, Avicenna’s Doctrine, p. 22. Later in the same chapter, Avicenna lists thing, being, and one (rather than necessary) as the notions that are common to everything. On the difference between the two lists, see Aertsen, Avicenna’s Doctrine, pp. 25 – 6. Avicenna, The Metaphysics of ‘The Healing’, I, 5, pp. 23 – 4. See the Latin translation in Avicenna Latinus, Liber de philosophia prima, vol. 1, pp. 33 – 4: ‘Similiter est etiam hoc

368 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Giorgio Pini

Second and more specifically, a thing can be described as something having a certain reality (certitudo, in the Latin translation), and this is the proper being of that thing: ‘The thing,’ or its equivalent, may be used in all languages to indicate some other meaning. For, to everything there is a reality by virtue of which it is what it is. Thus, the triangle has a reality in that it is a triangle, and whiteness has reality in that it is whiteness. It is that which we should perhaps call ‘proper existence’ …7.

Avicenna’s readers in the Latin West made a real effort to clarify these two descriptions. What exactly is Avicenna doing? Is he distinguishing between two definitions of the term ‘thing’? Or is he rather distinguishing between two kinds of things? Moreover, what is the relationship between these two descriptions? Is the former just a first and generic attempt to clarify what a thing is, and in that case, is it superseded by the second description? Or should these two descriptions be considered as two equally valuable ways to pick out different kinds of things? Finally, what is the ontological commitment of each of these descriptions?8 By wondering about Avicenna’s intentions in this dense passage, Latin medieval authors were trying to come to terms with an important if elusive metaphysical point. For no matter how one interprets Avicenna, there seems to be much merit in the attempt to distinguish between two descriptions or

7

8

quod dicitur quod res est id de quo potest aliquid vere enuntiari; certe potest aliquid minus notum est quam res, et vere enuntiari minus notum est quam res. Igitur quomodo potest hoc esse declaratio? Non enim potest cognosci quid sit potest aliquid vel vere enuntiari, nisi in agendo de unoquoque eorum dicatur quod est res vel aliquid vel quid vel illud; et haec omnia multivoca sunt nomina rei. Quomodo ergo vere potest sciri res per aliquid quod non potest sciri per eam? Sed fortasse hoc et consimile erit innuitio aliqua. Nam cum dicis quod res est id de quo vere potest aliquid enuntiari, idem est quasi dicens quod res est res de qua vere potest aliquid enuntiari; nam id et illud et res eiusdem sensus sunt. Iam igitur posuisti rem in definitione rei, quamvis nos non negamus quod haec et consimilia, cum sint vitiosa, tamen aliqua designatio rei sunt’. See, on this chapter, Marmura, Avicenna on Primary Concepts. On the background to Avicenna’s distinction between thing and existent, and on its significance, see Jolivet, Aux origines; Wisnovsky, Avicenna’s Concept of Thingness, which was revised and expanded in Wisnovsky, Avicenna’s Metaphysics in Context, pp. 145 – 80. On some Latin interpretations of Avicenna’s notions of being and thing (by Thomas Aquinas, Henry of Ghent, and Francis of Marchia), see Aertsen, Avicenna’s Doctrine, pp. 27 – 33. Avicenna, The Metaphysics of ‘The Healing’, I, 5, 9, p. 24. See the Latin translation in Avicenna Latinus, Liber de philosophia prima, vol. 1, pp. 34 – 5: ‘Sed res et quicquid aequipollet ei, significant etiam aliquid aliud in omnibus linguis; unaquaeque enim res habet certitudinem qua est id quod est, sicut trangulus habet certitudinem qua est triangulus, et albedo habet certitudinem qua est albedo. Et hoc est quod fortasse appellamus esse proprium …’. As far as I know, these questions have not been adequately considered in the literature on Latin medieval authors, which focuses almost exclusively on the relationship between essence and existence.

Scotus and Avicenna on What it is to Be a Thing 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

369

characterizations of what it is to be a thing. It is certainly not the case that we can talk only about things with a certain essence, such as human beings, cats, trees and the like. We can say of some things, for example of elves, that they do not exist. In that case, we are still talking about something. But we certainly do not want to attribute any ontological status to what we are talking about. Rather, when we say that elves do not exist, our point is just to deny that elves are real things. Similarly, we may describe blindness as an unfortunate situation that some human beings and some animals experience. Thus, we do talk about blindness. At the same time, we may be understandably unwilling to grant blindness any ontological status. Avicenna’s distinction between two descriptions of what it is to be a thing captures exactly this point. Sometimes we speak about things to which we do not want to attribute any reality in the extramental world. More precisely, sometimes we speak about things without taking any stance as to whether those things have any ontological status and, if they do, as to what that is. Rather than distinguishing between different senses or uses of ‘to exist’ or ‘to be,’ Avicenna distinguishes between things as the objects of true statements and things as real constituents of the world, i. e. essences. Several Western readers of Avicenna tried to make the most of that distinction. In his commentary on the Sentences, Thomas Aquinas first distinguished between the term ‘thing’ and the term ‘being’ according to whether they refer to what something is, i. e. its quiddity and essential account, or to the fact that something exists, i. e. its existence. Aquinas goes on to claim that a quiddity can exist in two ways, i. e. in a singular thing outside the mind and in the mind itself as something apprehended by the intellect. For example, suppose that I have a concept of what a cat is. In that case, the quiddity cat exists both in individual cats and in my mind. Clearly, Aquinas is making use of Avicenna’s famous view, which Avicenna presented in Metaphysics V and elsewhere, that a certain essence can be considered in three ways, i. e. in itself, according to the being it has in extramental individuals and according to the being it has in the mind as something understood.9 What is particularly interesting, however, is the way Aquinas connected the passage from Metaphysics V, where Avicenna introduced the distinction between the three ways an essence can be considered, with the passage from Metaphysics I.5, which we are currently taking into account. For Aquinas went on to notice that to the two ways a quiddity can exist, namely in the mind and outside the mind, there correspond two meanings of the word ‘thing’. To the quiddity existing in extramental individuals there corresponds the meaning of ‘thing’ as something determined 9

Avicenna, The Metaphysics of ‘The Healing’, V, 1, 4, p. 149. See Avicenna Latinus, Liber de philosophia prima V, 1, vol. 1, pp. 228 – 9. See Marmura, Quiddity and Universality; id., Avicenna’s Chapter on Universals; de Libera, La querelle, pp. 185 – 91; Black, Mental Existence, pp. 47 – 51.

370 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Giorgio Pini

and fixed in nature (res dicitur quasi aliquid ratum et firmum in natura). To the quiddity existing in the soul there corresponds the meaning of ‘thing’ (in Latin, res) taken from the verb reor-reris, namely ‘to opine’ or ‘to think’.10 By drawing this distinction between two meanings of res – as what is in the mind when I think about something and as a real constituent of the world – Aquinas tried to capture the distinction he found in Avicenna between, on the one hand, a thing as that about which a true statement is made and, on the other hand, a thing as an entity having a certain essence. Aquinas’s assumption is that, when I think about a certain object, that object is both in the extramental world and in my mind. Also, that object, say the essence of cats, can be called a ‘thing’ both when it is in the extramental world and when it is in my mind. But when I say that the object I am thinking about is a thing in the world and is also a thing in the mind, I am using ‘thing’ in two different senses. Thus, things in the mind are not another kind of things over and above things in the world; rather, they are things in a different sense of the word ‘thing’. There is, however, a problem with this reading of Avicenna. Aquinas can easily account for our thoughts about objects such as cats and human beings. These objects are things in the world in the first meaning of ‘thing’ and, when I think about them, they are also things in my mind in the second meaning of ‘thing’. Suppose, however, that I think about a nonexistent object, such as a chimera or a goatstag (these are the standard medieval examples of imaginary beings). Clearly, I can make several true statements about those objects; for example, I can say that chimeras do not exist. In that case, there is something in my mind but nothing corresponds to it in the extramental world, according to Aquinas’s account in the passage we are considering. But then, what is this ‘thing in the mind’? Surprisingly, Aquinas is committed to saying that it is a quiddity or essence. For Aquinas, in the passage I am referring to, explicitly identified a thing as that about which a true statement is made with a quiddity as existing in the mind (quidditas … in anima secundum quod est apprehensa ab intellectu). Admittedly, there is no individual exemplifying that essence and accordingly 10 Thomas Aquinas, In I Sent., d. 25, q. 1, a. 4: ‘Secundum Avicennam, ut supra dictum est, hoc nomen “ens” et “res” differunt secundum quod est duo considerare in re, scilicet quidditatem et rationem eius, et esse ipsius. Et a quidditate sumitur hoc nomen “res”. Et quia quidditas postest habere esse et in singulari quod est extra animam et in anima secundum quod est apprehensa ab intellectu, ideo nomen rei ad utrumque se habet: et ad id quod est in anima, prout “res” dicitur a “reor, reris”, et ad id quod est extra animam, prout “res” dicitur quasi aliquid ratum et firmum in natura. Sed nomen entis sumitur ab esse rei.’ Aquinas distinguished between the two meanings of ‘thing’ also in In II Sent., d. 37, q. 1, a. 1. Before Aquinas, Bonaventure had drawn the same distinction in In II Sent., d. 37, dub. I, p. 876. On Bonaventure’s and Aquinas’s passages, see Aertsen, Transcendental Thought, pp. 11 – 14. On the claim that res comes from reor–reris, see Hamesse, Res chez les auteurs philosophiques, pp. 91 – 104.

