Penser Avec Avicenne: De L'héritage Grec À La Réception Latine, En Hommage À Jules Janssens (Recherches De Theologie Et Philosophie Medievales - Bibliotheca, 20) (French Edition) 9789042949195, 9789042949201, 9042949198

Citation preview

Recherches de Théologie et Philosophie médievales Bibliotheca 20 Penser avec Avicenne De l’héritage grec à la réception latine, en hommage à Jules Janssens

Mélanges édités par

Daniel De Smet et Meryem Sebti

PEETERS

PENSER AVEC AVICENNE

Recherches de Théologie et Philosophie médiévales

BIBLIOTHECA

Editorial Board Russell L. FRIEDMAN, Wouter GORIS, Guy GULDENTOPS, Maarten HOENEN, Fiorella RETUCCI, Andreas SPEER, Carlos STEEL, David WIRMER

Bibliotheca is a peer-reviewed series published by the editorial board of Recherches de Théologie et Philosophie médiévales. It contains studies on medieval thought and editions of medieval philosophical and theological texts.

Bibliotheca 20

Penser avec Avicenne De l’héritage grec à la réception latine, en hommage à Jules Janssens Mélanges édités par

Daniel De Smet et Meryem Sebti

PEETERS LEUVEN – PARIS – BRISTOL, CT 2022

A catalogue record for this book is available from the Library of Congress All rights reserved. No part of this book may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise, without the prior permission of the publisher. © 2022 – Peeters – Bondgenotenlaan 153, 3000 Leuven, Belgium. D/2022/0602/57 ISBN 978-90-429-4919-5 eISBN 978-90-429-4920-1

TABLE DES MATIÈRES Introduction.............................................................................

3

Acte, puissance et réveil dans le VIIIe mīmar de la Théologie d’Aristote ............................................................................ Michael Chase

21

Le commentaire d’Avicenne à la Théologie d’Aristote : étude préliminaire à l’histoire d’une transmission complexe ............ Meryem Sebti

45

L’imam, philosophe et prophète de la cité vertueuse (al-madīna al-fāḍila) d’al-Fārābī, est-il un imam shi‘ite ? .................... Daniel De Smet

63

Tashkīk al-Wujūd and the Lawāzim in Avicenna’s Metaphysics Damien Janos

91

Freedom and Responsibility in Avicenna ................................. 149 Jari Kaukua Ittiṣāl, Contact, of the Human and Supernal Intellects in Avicenna: The Goal of Ethics, the Foundation of Epistemology ...... 169 Dimitri Gutas Avicenna on the Household? ................................................... 179 Thérèse-Anne Druart On the Authenticity of the Throne Epistle (al-Risāla al-‘arshiyya) Ascribed to Avicenna ........................................................ 193 Frank Griffel Approaching a Crisis: al-Ghazālī in 488/1095 ......................... 231 Taneli Kukkonen Avicenna’s Shifā᾿ and a Commentary on It in Twenty-Five Volumes: The Bodleian Pococke Collection and Fakhr al-Dīn al-Rāzī.................................................................... 255 Amos Bertolacci

2

TABLE DES MATIÈRES

Philosopher au XIIe siècle. Notes autour d’Avicenne et de sa première réception latine au Moyen Âge .......................... 283 Olga L. Lizzini Henry Bate on an Apparent Contradiction in Avicenna’s Doctrine of the Soul ........................................................................ 311 Carlos Steel Natura intendit speciem: A Brief Reception History of an Avicennian Topos from Peter of Spain to Francesco Piccolomini 335 Guy Guldentops A Note on the Solid Circle: Gentile, Rosmini, Contarini and the History of Philosophy ....................................................... 387 Andrea Aldo Robiglio