Scotus and Avicenna on What it is to Be a Thing 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

371

that essence is not something ‘determined and fixed in nature’. Nevertheless, only because I can think about chimeras, there must be an essence of chimeras, even though that essence is not instantiated. This claim, however, is extremely costly from an ontological point of view. Elsewhere, Aquinas himself did not seem to be willing to endorse it. Thus, we should conclude that Aquinas’s interpretation of Avicenna’s distinction between two descriptions of what it is to be a thing does not account for our thoughts about nonexistents, unless we are willing to populate the world with uninstantiated essences. The same reasoning can be extended from nonexistent things such as chimeras to privations such as blindness, of which Aquinas explicitly denied that they possess any positive ontological status. The root of the problem is the correlation that Aquinas here posited between the meaning of ‘thing’ as something in the mind and an essence’s way of being in the mind. Accordingly, it does not come as a surprise that Aquinas did not come back to this view when he considered the issue more carefully. Elsewhere, he preferred to speak not of two meanings of ‘thing’ correlated to two ways of being of an essence; rather, he introduced two main meanings of ‘to be’ (esse). In its first meaning, ‘to be’ is said of an extramental essence (or its actual existence). For example, we say that cats and human beings are or are beings. In its second meaning, ‘to be’ is said of a truthmaker, i. e. of what a certain statement is about and what makes a statement true. For example, we say that Socrates is or is a being because Socrates is what the statement ‘Socrates is white’ is about.11 Now Aquinas held that the second meaning of ‘to be’, i. e. as that about which a statement is made, does not entail any direct reference to an essence, whether extramental or merely in the mind. Thus, we can speak about chimeras and goatstags without positing any quiddity of chimeras and goatstags. Similarly, we can speak about blindness without positing any quiddity of blindness.12 Some years after Aquinas, Henry of Ghent referred repeatedly to the distinction between two meanings or, as he said, concepts of what it is to be a 11 Aquinas explicitly takes the second meaning of ‘to be’ from Aristotle, Met. V, 7, 1017a31 – 5. 12 See for example Aquinas, ST I, q. 3, a. 4; De pot., q. 7, a. 2, ad 1; In V Met., lect. 9, nn. 889 – 96; see also In I Sent., d. 33, q. 1, a. 1, ad 1, ed. Mandonnet, pp. 765 – 6 (where Aquinas distinguishes between not two but three meanings of ‘to be’). See Wippel, The Metaphysical Thought, pp. 24 – 5. It seems that, according to Aquinas, chimeras, goatstags and the like count as things that are both not actual and not possible. Such things have no essence and no ontological status whatsoever. By contrast, nonexisting possible things, such as my nonexisting twin brother, do have some sort of ontological status. Their ontological status as possible things is grounded on divine ideas. It should be remembered, however, that Aquinas held that divine ideas are really identical with God’s essence. As a consequence nonexisting possibles have no reality outside and independently of God’s essence. See Wippel, The Reality of Nonexisting Possibles.

372 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Giorgio Pini

thing.13 Henry was not always consistent in his interpretation of the two concepts of what it is to be a thing. Here I will focus on his treatment in Quodlibet VII, q. 1 – 2. In that work, Henry held that there are three main concepts of what it is to be a thing. First, according to the most general concept, a thing is ‘what is not nothing’, where ‘nothing’ means what neither is nor can be, either in the mind or outside the mind. This most general concept of thing is divided into two further concepts.14 Thus, second, a thing is what is called ‘thing’ or ‘res’ from the verb reor-reris, i. e. something that is or can be only in the mind as an object of thought. It is in this sense that fictitious beings, such as golden mountains or goatstags, are things.15 Third, a thing is what is called ‘thing’ or ‘res’ from the term ratitudo, which indicates the possession of an essence. Only things that exist or can exist outside the mind are things in this sense.16 These things are further divided into created and uncreated things, and created things are further divided into the ten Aristotelian categories. Thus, we can think and speak about golden mountains and goatstags even though those objects do not have an essence and are called ‘things’ only in the sense that they are objects of thought. Accordingly, Henry held that the scope of what we can think and speak about is not limited to real being, i. e. what is or can be in a natural kind by virtue of its essence. Nevertheless, Henry still linked the concept of what it is to be a thing as an object of thought to the concept of what it is to be a thing as a real essence. He actually made the former concept dependent on the latter. Admittedly, golden mountains and goatstags are not 13 See for example Henry of Ghent, Quodl. V, q. 2, fol. 154rD; Summa (Quaestiones ordinariae), art. 34, q. 2, pp. 174 – 5; Summa a. 21, q. 4, fols 127rO–127vO. See Aertsen, Transcendental Thought, pp. 2 – 10; Porro, Possibilit ed esse essentiae, pp. 226 – 228. See also Porro, Universaux et esse essentiae, pp. 9 – 51. 14 Henry of Ghent, Quodl. VII, q. 1 – 2, pp. 26 – 7: ‘… sciendum quod omnium communissimum, omnia continens in quodam ambitu analogo, est res sive aliquid, sic consideratum ut nihil sit ei oppositum nisi purum nihil, quod nec est nec natum est esse, neque in re extra intellectum, neque etiam in conceptu alicuius intellectus, quia nihil est natum movere intellectum nisi habens rationem alicuius realitatis. Res autem sive “aliquid” sic communissime acceptum, non habet rationem praedicamenti, – sic enim esset tantum unum praedicamentum continens Creatorem et creaturam –, sed distinguitur distinctione analogica in id quod est aut natum est esse tantum in conceptu intellectus sive in ipso intellectu, et in id quod cum hoc aut est aut natum est esse in re extra intellectum.’ 15 Ibid.: ‘Res primo modo est “res” secundum opinionem tantum, et dicitur “a reor, reris”, quod idem est quod “opinor, opinaris” quae tantum res est secundum opinionem, quoad modum quo ab intellectu concipitur, scilicet in ratione totius, ut est mons aureus, vel hircocervus habens medietatem cervi, medietatem hirci.’ 16 Ibid.: ‘Aliquid autem, sive res nata esse vel quae est aliquid extra intellectum, quae dicitur “res a ratitudine”, adhuc non habet rationem generis aut praedicamenti sicut neque prius, sed dividitur divisione analogica in id quod est aliquid quod est ipsum esse, et in id quod est aliquid cui convenit aut natum est convernire esse.’

Scotus and Avicenna on What it is to Be a Thing 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

373

real things. According to Henry, however, such entities are the result of the combination of real things. As Henry said in his Quodl. VII, the reason why a golden mountain and a goatstag can be thought about is that, even though they do not have a proper essence (and so they are not things in the second, stronger sense of ‘thing’), their concepts nevertheless result from the combination of the concepts of two or more real essences (the concept of a mountain and the concept of gold, the concept of a goat and the concept of a stag): [A thing as an object of opinion] is nevertheless a real thing with regard to what a mountain is and what gold is and the like. For such a combination could not be in the intellect and could not be a being according to opinion unless its parts were something real, because the intellect cannot be moved by anything different.17 (Transl. mine)

It is our imagination, it seems, that carries out the combination of two distinct essences resulting in entities such as golden mountains and goatstags. As to our intellect, it thinks about such imaginary things as golden mountains and goatstags as if each of them were a single essence, even though they are just the result of a combination that cannot obtain in the extramental world.18 As a matter of fact, it is not immediately clear why Henry thought that this is the case. It may be admittedly difficult to gather enough gold to build up a mountain, but there seems to be nothing physically impossible about this. Probably, Henry would contend that it is part of the very concept of what it is to be a mountain that a mountain is constituted not entirely by gold. Similarly, it may be contended that it is part of the very concept of what it is to be a goat that what is a goat cannot be combined with a stag, because goats and stags have incompatible essential properties. Be that as it may, Henry’s conclusion is that something can be an object of thought if either it is a real thing, i. e. an instantiated or instantiable essence, or it is an imaginary combination of real essences. Accordingly, the weaker meaning of what it is to be a thing (i. e., a thing as an object of thought) is dependent on the stronger meaning of what it is to be a thing (i. e., a thing as an essence). We could not have concepts of things, in the weak sense of ‘thing’, if we had not been previously acquainted 17 Ibid., p. 27: ‘Est tamen res secundum veritatem quoad partes eius quae sunt mons et aurum et huiusmodi; aliter enim non posset totum esse in intellectu et ens secundum opinionem, nisi partes essent aliquid secundum veritatem, quia ab alio non potest moveri intellectus.’ The importance of this passage was stressed by Aertsen, Transcendental Thought, p. 5. 18 Henry of Ghent, Summa, a. 34, q. 2, pp. 174 – 5: ‘Ita quod ratio entis sit ratio primi conceptus obiective in intellectu, quia “quod quid est, est proprium obiectum intellectus” secundum Philosophum, ut etiam ratio rei a reor dictae non potest concipi ab intellectu – licet possit ab imaginatione – nisi sub ratione entis quidditativi … Quia autem verum est prima ratio qua aliquid est conceptibile ab intellectu, ut praedictum est, ideo tertia ratio veri fundatur in ratione entis quidditativi, quod dicitur res a ratitudine …’

374 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Giorgio Pini

with things in the strong sense. We could not think about goatstags if we did not have the concept of what a goat is and of what a stag is. The reason for this is clear, according to Henry. Only what has a real essence can act upon our mind and ‘move’ our mind, i. e. cause a concept in our mind. The general idea behind Henry’s position is that human beings are naturally geared or predisposed to think about real essences because our minds are such that they can be acted upon only by real essences.19 And Henry’s assumption is that what we can think about is primarily what acts upon our minds. Thus, it is true that we can think about imaginary beings such as golden mountains and goatstags, but these are deviant cases that must be accounted for in terms of imaginary combination of real essences. The scope of what our mind can think about is primarily defined by what is real, i. e. (for Henry) what either is or can become actual. Only secondarily and in a derivative sense can we think about what is not real, i. e. what neither is nor can become actual, and we can do that only to the extent to which what is not real is the result of a combination of what is real. A combination such as golden mountain or goatstag neither is nor can become actual, but its components are or can become actual. In conclusion, Henry’s division is basically equivalent to the early Aquinas’s distinction of the two meanings of ‘thing’, apart from an important difference. Something, in order to be thought about, does not have to be a quiddity or essence. Only its components do. Thus, Henry was not committed to the implausible claim that when I think about chimeras I think about a certain quiddity or essence. Still, conceivability is connected with the possession of an essence, even though in a roundabout way, according to Henry. Admittedly, there is no such an essence as the essence of a chimera or the essence of a golden mountain. But what it is to be a chimera and what it is to be a golden mountain must be analyzed into their components. And those components do have an essence. Thus, a necessary condition for something to be thought about or talked about is for it to have some sort of being – because, as Avicenna had said, nothing affirmative can be stated of what is absolutely non-being. And the sort of being that what is thought or talked about has is either essential being or dependent on essential being (i. e. an imaginary combination of two essences).

19 Admittedly, extramental things ‘act upon’ our minds in a complicated way, which is supposed to save our mind’s active role in the act of thinking. Henry, however, admitted that our mind is passive with regard to its object in the first stage of cognition. See Henry of Ghent, Quodlibet 9.2, p. 26: ‘Est igitur primo sciendum quod nihil intelligit aliquid nisi id quod est per se obiectum virtutis qua cognoscit, vel id cuius illud per se obiectum est ratio cognoscendi, quia virtus qua intellectus intelligit, quae est intellectus eius, passiva est, quae non movetur ad actum intelligendi nisi a per se obiecto’ (italics mine).

Scotus and Avicenna on What it is to Be a Thing 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

375

II In the third question of his Quodlibet, Scotus endorsed a distinction of the meanings of ‘thing’ and ‘being’ that is very similar to that adopted by Henry of Ghent, who was most probably Scotus’s immediate source. Like Henry, Scotus started with the remark that ‘thing’, in its most general meaning, is opposed to nothing. He then went on to notice that the term ‘nothing’ has two meanings. First, ‘nothing’ means what includes a contradiction, i. e. what is constituted by two incompatible notions, such as ‘round square’ (my example). In this sense of ‘nothing’, nothing is what cannot exist either in the mind (as an object of thought) or outside the mind (as a real thing). Accordingly, here we are not dealing with any object – either an object of thought or a thing in the real world. Rather, we are just dealing with two incompatible notions. When we say that a round square is nothing we are not saying that there is a special kind of object that has the property of not existing in reality; rather, we are saying that the notions round and square are such that they are logically incompatible.20 Second, ‘nothing’ means what does not and cannot exist extramentally.21 Since ‘thing’ (or ‘being’) means ‘not nothing’, corresponding to these two meanings of ‘nothing’ there are two meanings of ‘thing’ (or ‘being’). The first meaning of ‘thing’ and ‘being’ is ‘something that does not include a contradiction’.

20 Duns Scotus, Quodl., q. 3, n. 2, p. 113: ‘… hoc nomen res potest sumi communissime et strictissime. Communissime, prout se extendit ad quodcumque quod non est nihil. Et hoc potest esse dupliciter. Verissime enim illud est nihil quod includit contradictionem, et solum illud, quia illud excludit omne esse extra intellectum et in intellectu. Quod enim est sic includens contradictionem, sicut non potest esse extra animam, ita non potest esse aliquid intelligibile ut aliquod ens in anima, quia nunquam contradictorium cum contradictorio constituit unum intelligibile, neque sicut obiectum cum obiecto neque sicut modus cum obiecto.’ See also Ord. I, d. 43, q. unica, n. 15, ed. Vat. VI, p. 359: ‘Et ex hoc apparet quod falsa est imaginatio quaerentium impossibilitatem aliquorum quasi in aliquo uno, quasi aliquid unum – vel intelligibile vel qualecumque ens – sit ex se formaliter impossibile sicut Deus ex se formaliter est necesse esse. Nihil enim est tale primum in non-entitate, nec etiam entitatis oppositae tali non-entitati est intellectus divinus ratio possibilitatis oppositae; nec etiam intellectus divinus est praecisa ratio possibilitatis oppositae de nihilo, quia tunc teneret illud argumentum “de causis praecisis in affirmatione et negatione”. Sed omne “simpliciter nihil” includit in se rationes plurium, ita quod ipsum non est primo nihil ex ratione sui, sed ex rationibus illorum quae intelligitur includere, propter formalem repugnatiam illorum inclusorum plurium; et ista ratio repugnatiae est ex rationibus formalibus eorum, quam repugnatiam primo habent per intellectum divinum’. It should be noticed that in these passages, Scotus took ‘thing’ (res) and ‘being’ (ens) as synonymous. 21 Quodl, q. 3, n. 2, p. 113: ‘Alio modo dicitur nihil quod nec est nec esse potest aliquod ens extra animam.’

376 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Giorgio Pini

In this sense of ‘thing’ and ‘being’, both purely mental and extramental entities are things and beings: Being [ens] or thing [res] in the first of its very broad meanings, therefore, covers anything that does not include a contradiction whether it be a conceptual being [ens rationis], i. e., having being or existence only in the thinking intellect, or a real being [ens reale], i. e., having some entity outside the consideration of the intellect.22 (Transl. Alluntis and Wolter)

By contrast, the second meaning of ‘thing’ (or ‘being’) is ‘something that has or can have some entity independently of the intellect’s consideration’. In this sense of ‘thing’ and ‘being’, only objects that exist or can exist in the extramental world are said to be things or beings.23 Scotus held that there are some objects that are things in the first, more general sense of the word ‘thing’ but not in the second, less general sense of the word ‘thing’. His example is logical concepts and conceptual (or rational) relations.24 Logical concepts are concepts such as species and genus, i. e. the socalled ‘second intentions’. Conceptual (or rational) relations are relations that hold among things only because of the way we describe them but are not grounded on any real feature and as such can be acquired and lost without any real change in either of the extremes (an example is the relation of being to the left of something; depending on our position, a chair can be described as to the left or to the right of a table, even though the chair has undergone no change of position). Let us focus on logical concepts. There is no contradiction included in the notion of what it is to be a species or what it is to be a genus. Accordingly, we can think about what it is to be a species and what it is to be a genus. Nevertheless, there is nothing in the extramental world whose essence is to be a species or to be a genus. Rather, in the extramental world there are real things such as cats and human beings, which can be classified in species and genera. But to be a species or to be a genus is not something in the world; rather, it is a conceptual (or rational) relation holding between concepts representing things

22 Ibid.: ‘Ens ergo vel res isto primo modo accipitur omnino communissime, et extendit se ad quodcumque quod non includit contradictionem, sive sit ens rationis, hoc est praecise habens esse in intellectu considerante, sive sit ens reale, habens aliquam entitatem extra considerationem intellectus’. The English translation is taken from John Duns Scotus, God and Creatures, p. 61. 23 Ibid.: ‘Et secundo accipitur in isto membro minus communiter pro ente quod habet vel habere potest aliquam entitatem non ex consideratione intellectus’. 24 Ibid.: ‘Et istorum duorum membrorum … primum videtur valde extendere nomen rei, et tamen, ex communi modo loquendi, satis probatur. Communiter enim dicimus intentiones logicas esse res rationis, et relationes rationis esse res rationis, et tamen ista non possunt esse extra intellectum.’ On second intentions, see G. Pini, Categories and Logic, pp. 45 – 137.

Scotus and Avicenna on What it is to Be a Thing 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

377

in the world (i. e. objects that are said to be things in the second, less general meaning of ‘thing’). I think that Scotus would say that chimeras and goatstags are things in this first, more general sense of ‘thing’. Chimeras and goatstags are nothing in the sense that they are nothing actual. Scotus also agreed with Henry that chimeras and goatstags cannot become actual. Scotus’s reason for this, however, is not that there is no single essence corresponding to the term ‘chimera’ or the term ‘goatstag’. Rather, Scotus’s claim is that the notion of a chimera and the notion of a goatstag are such that they cannot be instantiated in the extramental world, and apparently he takes this as a primitive fact. There is a contradiction between what it is to be a chimera and to exist in the extramental world. The notion of existing in the extramental world is ‘repugnant’ or logically incompatible with what it is to be a chimera. It seems that the impossibility of existing in the extramental world is just part of what it is to be a chimera. Nevertheless, even though the notion of a chimera and the notion of extramental existence are in contradiction with each other, there is no contradiction included in the notion of what it is to be a chimera. Thus, a chimera can be an object of thought, even though it cannot possibly exist.25 Scotus claimed that it is in the first meaning of ‘being’ and ‘thing’ that being and thing are the first object of the intellect, because the intellect can think about anything that does not include a contradiction. Scotus, however, was hesitant to say whether Avicenna spoke of ‘thing’ in this most general sense in his Metaphysics I.5. At first, Scotus stated that when Avicenna spoke of ‘thing’ and ‘being’, he spoke of both in the second, less general sense of those words, i. e. as what has or can have existence outside the mind. Accordingly, when Avicenna said that being and thing are the first concepts occurring to our mind and that those concepts are common to everything, he should be interpreted as saying that the first concepts our mind acquires are the concepts of things that exist or can exist.26 Later in the same question, however, Scotus showed some hesitation in construing Avicenna’s intention. He conceded that Avicenna might be speaking of ‘being’ and ‘thing’ in the first, more general sense. Accordingly, 25 Scotus considered entities such as chimeras as ‘fictitious beings’ (figmenta) and he defined what is not a fictitious being as that for which it is possible to exist. This sort of existence is extramental existence. Accordingly, a fictitious being is an object of thought such that real existence is repugnant to it, i. e. for which it is impossible to exist extramentally. See Ord. I, d. 36, q. unica, nn. 48 – 50, ed. Vat. VI, pp. 290 – 91; nn. 60 – 63, pp. 296 – 7. 26 Quodl. q. 3, n. 2, p. 114: ‘In secundo autem membro istius primi membri, dicitur res quod habere potest entitatem extra animam. Et isto modo videtur loqui Avicenna, 1 Metaphysicae, c. 5, quod ea quae sunt communia omnibus generibus sunt res et ens, nec potest illud intelligi de vocabulis in una lingua, quia in unaquaeque lingua est unus conceptus indifferens ad omnia illa quae [ed.: qua] sunt extra animam …’

378 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Giorgio Pini

Avicenna should be interpreted as saying that the first concept common to everything is the concept of what is non-contradictory. Scotus, however, added that Avicenna might also be referring to both senses of ‘thing’ and ‘being’. In any case, he was quite certain that Avicenna was referring at least to the second sense of ‘thing’ and ‘being’, i. e. as what has or can have extramental existence.27 Scotus’s distinction between the two main meanings of ‘thing’ and ‘being’ is clearly indebted to Henry of Ghent. Through Henry, Scotus was also giving his interpretation of Avicenna’s teaching on ‘thing’ and ‘being’ and on Avicenna’s distinction between two descriptions of what it is to be a thing. There are, however, two important points with regard to which Scotus’s account of the meaning of the word ‘thing’ (and, correspondingly, Scotus’s interpretation of Avicenna) is different from Henry’s treatment. The first difference concerns Scotus’s interpretation of the first meaning of ‘thing’. The second difference concerns Scotus’s interpretation of the second meaning of ‘thing’. As to the first meaning of the word ‘thing’, Scotus’s description of a thing as what does not include any contradiction does not make any reference to the notion of essence. Henry of Ghent had maintained that a thing as a purely mental item is not a single essence but is nevertheless an imaginary combination of several real essences. As a consequence, Henry held that even items that are things only in the sense of possible objects of thought can be thought about at all because there are things that have real essences, and the latter things are the elements out of which purely mental things are constituted. Thus, Henry of Ghent’s assumption was still that the mind is naturally turned to grasping essences, i. e. things that either are or can become actual constituents of the world. By contrast, Scotus shifted the focus from essences to what is not contradictory. In the question we are currently taking into consideration, he explicitly stated that the object of the intellect is anything that does not include a contradiction. Thus, he defined the object of the intellect by way of a merely formal and logical requirement. Quite simply, in order for something to be thinkable, it is sufficient for it not to include a contradiction. It is not necessary for it to be an essence or a combination of essences. Scotus’s method to find out what can be thought about and what cannot be thought about has a considerable advantage. Scotus could decide if something is thinkable by simply testing whether it satisfies a formal requirement. Thus, logical intentions and conceptual or rational relations can be accounted for as 27 Quodl. q. 3, n. 3, pp. 114 – 5: ‘Habemus ergo primum membrum sic, scilicet communissime, bipartitum in illud, scilicet quod non includit contradictionem, qualecumque esse habeat, et in illud quod habet vel habere potest proprium esse extra intellectum, et isto modo, vel utroque modo, vel saltem secundo modo accipit Avicenna rem et ens, ut dictum est’.