INTRODUCTION Depuis plus de vingt ans, il est bien difficile de lire un article ou un livre sur Avicenne sans voir le nom de Jules Janssens mentionné dans la bibliographie. C’est dire l’importance de ses travaux, notamment ceux consacrés à Avicenne et à sa réception tant dans le monde musulman que dans le monde latin. Parmi la centaine d’articles publiés, dont une belle majorité portant sur Avicenne, Jules Janssens a construit avec patience et minutie une œuvre importante qui souligne tout en les éclairant les nombreux points de tension de la pensée du grand philosophe persan. Ainsi, il est l’un des rares – avec Fazlur Rahman (1914-1974) – à avoir abordé la question de savoir quel statut le philosophe accorde à la révélation coranique. Il montre que tout en s’inscrivant dans la filiation de l’héritage aristotélicien, Avicenne cherche à intégrer dans son système philosophique – en vue d’en rendre raison rationnellement – les données coraniques : le statut de la prophétie, la réalité ontologique de l’unité et de l’unicité absolues de Dieu (tawḥīd), l’angélologie… (« Ibn Sīnā’s Ideas on Ultimate Reality and Meaning. Neoplatonism and the Qur’ān as Problem-Solving Paradigms in the Avicennian System » [1987] et « Ibn Sīnā (Avicenne) : un projet religieux de philosophie ? » [1998]). Dans son « Avicenna and the Qur’ān. A Survey of His Qur’ānic Commentaries » (2004), Jules Janssens poursuit son analyse en étudiant l’herméneutique coranique d’Avicenne. Il affinera ce travail sur la lecture philosophique du Coran en étudiant celle d’al-Kindī et en posant de manière plus générale la question importante du statut des commentaires philosophiques du livre fondateur de l’islam (« Al-Kindī : The Founder of Philosophical Exegesis of the Qur’ān » [2007] et « Philosophical Commentaries » [2020]). Ces études fondées sur des analyses textuelles minutieuses lui permettent d’aborder de façon plus large le thème du rapport entre foi et raison en enrichissant ce débat de deux contributions significatives (« La philosophie peut-elle contribuer au dialogue interreligieux ? Réflexions à partir de Ghazālī, Maïmonide, Thomas d’Aquin et Yaḥyā’ ibn ‘Adī » [2013] et « Foi et raison chez Avicenne » [2014]).

4

INTRODUCTION

Aucune des zones d’ombre du corpus avicennien n’a échappé à l’acribie de Jules Janssens. Il n’a eu cesse d’aborder la question épineuse des problèmes textuels dus aux transmissions complexes des textes d’Avicenne (« Le Dānesh-Nāmeh d’Ibn Sīnā : un texte à revoir ? » [1986] ; « Les Ta‘līqāt d’Ibn Sīnā, essai de structuration et de datation » [1997] ; « Ibn Sīnā’s Ta‘līqāt : The Presence of Paraphrases of and Super-Commentaries on the Ilāhiyyāt of the Shifā’ » [2012] ; « The Significance of Avicenna’s Additional Remarks to the DānešNāmeh in his Ta‘līqāt, with Special Attention to the Sections on Divine Knowledge and Will » [2020]). De même, il n’a eu cesse de faire face aux questions les plus difficiles qui confrontent tout lecteur attentif d’Avicenne, en particulier celle de la liberté humaine, qui se pose de manière aigüe. En effet, tout en affirmant que la destinée de l’homme est d’atteindre à la perfection de son âme rationnelle – laissant ainsi entendre que la liberté de se déterminer est au fondement de l’existence humaine – Avicenne construit un système métaphysique déterministe. Jules Janssens analyse avec finesse ces tensions (« The Problem of Human Freedom in Ibn Sīnā » [1996]), tout comme il aborde avec rigueur la place de l’expérience dans la constitution de la connaissance, autre question complexe étant donné la doctrine extrinséciste de l’origine de la connaissance humaine chez Avicenne articulée cependant à une valorisation de l’apport de l’expérience sensible dans l’acquisition de la connaissance (« “Experience” (tajriba) in Classical Arabic Philosophy (al-Fārābī - Ibn Sīnā) » [2004] ; « Ibn Sīnā on “Experience” (tajriba) and Its Particular Significance for the Evaluation of Simple Drugs » [2017]). On doit aussi à Jules Janssens des articles éclairants tant sur la logique (« Ibn Sīnā on Substance in Chapter Two of the Maqūlāt (Categories) of the Shifā’ » [2013]) que sur de subtils points de doctrine, tels la distinction entre création et émanation ou l’usage des notions de Wāhib al-ṣuwar et de Wāhib al-‘aql par Avicenne (« Creation and Emanation in Ibn Sīnā » [1997] ; « The Notions of Wāhib al-ṣuwar (Giver of Forms) and Wāhib al-‘aql (Bestower of Intelligence) in Ibn Sīnā » [2006]). Jules Janssens ne cède jamais, dans l’examen qu’il fait de ces tensions doctrinales, à la tentation d’une réponse univoque qui esquiverait les difficultés. Il analyse les textes avec rigueur, fait un état des lieux des problèmes et laisse quand il le faut les problèmes ouverts à la discussion. C’est aussi en cela que ses études sur