Scotus and Avicenna on What it is to Be a Thing 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

379

objects of thought just because they do not include any contradiction. There is no need to ground logical concepts and conceptual or rational relations on real essences in order to account for their thinkability. There is no need to know what there is out in the world to find out whether something is thinkable or not. The test of thinkability is the satisfaction of the logical requirement of being internally consistent, i. e. non-contradictory. Concepts do not have to be tested against extramental essences in order to be thinkable. Accordingly, Scotus’s approach, has a remarkable consequence. According to Scotus, the mere fact that we can think about something leaves any question concerning its ontological status undecided. Just because I am able to think about something I am not in a position to say that that thing exists extramentally or even that it can exist extramentally. Undeniably, logical intentions, conceptual relations and chimeras can be thought about. Nevertheless, they cannot exist extramentally. Thus, there are things (in the first meaning of ‘thing’) that God cannot bring about in extramental existence.28 Ontological questions concerning the real possibility of something cannot be decided by virtue of merely logical criteria. In other words, Scotus distinguished between, on the one hand, logical possibility as a formal relation holding between concepts and, on the other hand, real or metaphysical possibility, if we interpret ‘metaphysical possibility’ as the property of not being incompatible with actual existence in at least one possible world (be it the actual world or another world).29 There are things that satisfy the logical criterion of not including a contradiction and there are things that satisfy the metaphysical criterion of being compatible with actual existence. These, however, are not two kinds of things that can be compared as more or less real; rather, these are things in two different senses of the word ‘thing’. When we say that a chimera or a logical concept is less real than a human being we are making what in contemporary jargon is called a ‘category mistake’. Chimeras and logical concepts are not less real than human beings; rather, they are not real at all. To be real is a predicate that can be applied to what is a thing only in the second, less general sense of thing, i. e. as what either is or can become actual. As to the second meaning of ‘thing’, Scotus agreed with Henry of Ghent that the word ‘thing’, in its second meaning, refers to all real essences, because, necessarily, if x is a real essence, x is a thing in the second meaning of ‘thing’, i. e. as that which has or can have extramental existence. From what Scotus said 28 Nevertheless, Scotus still endorsed the common claim that God can do whatever is noncontradictory, for he held that it is contradictory for a logical concept or a chimera to exist extramentally. 29 For a characterization of logical and metaphysical possibility, as well as of nomological possibility, see Conceivability and Possibility, eds. T.S. Gendler and J. Hawthorne, Oxford: Clarendon Press, 2002, pp. 4 – 5.

380 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Giorgio Pini

elsewhere, however, it appears that he rejected the converse claim that, necessarily, if x can have extramental existence, then x is an essence. It is not difficult to see why this is the case if we consider that God could have created some kinds of items that He actually did not create. The things that God could have created but did not are things in the second sense of the word ‘thing’, because it is possible for them to exist extramentally, or, as Scotus said, extramental existence is not repugnant to (i. e., incompatible with) them. Nevertheless, the things that God could have created but did not create are not essences, according to Scotus. Something is an essence if and only if it was made the object of God’s act of creation. As a consequence, it appears that Scotus, unlike Henry of Ghent, held that there are no possible, i. e. non-actual, essences (even though, given an actual essence, there are possible, i. e. non-actual individuals that could but do not instantiate that essence).30 There are, however, things that could have been created and could exist but do not. Such non-actual things are called ‘things’ because existence is not repugnant to (i. e., incompatible with) them, but they are not essences. Scotus argued that the same point holds even for essences, i. e. things that God actually created, when they are considered as they were in themselves before God created them. By contrast, Henry of Ghent had identified the second way something is said to be a thing (as a res a ratitudine) with what it is to be an essence. According to Henry, something can have extramental existence because it is a thing in the sense of being or having an essence. Admittedly, an essence can fail to be instantiated. In Henry’s jargon, an essence does not have to have existential being (esse existentiae). Necessarily, however, if something can have extramental existence, it is a thing in the sense of ‘essence’, i. e. it has essential being (esse essentiae). Against Henry’s position, Scotus held that, before creation, a human being and a chimera both lack an essence. Thus, before creation, the only difference between a human being and a chimera is that it is not impossible for a human being to be brought to existence, whereas it is impossible for a chimera to be brought to existence. But there is an essence human being only once God creates actual human beings in the world. Accordingly, Scotus held that to be metaphysically possible (i. e., to be a thing in the second sense of ‘thing’ that Scotus had distinguished in his Quodlibet, q. 3) and to be an essence are two distinct notions. Whereas the latter entails the former, the former does not entail the latter. Thus, what is metaphysically possible, i. e. what exists or can exist extramentally, is not necessarily an essence. Rather, it is a contingent fact that something metaphysically possible, for example the concept human being, is also 30 Scotus’s position is, in this respect, just the reverse of that of Henry of Ghent. For Henry, there are non-actual essences but only actual individuals, because what accounts for an essence’s being instantiated in a certain individual is that such an individual actually exists. By contrast, for Scotus there are non-actual individuals but only actual essences.

Scotus and Avicenna on What it is to Be a Thing 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

381

an essence, because this is a fact that depends entirely on God’s will to create human beings. There is nothing in the concept human being that entails its being a certain essence. What is included in the concept human being is merely the possibility of being a certain essence.31 To sum up, Scotus distinguished between (1) what is internally consistent, i. e. logically possible; (2) what is not incompatible with extramental existence, i. e. metaphysically possible; and (3) what has an essence, i. e. what belongs to a natural kind. In order for a thing (or a being) to be thinkable, it is sufficient to be a thing according to the first meaning of ‘thing’. What is a thing (or a being) in this first meaning, however, is not necessarily compatible with extramental existence. In turn, what is a thing (or a being) in the second meaning of ‘thing’, i. e. as something with which extramental existence is not incompatible, does not necessarily have an essence. Only an actual thing has an essence. Accordingly, both actual entities and nonexistent possibles are things in the second sense of ‘thing’, even though the former lack and the latter have an essence. To say of something that it is a thing or a being – even in the second, stronger sense of ‘thing’ or ‘being’ – does not tell us anything about its ontological status or even whether it has some ontological status.

III So far I have considered what Scotus took a thing to be and how he read Avicenna accordingly. As I have indicated, Scotus held that the term ‘thing’ is equivocal between something that does not include a contradiction and something that does or can exist extramentally. And he maintained that probably what Avicenna meant by ‘thing’ was the latter. Typical of Scotus’s approach is also his view that the first sense of ‘thing’ as something internally consistent is not dependent on the second sense of ‘thing’ as something that does or can exist extramentally. Now I would like to turn to the third issue I mentioned at the beginning of this paper, namely what role our concept thing or being as something that is or can be extramental plays in cognition. It turns out that the concept thing or being as something at least possibly extramental plays an extremely important role in our cognition, both of the material and of the immaterial world. In this respect, Scotus qualified what he read in Avicenna. At the beginning of Metaphysics I.5, Avicenna made the famous claim that thing and being are the first notions impressed in our souls.32 Scotus, however, was committed to the view that, in our current situation, our 31 Ord. I, d. 36, q. unica, nn. 60 – 63, ed. Vat. VI, pp. 296 – 7. 32 Avicenna, The Metaphysics of ‘The Healing’, I, 5, 1, p. 22. See the Latin translation in Avicenna Latinus, Liber de philosophia prima, I, 5, vol. 1, p. 31.

382 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Giorgio Pini

intellect is dependent on the senses for the information it receives about the world.33 Now Aristotle had claimed that only sensible qualities such as colors, sounds and the like are the per se objects of the senses.34 And Scotus fully agreed with this claim. More precisely, Scotus argued that our senses are directly acquainted not with individual instances of sensible qualities, such as this singular instance of red or this singular instance of a certain sound. Rather, Scotus held that our senses are directly acquainted with replicable features – things such as this hue of red or this type of sound.35 These are what Scotus, following a common usage, called species specialissimae. Since all cognitive content comes to the intellect through the senses, it follows that the first thing the intellect cognizes, in the order of origin, are replicable sensible features, such as this kind of color or this kind of sound – the species specialissimae of sensible qualities.36 The concepts being and thing become relevant only at the second stage of the cognitive process, when we pass to what Scotus called ‘cognizing something in a distinct way’, i. e. when we provide descriptions and definitions of what we cognize.37 At first, however, we do not grasp the objects of our cognition as beings and things, but as this sort of color, this sort of sound, and so on–namely, as the least universal kinds of sensible qualities. So how do we get from these sensible qualities to the concepts being and thing? Scotus held that we arrive at the concepts being and thing, quite simply, by abstraction from sensible qualities.38 Accordingly, Avicenna’s claim that thing and being are the first notions impressed in our minds should be strongly qualified. On the face of it, such a claim seems to be in contrast with Aristotle’s view that the first objects of our cognition are sensible qualities. Thus, Scotus reinterpreted Avicenna’s claim as referring to what is cognized in a distinct way, i. e. when we set to provide descriptions and definitions of what we cognize. Then and only then do being and thing have a primacy as the most general concepts that are contained in the description of any other less general concept. For example, if I have to give a definition or description of a human being, the most general description I can provide is probably something of this sort: ‘something that has a head, two legs, etc. and displays an intelligent behavior’. 33 Scotus, Ord. I, d. 3, p. 3, q. 1, n. 392, ed. Vat. III, p. 239. See also Ord. I, d. 3, p. 1, q. 1 – 2, nn. 35 – 6, ed. Vat. III, pp. 21 – 2; Ord. II, d. 11, q. unica, n. 13, nn. 28 – 31, ed. Vat. VIII, pp. 214 – 5, 222 – 4. 34 Aristotle, Cat. 8, 9a35-b7; De gen. et corr. II, 2, 329b19. 35 Scotus, Quaestiones super Metaphysicam, VII, q. 15, nn. 20 – 21, OPh IV, pp. 301 – 2; Ord. II, d. 3, p. 1, q. 1, n. 21, ed. Vat. VII, pp. 399 – 400. See South, Scotus and the Knowledge; Pini, Scotus on the Objects. 36 Ord. I, d. 3, p. 1, q. 1 – 2, nn. 72 – 3, ed. Vat. III, pp. 49 – 50. 37 Ord. I, d. 3, p. 1, q. 1 – 2, n. 80, ed. Vat. III, pp. 54 – 5. 38 Ord. I, d. 3, p. 1, q. 3, nn. 139 – 40, ed. Vat. III, pp. 86 – 8. See also Lect. I, d. 3, p. 1, q. 1 – 2, n. 110, ed. Vat. XVI, p. 265.