INTRODUCTION

5

la philosophie d’Avicenne sont si précieuses et qu’elles ont nourri et inspiré le travail de tant de chercheurs. Les travaux de Jules Janssens ont également ouvert la voie à la recherche sur le post-avicennisme dans le monde musulman. Il est l’un des premiers à s’être intéressé aux disciples d’Avicenne. Le plus connu d’entre eux, l’auteur du Kitāb al-Taḥṣīl Bahmanyār ibn Marzubān (m. 1066), est un disciple de la première génération d’Avicenne. Il a infléchi la métaphysique de son prestigieux maître tout comme il a remis en question certaines doctrines de sa psychologie (« Bahmanyār ibn Marzubān : A Faithful Disciple of Ibn Sīnā ? » [2003] ; « Bahmanyār and his Revision of Ibn Sīnā’s Metaphysical Project » [2007] ; « Bahmanyār ibn Marzubān » [2020]). Le second, disciple de la deuxième génération, ayant eu pour maître Bahmanyār, Abū ‘Abbās al-Lawkarī (m. 1123/24), auteur du Bayān al-ḥaqq bi-ḍimān al-ṣidq, a renouvelé la question de la relation de prédication telle que la conçoit Avicenne (« Al-Lawkarī’s Reception of Ibn Sīnā’s Ilāhiyyāt » [2012]). Jules Janssens a été précurseur dans l’étude du lien entre la philosophie d’Avicenne et celle du théologien ash‘arite al-Ghazālī (10581111). Les spécialistes d’al-Ghazālī s’accordent aujourd’hui pour considérer que ses ouvrages marquent un tournant dans l’histoire de la pensée musulmane, notamment en raison de la synthèse qu’il a effectuée de la philosophie d’Avicenne, laquelle devient une partie intégrante de la théologie islamique à partir du XIIe siècle. Or Jules Janssens est là encore un des premiers à avoir montré l’influence d’Avicenne sur al-Ghazālī au travers de nombreux articles (« Al-Ghazzālī’s Tahāfut : Is It Really a Rejection of Ibn Sīnā’s Philosophy ? » [2001] ; « Al-Ghazzālī’s Mi’yār al-‘ilm fī fann al-mantiq : sources avicenniennes et farabiennes » [2002] ; « Al-Ġazālī and His Use of Avicennian Texts » [2003] ; « Al-Ghazzālī’s Political Thought : Elements of Greek Philosophical Influence » [2004] ; « Al-Ghazālī’s Mizān al-‘amal : An Ethical Summa Based on Ibn Sīnā and Rāghib al-Isfahānī » [2008] ; « L’âme-miroir : al-Ġazālī entre philosophie et mysticisme » [2008] ; « Al-Ghazzālī : The Introduction of Peripatetic Syllogistic Theory in Islamic Law (and Kalām) » [2010] ; « Al-Ghazālī between Philosophy (Falsafa) and Sufism (Taṣawwuf) : His Complex Attitude in the Marvels of the Heart (‘Ajā’ib al-qalb) of the Ihyā’ ‘Ulūm al-Dīn » [2011] ; « Éléments avicenniens dans le livre al-Maqṣad