Scotus and Avicenna on What it is to Be a Thing 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

383

In that description, there is no way to avoid such general labels as ‘something’, ‘thing’, ‘being’. This is the less general meaning of ‘thing’ and ‘being’ Scotus had distinguished in the passage we have considered above, i. e. ‘thing’ and ‘being’ as something that has or can have extramental existence. Scotus held that it is thanks to this concept thing and being that we can extend our cognition beyond the species specialissimae of the sensible qualities we are directly acquainted with. I see black and I touch soft. How do I pass from these sensations to the perception of a cat? Scotus’s answer is that I infer the presence of something underlying the sensible accidents. In this inference, the notions of thing and being play a central role. After noticing that some sensible qualities, say a certain color and a certain tactile impression, constantly occur together, I infer that there must be something that accounts for their constantly occurring together. Thus, I conclude that these sensible qualities occur together because they inhere in the same thing, which I call a substance. With regard to that underlying thing, I only know that it is something – a thing or being – that accounts for the unity of certain sensible qualities, i. e. something in which certain sensible qualities inhere.39 Accordingly, without the concept of being or thing, it would be impossible for us to reconstruct the structure of the world. Scotus admitted that the fact that we cannot grasp substances directly and we rather have to infer that there are substances (if we want to account for the regular conjoined occurrence of certain sensible qualities) is a lamentable situation. He also thought that it is a contingent situation. Human beings were not originally intended to cognize the world in this way. Rather, human beings were supposed to get a direct grasp of substances – to know cats, dogs and other human beings not on the basis of their sensible qualities but by inspecting their very essences. To some extent, this will be the condition of the blessed in the next life.40 The fact that we currently can cognize substances only by inference from sensible qualities and through the concept thing and being is probably a consequence of the fall of Adam. After the fall, our intellect became dependent on the senses as its only source of information.41 Thus, Scotus thought that Aristotle had given us an accurate description of the postlapsarian situation, but not of our natural condition. Scotus, however, held that, even given the cognitive limitations that characterize our current situation, we are still able to get a fairly accurate picture of the world. What allows us to reach such a result is our possession and use of the most general concepts being, thing, something and the like. Thus, Avicenna had come across a very important point when he had emphasized the role of thing and being as our most basic concepts, even though 39 Ord. I, d. 22, q. unica, n. 7, ed. Vat. V, pp. 344 – 5; Quaestiones super Metaphysicam, II, q. 2 – 3, n. 115, OPh III, p. 233. See Pini, Scotus on Doing Metaphysics in statu isto. 40 Ord. II, d. 11, q. unica, n. 31, ed. Vat. VIII, p. 224. 41 See above, n. 33.

384 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Giorgio Pini

he had not been aware of the reason why those concepts are so important in our current situation, namely because they allow us to overcome our current cognitive limitations and to reconstruct the structure of reality on the basis of what we are directly acquainted with, i. e. sensible qualities. In that respect, Avicenna went beyond Aristotle. Avicenna’s views concerning thing and being also went beyond Aristotle with regard to our current cognition of God. When Avicenna posited being and thing as the objects of our intellect, taken in all their generality, he did not take into account that in the current situation we are indeed limited to having cognitive access only to sensible things. Thus, Avicenna was ignorant of the fact that there is a sharp distinction between what our condition ought to be and what it is now in the current, postlapsarian situation. We were not intended to rely on the senses as we do now. Avicenna judged our cognitive powers according to their own nature, not according to their current condition. According to Scotus, Avicenna did so because he was being unwittingly influenced by his own religious belief, which required that we should be able to see God in the next life and thus that we should not be constitutionally limited to having cognitive access only to sensible things.42 Scotus held that the influence of Islamic beliefs on Avicenna’s philosophical positions, even though unconscious, was beneficial. By ignoring our present cognitive limitations, Avicenna gave us a glimpse of true nature, i. e. of what we were originally intended to know and what we shall eventually know in the next life. Thus, I think we should conclude that Scotus saw Avicenna as a salutary complement to Aristotle. Both Aristotle and Avicenna were ignorant of the difference between the present, postlapsarian condition and our original condition. Accordingly, they both failed to appreciate the consequences that the fall of Adam had on our cognitive powers. But their ignorance had opposite results. Aristotle (and, to some extent, his follower Aquinas) took our present cognitive limitation to sensible things as normal and constitutive of the human condition. Accordingly, Aristotle and Aquinas thought that the proper object of our intellect is the quiddity of sensible things, i. e. that our intellect is 42 Ord. prol., p. 1, q. unica, n. 33, ed. Vat. I, p. 19: ‘Ad aliud, negandum est illud quod assumitur, quod scilicet naturaliter cognoscitur ens esse primum obiectum intellectus nostri, et hoc secundum totam indifferentiam entis ad sensibilia et insensibilia, et quod hoc dicit Avicenna quod sit naturaliter notum. Miscuit enim sectam suam – quae fuit secta Machometi – philosophicis, et quaedam dixit ut philosophica et ratione probata, alia ut consona sectae suae: unde expresse ponit libro IX Metaphysicae cap. 7 animam separatam cognoscere substantiam immaterialem in se, et ideo sub obiecto primo intellectus habuit ponere substantiam immaterialem contineri. Non sic Aristoteles; sed secundum ipsum, primum obiectum intellectus nostri est vel videtur esse quiditas sensibilis, et hoc vel in se sensibilis vel in suo inferiori; et haec est quiditas abstrahibilis a sensibilibus’.

Scotus and Avicenna on What it is to Be a Thing 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

385

constitutionally limited to what is sensible. By contrast, Avicenna correctly singled out being as the object of our cognitive powers, but he failed to consider that in the current situation our grasp is, as a matter of fact, limited to sensible things. Aristotle focused on what we are and neglected what we were and will be. By contrast, Avicenna focused on what we should and will be, but he ignored our present cognitive limitations. His neglect of our present state had the fortunate effect of putting much emphasis on the concepts thing and being as the objects of our intellect. It is by way of those concepts that we can measure up what we lost, i. e. direct cognitive access to substances, and what we are supposed to reach in the next life, i. e. direct cognitive access to God in His individuality. Thanks to the concepts thing and being we can also reconstruct the structure of reality, to the extent that we can do this in our present situation. By the concepts thing and being, we can even get a sort of pale anticipation of the knowledge of God’s essence that Scotus thought we shall experience in the next life.

Bibliography Primary Sources Avicenna, Liber de philosophia prima sive de Scientia divina, 2 vols, ed. S. Van Riet, Leuven/Leiden: Peeters/Brill, 1977 and 1980. –– , The Metaphysics of ‘The Healing’, transl. and ed. M. Marmura, Provo, Utah: Brigham Young University Press, 2005. Bonaventure, In secundum librum Sententiarum (= In II Sent.), in id., Opera omnia II, Quaracchi: Typographia Collegii S. Bonaventurae, 1885. Henry of Ghent, Summa (Quaestiones ordinariae), a. 34, q. 2, in id., Opera omnia XXVII, ed. R. Macken, Leuven: Leuven University Press, 1991. –– , Summa quaestionum ordinariarum a. 21, q. 4, in id., Summa quaestionum ordinariarum, Paris, 1520; reprinted St. Bonaventure, New York: Franciscan Institute, 1953. –– , Quodlibet V, q. 2, in id., Quodlibeta, Paris, 1518; reprinted Louvain: Publications de l’Institut de Philosophie, 1961. –– , Quodlibet VII, q. 1 – 2, in id., Opera omnia XI, ed. G. Wilson, Leuven: Leuven University Press, 1991. –– , Quodlibet IX, q. 2, in id., Opera omnia XIII, ed. R. Macken, Leuven: Leuven University Press, 1983. John Duns Scotus, Opera omnia, Vatican City: Typis Polyglottis Vaticanis, 1950– (= ed. Vat.). –– , Quaestiones super libros Metaphysicorum Aristotelis, in id., Opera philosophica III and IV (= OPh), St. Bonaventure, New York: Franciscan Institute, 1997 – 9. –– , Quodlibet, in id., Opera omnia, 25 – 6, Paris: L. Vivs, 1895. –– , God and Creatures. The Quodlibetal Questions, transls and eds F. Alluntis and A.B. Wolter, Princeton/London: Princeton University Press, 1975.

386 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Giorgio Pini

Thomas Aquinas, Scriptum super primum librum Sententiarum (= In I Sent.), ed. P. Mandonnet, Paris: Lethielleux, 1929. –– , Scriptum super secundum librum Sententiarum (= In II Sent.), ed. P. Madonnet, Paris: Lethielleux, 1929. –– , De potentia (= De pot.), ed. P.M. Pession, in id., Quaestiones disputatae, vol. 2, Rome/Turin: Marietti, 1965, pp. 1 – 276. –– , Expositio super duodecim libros Metaphysicorum Aristotelis (= In Met.) eds. M.R. Cathala and R.M. Spiazzi , Rome/Turin: Marietti, 1964.

Secondary Sources J. Aertsen, Transcendental Thought in Henry of Ghent, in Henry of Ghent. Proceeding of the International Colloquium on the Occasion of the 700th Anniversary of His Death (1293), ed. W. Vanhamel, Leuven: Leuven University Press, 1996, pp. 1 – 18. –– , Avicenna’s Doctrine of the Primary Notions and its Impact in Medieval Philosophy, in Islamic Thought in the Middle Ages: Studies in Text, Transmission and Translation, in Honour of Hans Daiber, eds A. Akasoy and W. Raven, Leiden: Brill, 2008, pp. 21 – 42. G.C. Anawati, Saint Thomas d’Aquin et la Mtaphysique d’Avicenne, in Saint Thomas Aquinas, 1274 – 1974: Commemorative Studies, vol. 1, eds A. Maurer et al., Toronto: Pontifical Institute of Mediaeval Studies, pp. 449 – 65. D. Black, Mental Existence in Thomas Aquinas and Avicenna, Mediaeval Studies, 61, 1999, pp. 45 – 79. A. De Libera, La querelle des universaux. De Platon  la fin du Moyen Age, Paris: Editions du Seuil, 1996. M.-T. Druart, Ibn Sina (Avicenna) and Duns Scotus, in Duns Scotus, Philosopher, eds M.B. Ingham and O. Bychkov, Archa Verbi, Subsidia 3, Mnster: Aschendorff, 2010, pp. 13 – 27. . Gilson, Avicenne et le point de depart de Duns Scot, Archives d’histoire doctrinale et littraire du Moyen Age, 2, 1927, pp. 89 – 149, reprinted in . Gilson, Pourquoi saint Thomas a critiqu saint Augustin, suivi de Avicenne et le point de depart de Duns Scot, Paris: Vrin, 1986, pp. 129 – 89. –– , Avicenne en Occident au Moyen Age, Archives d’historire doctrinale et littraire du Moyen Age, 44, 1969, pp. 89 – 121. J. Hamesse, Res chez les auteurs philosophiques des 12e et 13e sicles ou le passage de la neutralit  la spcificit, in Res. IIIo Colloquio Internazionale del Lessico Intellettuale Europeo, eds M. Fattori and M. Bianchi, Rome: Edizioni dell’Ateneo, 1982, pp. 91 – 104. J. Jolivet, Aux origines de l’ontologie d’Ibn Sina, in tudes sur Avicenne, eds J. Jolivet and R. Roshdi, Paris: Les Belles Lettres, 1984, pp. 11 – 18, reprinted in J. Jolivet, Philosophie mdivale arabe et latine, Paris: Vrin, 1995, pp. 221 – 36. M. Marmura, Avicenna on Primary Concepts in the Metaphysics of his al-Shifa¯’, in Logos Islamikos: Studia Islamica in honorem Georgii Michaelis Wickens, eds R.M. Savory and D.A. Agius, Toronto: Pontifical Institute of Mediaeval Studies, 1984, pp. 219 – 39. –– , Avicenna’s Chapter on Universals in the Isagoge of his Shifa’, in Islam: Past Influence and Present Challenge, eds A.T. Welch and P. Cacia, Edinburgh: Edinburgh University Press, 1979, pp. 34 – 56.

Scotus and Avicenna on What it is to Be a Thing 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

387

–– , Quiddity and Universality in Avicenna, in Neoplatonism and Islamic Thought, ed. P. Morewedge, Albany, New York: SUNY Press, 1992, pp. 77 – 87. G. Pini, Categories and Logic in Duns Scotus: An Interpretation of Aristotle’s ‘Categories’ in the Thirteenth Century, Leiden: Brill, 2002. –– , Scotus on the Objects of Cognitive Acts, Franciscan Studies, 66, 2008, pp. 281 – 315. –– , Scotus on Doing Metaphysics in statu isto, in Duns Scotus, Philosopher, eds M.B. Ingham and O. Bychkov, Archa Verbi, Subsidia 3, Mnster: Aschendorff, 2010, pp. 29 – 55. P. Porro, Possibilit ed esse essentiae in Enrico di Gand, in Henry of Ghent. Proceedings of the International Colloquium on the Occasion of the 700th Anniversary of His Death (1293), ed. W. Vanhamel, Leuven: Leuven University Press, 1996, pp. 211 – 53. –– , Universaux et esse essentiae: Avicenne, Henri de Gand et le ‘Troisime Reich’, Cahiers de philosophie de l’Universit de Caen, 38 – 9, 2002, pp. 9 – 51. –– , Duns Scot et le point de rupture avec Avicenne, in Duns Scot  Paris 1302 – 2002, eds. O. Boulnois et al., Turnhout: Brepols, 2004, pp. 195 – 218. G. Sondag, La rception de la Mtaphysique d’Avicenne par Duns Scot, in Wissen ber Grenzen: Arabisches Wissen und Lateinisches Mittelalter, ed. A. Speer, Berlin/New York: Walter de Gruyter, 2006, pp. 591 – 611. J.B. South, Scotus and the Knowledge of the Singular Revisited, History of Philosophy Quarterly, 19, 2002, pp. 125 – 47. J.R. Teske, Henry of Ghent’s Debt to Avicenna’s Metaphysics, The Modern Schoolman, 85, 2007, pp. 51 – 70. C. Vansteenkiste, Avicenna-Citaten bij S. Thomas, Tijdschrift voor Philosophie, 15, 1953, pp. 457 – 507. J.F. Wippel, The Latin Avicenna as a Source of Thomas Aquinas’s Metaphysics, Freiburger Zeitschrift fr Philosophie und Theologie, 37, 1990, pp. 51 – 90. –– , The Reality of Nonexisting Possibles According to Thomas Aquinas, Henry of Ghent, and Godfrey of Fontaines, Review of Metaphysics, 34, 1981, pp. 729 – 58, reprinted as Thomas Aquinas, Henry of Ghent and Godfrey of Fontaines on the Reality of Nonexisting Possibles, in J.F. Wippel, Metaphysical Themes in Thomas Aquinas, Washington, DC: Catholic University of America Press, 1984, pp. 163 – 89. –– , The Metaphysical Thought of Thomas Aquinas: From Finite Being to Uncreated Being, Washington, DC: Catholic University of America Press, 2000. R. Wisnovsky, Avicenna’s Metaphysics in Context, Ithaca, New York: Cornell University Press, 2003.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Index of Avicenna’s Works with Passages Cited

al-Birr wa l-itm (Piety and Sin) ¯ 22n pp. 360-68: ˘

al-Hikma al- Aru¯diyya (Philosophy for ˙ ¯ d¯ı) ˙ Aru v ˙ fol. 2 : 31n fols 3v-4r : 31n al-Isˇa¯ra¯t wa-l-tanbı¯ha¯t (Pointers and Reminders) (ed. Dunya¯) vol. II, § 3.1, pp. 344-5: 106n vol. II, § 3.16, pp. 404-8: 102 vol. II, § 3.17: 105n vol. II, § 3.19-20: 111n vol. II, § 3.19: 107, 108, 112, 113, 115 vol. II, § 3.19, pp. 415-20: 107 vol. II, § 3.20: 110 vol. II, § 3.20, p. 422: 100 vol. II, § 3.20, pp. 422-4: 100 vol. II, § 3.22, pp. 429–30: 105 vol. III, § 4.12-15, pp. 23-7: 12n vol. III, § 4.28, p. 53: 98, 99n vol. III, § 7.13, pp. 275-6: 114 vol. III, § 7.14-16: 115 vol. III, § 7.14: 115 vol. III, § 7.15: 115 vol. III, § 7.16, pp. 279-80: 115 vol. III, § 7.17, pp. 281-5: 117 vol. IV, pp. 153-9: 19n al-Isˇa¯ra¯t wa-l-tanbı¯ha¯t (ed. Forget) IV: 30 IV, 29: 28 pp. 138-9: 31n pp. 140-41: 31n al-Isˇa¯ra¯t wa-l-tanbı¯ha¯t (transl. Inati) pp. 83-4: 295

˘

al-Mabda wa-l-ma a¯d (Provenance and Destination) I, 1, p. 1: 7n pp. 2-3: 31n II, 1-8: 13 III, 16-20: 13 ˘

˘

Risa¯la l-Adhawiyya fı¯ l-ma a¯d (Adhawiyya ˙ Destination) ˙˙ Treatise˙ on p. 197: 15n

˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50

al-Masˇriqiyyu¯n (The Easterners) p. 8: 7n Muba¯hata¯t (Discussions) ¯ pp.˙87-8, §§ 150-54: 19 p. 94, §§ 181-2: 39n p. 112, § 261: 17n pp. 140-41, §§ 386-90: 17n pp. 155-6, §§ 427-8: 19 p. 226, § 674: 17n pp. 271-2, § 787: 17n pp. 301-2, § 844-6: 17 p. 316, § 888: 19 p. 318, §§ 892-3: 19 p. 366, § 1141: 17n al-Nagˇa¯t (The Salvation) (ed. Fakhry) pp. 54-5: 296n p. 245: 52n pp. 320-26: 161n al-Nagˇa¯t (ed. ‘Amı¯ra) vol. II, pp. 153-5: 157n vol. II, pp. 165-6: 157n Sˇarh Kita¯b Utu¯lu¯gˇiya¯ (Commentary on the ¯ ˙ Theology) pp. 35-74 (47-9): 14 pp. 46-7: 17n pp. 56-7: 17n pp. 49-50: 18 Fı¯ l-Siya¯sa l-manziliyya (On Domestic Politics) pp. 232-60 (240): 21n

390 ˘ ˘

Sˇifa¯ (The Cure) Sˇifa¯ : al-Mantiq (Logic): al-Madhal (Isago˙ ˘ ge) (ed. Anawati) I, 1, p. 10: 23n I, 2, p. 15: 28n

V, 6, p. 234: 182n V, 6, p. 245: 190n V, 6, p. 248: 183n al-Nafs (Latin, ed. Van Riet) V, 3: 346n V, 3, pp. 105f: 346n Sˇifa¯ : al-Ila¯hiyya¯t (Metaphysics) (ed. Marmura) I: 289 I, 1-4: 8 I, 2: 10 I, 2, p. 7: 70n I, 2, p. 10: 296 I, 3, pp. 14-5: 156 I, 4: 289-90 I, 5: 29, 30, 52n, 53n, 366-7, 369, 377, 381 I, 5, p. 22: 381n I, 5, pp. 23-4: 367n I, 5, p. 24: 70, 368n I, 5, pp. 24-5: 73 I, 5, pp. 25-9: 71 I, 6: 9 I, 6, p. 30: 156 I, 7: 9, 11 I, 8: 8n, 70n I, 8, pp. 38-9: 70 II, 1: 8, 256, 258n II, 1, p. 45: 256n II, 1, p. 46: 10n II, 1, pp. 46-8: 257n II, 1, p. 48: 256n II, 4: 9, 257, 259, 262 II, 4, p. 65: 259n II, 4, p. 67: 259n II, 4, p. 68: 260n III: 302 III, 1: 8 III, 3: 9, 53 III, 3-5: 302 III, 3, p. 80: 83, 92 III, 3, p. 81: 79, 80 III, 3, pp. 82-3: 79, 80 III, 3, pp. 82-4: 87 III, 3, p. 84: 80 III, 5-6: 53 III, 5, p. 91: 80 III, 5, pp. 91-3: 83n ˘

al-Madhal (tr. Marmura) ˘ pp. 50f: 345n ˘

Sˇifa¯ : al-Mantiq: al-Maqu¯la¯t (Categories) I, 4: 8, 10˙ ˘

Sˇifa¯ : al-Mantiq: al-Safsata (Sophistici ˙ Elenchi) ˙ I, 1, p. 7: 46n II, 2, pp. 67-9: 46n ˘

˘

Sˇifa¯ : al-Tabı¯ ¯ıya¯t (Natural Philosophy): alSama¯ ˙ al-Tabı¯ ¯ı (Physics) ˙ 263n, 264 I, 10: 262, I, 10, pp. 48-9: 262n I, 10, p. 49: 259n, 264n I, 13, pp. 63-6: 9n I, 14: 9 ˘