6

INTRODUCTION

d’al-Ghazālī » [2014] ; « Al-Ghazālī’s Use of Avicennian Texts in His Maqāṣid al-falāsifa » [2019]). Par-delà la question de l’influence d’Avicenne, Jules Janssens a renouvelé notre lecture d’al-Ghazālī, peut-être le théologien musulman le plus influent de l’histoire intellectuelle en terre d’islam (« Al-Ġazālī : How to Read the Munqiḏ ? » [2013]). On lui doit également des études précieuses consacrées à l’influence d’Avicenne sur l’autre grand théologien ash‘arite de la fin du XIIe - début du XIIIe siècle, à savoir Fakhr al-Dīn al-Rāzī (1150-1210). Là encore, les travaux de Jules Janssens ont réorienté une lecture désormais obsolète qui a longtemps prévalu. La philosophie aurait déserté le paysage intellectuel de l’islam après la mort en Occident musulman d’Averroès (1126-1198). Or comme le montrent les travaux de Jules Janssens, la philosophie a affecté en profondeur la structure même de la pensée théologique, s’inscrivant ainsi durablement dans ce paysage intellectuel (« Ibn Sīnā’s Impact on Fakhr ad-Dīn ar-Rāzī’s Mabāḥith al-Mashriqiyya, with Particular Regard to the Section Entitled al-Ilāhiyyāt al-maḥḍa : An Essay of Critical Evaluation » [2010] ; « Fakhr ad-Dīn ar-Rāzī on the Soul : A Critical Approach to Ibn Sīnā » [2012] ; « Fakhr al-Dīn al-Rāzī’s Use of al-Ghazālī’s Mishkāt in His Commentary on the Light Verse (Q. 24:35) » [2016] ; « Avicennian Elements in Faḫr al-Dīn al-Rāzī’s Discussion of Place, Void and Directions in the al-Mabāḥiṯ al-mašriqiyya » [2018]). Il nous serait difficile de mentionner ici l’intégralité des contributions de Jules Janssens à notre discipline. Nous ne pouvons cependant nous empêcher – afin d’illustrer l’étendue de sa curiosité intellectuelle comme celle de son savoir – de citer son article concernant la reprise par le grand philosophe iranien Mullā Ṣadrā (1571/72- 1640/41) des Ta‘līqāt d’Avicenne (« L’utilisation des Ta‘līqāt d’Ibn Sīnā dans les Asfār de Mullā Sadrā Shīrāzī [2002]) ; ou encore celui traitant de l’influence d’Avicenne sur l’un des premiers philosophes ottomans al-Fanārī (m. 1431) (« Elements of Avicennian Influence in al-Fanârî’s Theory of Emanation » [2010]). Outre ses qualités d’arabisant, Jules Janssens est une autorité internationalement reconnue et estimée dans le cénacle des historiens de la philosophie médiévale de langue latine, pour ses éditions critiques et études innovantes des traductions arabo-latines d’Avicenne : l’Avicenna latinus. Lui revient le mérite d’avoir mené à bon terme, dans

INTRODUCTION

7

des circonstances parfois difficiles, un projet lancé en 1968 par Simone Van Riet et Gerard Verbeke, visant à établir des éditions critiques des traductions latines médiévales du Kitāb al-Shifā’. Établie à Louvain-la-Neuve après la scission de l’Université Catholique de Louvain, Simone Van Riet publia, avec le soutien de l’Académie Royale de Belgique et sous le patronage de l’Union Académique Internationale, et avec des introductions doctrinales de Gerard Verbeke, la Métaphysique (Liber de philosophia prima sive scientia divina), la première partie du livre I, ainsi que les livres III, IV et VI de la Physique (Libri Naturalium) : De causis et principiis naturalium, De generatione et corruptione, De actionibus et passionibus et De anima. Elle engagea des travaux préparatoires en vue de l’édition des parties 2 (De motu et de consimilibus) et 3 (De his quae habent naturalia ex hoc quod habent quantitatem) du premier livre de la Physique, mais sa mort en 1993 laissa son projet inachevé. L’Académie Royale et l’Union Académique Internationale chargèrent alors Jules Janssens de mener à bien l’édition des deux volumes, dont le premier parut en 2006 (Avicenna Latinus, Liber primus Naturalium. Tractatus secundus. De motu et de consimilibus) et le second en 2017 (Avicenna Latinus, Liber primus Naturalium. Tractatus tertius. De his quae habent naturalia ex hoc quod habent quantitatem). En acceptant cette proposition de poursuivre le projet de l’Avicenne latin, dont il est actuellement le directeur scientifique, Janssens prit sur lui une lourde responsabilité. Car la tâche était d’envergure : pour le premier volume, il y avait pour ainsi dire trop de manuscrits, avec des divergences parfois considérables, et pour la majeure partie du second, il n’y en avait pas assez : un codex unicus de la Vaticane. Janssens s’aperçut en cours de route que le travail préliminaire de Van Riet était à refaire et que le texte, outre les aléas de sa transmission manuscrite, posait de redoutables problèmes. Les difficultés s’accumulaient à différents niveaux : les traducteurs ont lutté avec la syntaxe et le vocabulaire technique d’Avicenne, de sorte que leur version latine s’avère souvent inintelligible si l’on n’a pas recours à l’original arabe. Or les différentes éditions du Kitāb al-Shifā’ sont peu fiables et le texte arabe n’est pas établi de façon satisfaisante. Janssens devait donc se confronter simultanément à deux difficultés, en essayant de reconstituer l’arabe à l’aide du latin et vice-versa, un exercice certes périlleux mais qu’il réussit à parachever grâce à ses compétences philologiques