˘

˘

Sˇifa¯ : al-Tabı¯ ¯ıya¯t: al-Kawn wa-l-Fasa¯d (On ˙ Generation and Corruption) (ed. Qassem) VI: 266n XIII, p. 187: 225n XIV, p. 190: 225n, 232n, 259n, 266n ˘

al-Kawn wa-l-Fasa¯d (Latin, ed. Van Riet) XIV, p. 139: 232n ˘

˘

˘

Sˇifa¯ : al-Tabı¯ ¯ıya¯t : Fı¯ l-af a¯l wa-l-infi a¯la¯t ˙ (On Actions and Passions) p. 256: 225n ˘

˘

Sˇifa¯ : al-Tabı¯ ¯ıya¯t: al-Nafs (On the Soul) ˙ (ed. Rahman) I, 1, p. 9: 46n I, 1, p. 16: 106n I, 5, p. 47: 190n II, 2, p. 62: 21n IV, 4: 21n V: 193 V, 2, pp. 209ff: 103n V, 5: 181 V, 6, p. 239: 106n ˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50

Index of Avicenna’s Works with Passages Cited

Index of Avicenna’s Works with Passages Cited 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50

III, 7-10: 8 IV: 298 IV, 2, pp. 131-2: 156 IV, 2, p. 134: 257n IV, 2, p. 142: 264n IV, 3: 8, 28 V: 289-90, 302, 369 V, 1: 52n V, 1, p. 149: 53, 74, 369n V, 4: 9 V, 5: 8 V, 6: 8, 9 V, 8-9: 8, 9 V, 9: 264n V, 9, p. 193: 265n VI: 289 VI, 1: 261 VI, 1, p. 198: 260n VI, 1-2: 270 VI, 1-4: 8 VI, 2: 259-60, 262, 268, 272 VI, 2, p. 201: 261n VI, 2, p. 202: 261n, 262n VI, 2, p. 203: 232n VI, 2, p. 204: 232n, 267n VI, 3: 263, 270 VI, 3, p. 205: 263n VI, 3, p. 206: 264n VI, 3, p. 207-8: 263n VI, 3, p. 213: 264n VI, 4, p. 219: 261n VI, 5, p. 220: 11n VII: 289-92, 302-303 VII, 1: 52n, 290, 292 VII, 2-3: 8, 292, 302 VII, 2, p. 243: 290 VII, 2, pp. 243-4: 157, 291 VII, 2, p. 247: 293, 297 VII, 2, pp. 248-9: 294 VIII-IX: 225 VIII, 1: 11, 12 VIII, 3-4: 11 VIII, 3, p. 272: 232n VIII, 4, pp. 276-7: 298 VIII, 5-7: 11 VIII, 6-IX, 3: 161n VIII, 6, p. 284: 185n VIII, 6, pp. 284-5: 114n VIII, 6, p. 288: 157 VIII, 7, pp. 291-2: 119n

391

VIII, 7, p. 295: 114n VIII, 7, p. 296: 156 IX, 1: 11 IX, 2, pp. 311-12: 156 IX, 2, p. 314: 156 IX, 4, p. 331: 225n IX, 5: 259, 265 IX, 5, pp. 334-6: 265n IX, 5, p. 335: 225n IX, 5, p. 337: 225n IX, 6: 14, 161 IX, 6, p. 339: 161n, 162n IX, 6, p. 340: 162n, 163n, 165n, 166n IX, 6, pp. 340-43: 163n IX, 6, p. 342: 169n IX, 6, p. 343: 166n IX, 6, pp. 343-4: 162n IX, 6, p. 345: 171n IX, 6, p. 346: 169n IX, 6, p. 347: 163n, 172n IX, 7: 384n IX, 7, pp. 350-51: 157 X, 1: 182 X, 1-2: 13, 19 X, 1, p. 359: 182n X, 1, pp. 362-3: 157 X, 1, p. 363: 182n X, 2, pp. 364-5: 157 X, 4-5: 16 al-Ila¯hiyya¯t (ed. Anawati) I, 5, pp. 29-30: 9 I, 5, pp. 31-3: 31n I, 5, p. 33: 46n I, 6, pp. 37-8: 31n I, 6, pp. 38-9: 43n II, 1, p. 57: 256n II, 1, p. 58: 10n II, 1, p. 59: 257n II, 1, p. 60: 256n, 257n II, 4, pp. 83-5: 259n II, 4, p. 87: 260n IV, 2, p. 175: 257n IV, 2, p. 184: 264n IV, 3, p. 188: 29n V, 2, p. 211: 23 V, 8, p. 247: 24 V, 9, p. 252: 265n VI, 1, p. 261: 260n VI, 2, pp. 264-5: 261n

392

VI, 2, p. 266: 262n VI, 2, p. 267: 267n VI, 4, p. 283: 261n VI, 3, p. 268: 263n VI, 3, p. 269: 264n VI, 3, pp. 270-71: 263n VI, 3, pp. 274-8: 10 VI, 3, p. 276: 264n VI, 4, p. 283: 9n VI, 5, p. 284: 11 VI, 5, pp. 288-300: 10n VII, 3, p. 318: 24 VIII, 4, pp. 343-7: 31n VIII, 7, pp. 368-9: 12 IX, 2, pp. 381-4: 12 IX, 2, p. 383: 13 IX, 2, p. 386: 12 IX, 2, pp. 389-92: 14 IX, 2, pp. 392-3: 12 IX, 3, pp. 402-9: 13 IX, 5, pp. 413-14: 12 IX, 5, p. 410: 13 IX, 5, pp. 410-11: 265n IX, 5, pp. 411-2: 13 IX, 5, p. 413: 13 IX, 6, pp. 415-18: 12 IX, 7: 12 X, 1, p. 435: 14 X, 3: 12, 18 al-Ila¯hiyya¯t (Latin, ed. Van Riet) I, 1: 209, 212 I, 1-3: 206 I, 2, p. 10: 212n, 213n I, 2, p. 13: 296n I, 2, pp. 14-17: 211n I, 5: 212, 218 I, 5, p. 29: 212n I, 5, p. 31: 381 I, 5, pp. 33-4: 367n I, 5, pp. 33-40: 312n I, 5, pp. 34-5: 368n I, 5, pp. 36-40: 313n I, 6: 218 I, 6, p. 44: 208n I, 6-7: 208 II, 1: 212 II, 1, p. 65: 256n II, 1, p. 67-9: 257n II, 1, pp. 68-9: 256n

II, 3: 209n II, 4: 209 II, 4, pp. 96-8: 259n II, 4, p. 100: 260n III, 1: 209 III, 1, p. 106: 312n III, 2, p. 114: 312n III, 3: 355 III, 3, p. 117: 312n, 343n III, 3, pp. 119-22: 312n IV, 2, p. 200: 257n IV, 2, p. 212: 264n V: 351n, 352n, 354-6, 358 V, 1: 311n, 312n V, 1-4: 344n V, 1, p. 228: 312n V, 1, p. 228-9: 369n V, 1, pp. 230-32: 312n V, 1, p. 232: 311n V, 1, pp. 233-6: 313n V, 1, pp. 233-7: 315n V, 1, pp. 234-6: 311n, 312n V, 2: 357-8 V, 2, pp. 236-7: 297n V, 2, p. 240: 343n V, 2, p. 243: 23n V, 3: 357 V, 5, p. 276: 345n V, 5, p. 277: 24 V, 9, pp. 289-90: 265n VI, 1, pp. 296-7: 260n VI, 2, p. 301-302: 261n VI, 2, p. 303: 262n VI, 2, p. 305: 267n VI, 3, p. 307: 263n VI, 3, p. 308: 264n VI, 3, p. 310: 263n VI, 3, pp. 317-8: 264n VII, 1, p. 349: 312n VII, 2, pp. 363-4: 293n VIII, 4, p. 402: 298n VIII, 7: 209 IX, 2: 209n, 212 IX, 2, p. 387: 212n IX, 2–5: 209 IX, 5: 209n IX, 5, p. 488: 265n IX, 5, pp. 489-90: 265n ˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50

Index of Avicenna’s Works with Passages Cited

Ta lı¯qa¯t (Notes)

Index of Avicenna’s Works with Passages Cited 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50

pp. 13-22: 17n p. 23: 18 p. 37: 18 pp. 47-8: 19 p. 49: 18 p. 52-4: 18 p. 62-3: 18

pp. 69-70: 19 p. 77: 19 p. 79: 19 p. 81-2: 19 p. 164: 28n pp. 193-194: 19

393

Index of Names

˘

˘

˘

˘

˘

˘

209-218, 228, 231n, 233, 235-6, 2389, 241-2, 243n, 244, 251, 253n, 255n, 256n, 261n, 264n, 268, 271n, 276-81, 286, 289-92, 295, 300-303, 319, 3267, 348-9, 358, 371n, 382-5 Asclepius 280 al-Asˇ arı¯ 30-31, 33n, 34n, 36, 40, 42-3, 44n, 45, 127, 131-2 Augustine 202, 206n, 208, 227-8 Avendauth (see also Abraham Ibn Daud) 160, 215n Averroes (Ibn Rusˇd) 3-4, 40, 42n, 47-8, 51-6, 57n, 58-62, 63n, 64-9, 72-88, 90-94, 97, 153n, 191, 193, 200-203, 205-206, 211n, 216, 229-31, 233, 235, 238, 241, 245, 252n, 255n, 256n, 258n, 268, 312n, 344, 355n Avner of Burgos 155, 179-85, 187-8, 190-93 Badawı¯ 14, 15n Bahmanya¯r 7-12, 15, 18n, 19n, 23, 42n al-Ba¯qilla¯nı¯ 35-6, 39, 40, 44 al-Bayda¯wı¯ 47, 123n ˙ ¯ 7, 38n al-Bayhaqı al-Bazdawı¯ 33 C. Belo 172n A. Bertolacci 7n, 11n, 24, 302, 330n al-Bı¯ru¯nı¯ 181, 184, 186, 189-90, 193 Boethius 199n, 202, 205, 206n, 207n, 208, 277, 281-2, 284, 286, 288, 299300, 339-40, 343, 347, 352-4, 356-60 Bonaventure 228, 231, 342n, 347, 359, 370n M. Bouyges 57n D. J. Brand 262n, 263n al-Buha¯rı¯ 186n ˘ Campanella, Tommaso 230 J. Carrier 288 Daniel of Morley 202n, 215n David (Greek commentator) 281n David of Dinant 215, 217n H. Denifle 216n R. Descartes 198 R. De Vaux 209n, 215n ˘

˘ abba¯r 34n, 40, 44 Abd al-G al-Abharı¯ 47, 123n, 125-34, 136, 142, 146 Abraham bar Hiyya 164, 166 ˙ Abraham Ibn Daud 4, 153, 155, 15961, 162n, 164-73 Abraham Ibn Ezra 164, 192n Abu¯ al-‘Abba¯s Ahmad ibn ‘Alı¯ al-Isfa˙ ˙ ha¯nı¯ 13 ˇ ubba¯ ¯ı 34n Abu¯ Alı¯ al-G Abu¯ l-Baraka¯t al-Bag˙da¯dı¯ 38 ˇ ubba¯ ¯ı 30, 35-6, 39-40, Abu¯ Ha¯ˇsim al-G 44 Abu¯ l-Hudayl 33, 36 ¯ Abu¯ l-Husayn al-Basrı¯ 42-3, 47, 127 ˙ Abu¯ Ma ˇsar 205n ˙ Adam Wodeham 366 Ala¯ al-Dawla Fara¯marz 38n Albertus Magnus 198-99, 201-202, 227-9, 231, 232n, 238-45, 252n, 342n, 343n Alexander of Aphrodisias 64, 66, 216n, 239, 241, 280 Alexander of Hales 254n Alfonso de Valladolid see Avner of Burgos Alfred of Sareshel 216n M.-T. d’Alverny 200n, 208n, 209-210 Amalricus of Bene 215 al-A¯midı¯ 40 Ammonius 280, 295n G. C. Anawati 10n, 11n Anaxagoras 238n, 253 Anonymous (d’Alverny) 202, 204, 209, 210n Anonymous (de Vaux), author of De causis primis et secundis 202n, 208, 209n, 226, 239n Anonymous (van Steenberghen) 226n Aristotle 1, 12, 42n, 46n, 51-3, 56, 589, 60n, 62, 64-6, 70, 72, 74-7, 79-81, 82n, 83-6, 87n, 88-91, 94n, 98-100, 102, 103n, 107-8, 119-21, 145, 169n, 178, 197, 200-204, 205n, 206, 207n,

˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50

396

Dexippus 60n al-Dı¯ba¯gˇ¯ı 9n, 12n, 13n, 14n, 16 Dominicus Gundisalvi (Gundissalinus) 2, 160, 197, 201-202, 204, 205n, 206-208, 210-11, 218, 226, 233 M. Dunlop 20n Eckhart of Hochheim (Meister Eckhart) 55n Elias (Greek commentator) 281n al-Fa¯ra¯bı¯ 16n, 18n, 20-23, 28, 51-3, 57, 59-62, 64-76, 79, 85-6, 88-92, 94, 108, 109n, 160, 206n, 207n, 209, 231, 268, 288-9, 295, 301, 358 Ficino, Marsilio 227, 233, 236 A. Fidora 206n, 207n Francis of Marchia 368n G. Galluzzo 309n Gaudapa¯da 184n ˙ ¯lı¯ 3-4, 42n, 51, 54, 66-7, 72, ˙ aza al-G 153, 155, 161, 168-9, 200n, 202, 206n, 207, 210-11, 212n, 213n, 217, 226, 236, 347, 360 Gilbert of Poitiers 199n . Gilson 199n, 365 Gerald of Wales 215n Gerard of Cremona 202n Giles of Rome 347, 360 ˇ ¯ılı¯ 29-30, 40 al-G Godfrey of Fontaines 341n, 342n, 343, 345n, 347, 353, 356 A.-M. Goichon 115n V. Govind 184n M. Grabmann 214n, 215n Gundissalinus see Dominicus Gundisalvi D. Gutas 178-80, 188-9, 192n ˘ uwaynı¯ 30, 33-40, 44-5 al-G Hasta¯mala 191n M. Heidegger 277 Henry of Ghent 198, 227, 343, 347-54, 365-6, 368n, 371-5, 377-80 Hermann of Carinthia 205n, 208n Hermann the German 216n M. Horten 10n, 11n G.F. Hourani 172n A. Hyman 258n Ibn al-Mala¯himı¯ 44, 47 Ibn al- Arabı¯˙ 48, 133 Ibn Falaquera 154, 156-7 Ibn Gabirol 202, 206n, 208, 210-11 ˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50

Index of Names

Ibn Hazm 39 ˙ Ibn Kammu ¯ na 45, 47 Ibn Kulla¯b 33-4, 36 Ibn Rusˇd s. Averroes Ibn Tufayl 177, 181n, 183n, 185n, ˙ 187n, 188n, 189n, 190n, 192-3 186n, Ibn Waqqar 155 al-I¯gˇ¯ı 123n ¯Ila¯qı¯ 39n Isaac Polgar 179 ¯I´svarakr¸¸sn¸a 184n A. Ivry 172n T. Izutsu 27n W. Jaeger 280 James of Venice 204 J. Janssens 172n John Blund 201-203, 212-3, 216-8 John Buridan 227, 233, 235-6, 238 John Damascene 347-50, 352, 354-8 John Duns Scotus 3, 198, 341n, 342n, 343, 347, 353, 357, 365-6, 375-85 John of Jandun 235 John of La Rochelle 227, 254n, 347, 358 John of Paris 343, 358 John Pecham 227 John Scot Eriugena 202, 208 J. Jolivet 202n, 206n Kapila 184n al-Ka¯tibı¯ al-Qazwı¯nı¯ see al-Qazwı¯nı¯ alKa¯tibı¯ al-Kindı¯ 52, 68-70, 85, 88, 90-92, 103n, 206n C. Lafleur 288 al-Lawkarı¯ 2, 7-16, 17n, 18n, 19-20, 22-4 P. Lee 262n, 263n G.W. Leibniz 198 A. de Libera 200n, 245n O. Lizzini 10n, 260n M. Mahdi 52, 90 Maimonides 4, 153-5, 159, 173 Ma¯lik-Sˇa¯h 39n R.P. Mandonnet 215n J.-M. Mandosio 216n R. Marcotte 14n M. Marmura 10, 11n, 268n C. Di Martino 216n Mauricius Hispanus 215

Index of Names

Meister Eckhart see Eckhart of Hochheim W. Mettmann 179 Michael of Ephesus 280 Michael Scot 201-202, 204-206, 21011, 216n Moses ben Judah 153 Moses ha-Levi 154 Mulla¯ Sadra¯ 11n, 28 ˙ ¯ 55n al-Na¯bulusı F.M. Nagˇgˇa¯r 19-20 P. Natorp 280 Nifo, Agostino 229, 230, 231n Olympiodorus 281n J.K. Otte 216n Parmenides 83, 84n PataÇjali 184n, 186 Peter Abelard 205, 342 Peter John Olivi 341n, 342n, 347, 360 Peter of Auvergne 342n, 343, 345n, 357 Peter of Corbeil 213 Peter of Falco 343, 354 Peter of Tarantaise 347n, 350n, 359 Philoponus 281n Pietro d’Abano 229 G. Pini 314-5 Plato 84-6, 103, 189, 227, 228n, 230n, 231-2, 235-6, 242n, 245, 254, 268, 271, 278, 291, 302 Plotinus 116, 173 Pomponazzi, Pietro 229, 230n Porphyry 178, 339, 340n, 343, 347, 350, 352-60 Proclus 91n, 173 Pseudo-Alexander of Aphrodisias 28081 Pseudo-Dionysius 208, 242-3 Pseudo-Ptolemy 13 Pythagoras 291, 299 al-Qazwı¯nı¯ al-Ka¯tibı¯ 47, 123n, 125, 128-30, 134, 147 al-Qusˇayrı¯ 34n Qutbaddı¯n al-Sˇ¯ıra¯zı¯ 45, 47 ˙¯zı¯, Fahr al-Dı¯n 2, 29-30, 32, 40al-Ra 47, 97-8,˘ 101, 103-106, 109, 110n, 111-21, 123-31, 134, 141-2, 155n Richard of Middleton 343, 355 Richard of St-Victor 359 Robert Grosseteste 201-203, 348n Robert Kilwardby 347, 360

Robert of CourÅon 214 Roger Bacon 201-203, 227, 347, 359 Roger Marston 227, 343, 347, 356 Roland of Cremona 202-203 M.-D. Roland-Gosselin 199n A. Rucquoi 206n Russiliano, Tiberio 227, 230, 233, 237 A.I. Sabra 70n al-Sˇahrasta¯nı¯ 39-40, 44, 47 al-Sˇahrazu¯rı¯ 40, 45 al-Samarqandı¯ 44n, 47, 123n, 126-7 S´amkara 190n ˙ Scaliger, Julius Caesar 230 A. Schwegler 278-9 Sennert, Daniel 230 Y. Shiffman 154 A. Shihadeh 41 Simplicius 281n, 347, 360 Socrates 291 Speusippus 291 Spinoza 198 A. Stone 258n, 266-7 Suarez, Francisco 228, 229n al-Suhrawardı¯ 27-30, 37, 40-41, 44-8, 125-31, 133, 141 Syrianus 280-81 R. Szpiech 177, 192n al-Tafta¯za¯nı¯ 123n, 133 Themistius 231n Theophrastus 295n Thomas Aquinas 3, 69, 198-9, 205n, 226n, 227n, 228-9, 232-3, 236, 238, 251-7, 258n, 262, 264, 267n, 268-72, 275, 277, 282-90, 298-301, 309-29, 330n, 331-4, 339, 342n, 349n, 350n, 365-6, 368n, 369-71, 374, 384 Thomas Sutton 350n Tignosi, Niccol 229 Todros ben Meshullam ben David Todrosi 154 al-Tu¯sı¯ 97, 98n, 104-9, 110n, 111n, ˙ 112-4, 115n, 116-21, 129 Umar Hayya¯m 30, 37-9, 41, 44-7 ˘ ¯ 123n al-Urmawı Usta¯t 59n ¯ Riet 10n, 11n, 23n S. ˙Van M.H. Vicaire 202n Vincent of Beauvais 210, 227 Vital du Four 227 S. Weil 170n

˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50

397

398

William Brito 214n William de la Mare 347, 349n, 360 William of Auvergne 198-9, 201-203, 218, 227, 233-5, 254n, 347, 358 William of Ockham 366 R. Wisnovsky 52n, 53n

Xenocrates 291 Yahya ibn Adı¯ 46n, 111n ˙ Zedler 258n B.H. Zimara, Marcantonio 229, 230n F.W. Zimmermann 60n, 263n ˘

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50

Index of Names

This index of names lists all authors of primary studies. Authors of secondary studies are included only if their views are discussed in some detail.