8

INTRODUCTION

et codicologiques remarquables, tout en pouvant bénéficier des conseils érudits de son ami, disparu trop tôt, Marc Geoffroy. Une collaboration analogue, à la fois fructueuse et amicale, avec Marc Geoffroy et Meryem Sebti fut pour Jules Janssens l’occasion de montrer les mêmes compétences dans le domaine de la philologie arabe. Il en résulta une édition critique, avec traduction annotée : Avicenne (Ibn Sīnā), Commentaire sur le livre Lambda de la Métaphysique d’Aristote (chapitres 6-10) (2014). Loin de se limiter à la seule édition de l’Avicenna Latinus, Janssens consacra plusieurs articles à sa réception dans la philosophie latine médiévale (« Some Elements of Avicennian Influence on Henry of Ghent’s Psychology » [1996] ; « L’Avicenne latin : un témoin (indirect) des commentateurs (Alexandre d’Aphrodise, Thémistius, Jean Philopon) » [1999] ; « L’Avicenne latin : particularités d’une traduction » [2002] ; « Elements of Avicennian Metaphysics in the Summa » [2003] ; « The Reception of Avicenna’s Physics in the Latin Middle Ages » [2007] ; « Henry of Ghent and Avicenna » [2010] ; « The Latin Translation of the Physics : A Useful Source for the Critical Edition of the Arabic Text ? » [2012] ; « The Physics of the Avicenna Latinus and Its Significance for the Reception of Aristotle’s Physics in the Latin West » [2012] ; « A Survey of Thomas’s Explicit Quotations of Avicenna in the Summa contra Gentiles » [2014] ; « The Liber primus naturalium, i.e. the Physics of the Avicenna Latinus » [(2017] ; « Les commentaires d’Andrea Alpago au Compendium de anima d’Avicenne » [2018]). Et comment citer Jules Janssens sans évoquer sa bibliographie annotée d’Avicenne ? Couvrant à la fois la tradition arabe et latine, il s’agit d’un outil indispensable des études avicenniennes (An Annotated Bibliography on Ibn Sînâ (1970-1989) [1991] ; An Annotated Bibliography on Ibn Sînâ: First Supplement (1990-1994) [1999] ; An Annotated Bibliography on Ibn Sînâ: Second Supplement (1995-2009) [2017]). Il faut beaucoup de générosité intellectuelle pour s’atteler à une tâche telle que la constitution de cette bibliographie. C’est précisément cette générosité qui caractérise Jules Janssens. Qui a eu comme nous le privilège de travailler avec lui le sait. Jules Janssens aime par-dessus tout transmettre son savoir. Tout au long de notre carrière universitaire, il a toujours été là pour enrichir, discuter, critiquer s’il le faut les travaux que nous lui soumettions. Toujours dans la bienveillance et le partage.

INTRODUCTION

9

Pour rendre hommage aux qualités intellectuelles et humaines de Jules Janssens nous avons pris l’initiative de cet ouvrage. Il réunit quatorze contributions originales qui couvrent à-peu-près tous les domaines explorés par Janssens dans ses nombreuses publications. Nous commençons, chronologiquement, avec la contribution de Michael Chase, consacrée à la paraphrase arabe d’une partie des Ennéades de Plotin connue sous le nom de Théologie d’Aristote, qui met en évidence la manière subtile dont l’auteur de cette paraphrase a rendu les notions d’acte et de puissance. Cette Théologie est une des sources majeures de la falsafa, la philosophie dans le monde musulman. Avicenne s’y intéressait de près, au point d’en avoir rédigé un commentaire, qui ne nous est parvenu que de façon fragmentaire. Jules Janssens, Meryem Sebti et Michael Chase en préparent actuellement une édition critique. Les nombreuses difficultés liées à cette édition, notamment en raison de la complexité de la transmission du texte, forment l’objet de l’article de Meryem Sebti. Daniel De Smet s’intéresse pour sa part à la philosophie politique d’al-Fārābī, le prédécesseur le plus connu d’Avicenne, en soulevant la question épineuse d’une possible influence shi‘ite, éventuellement ismaélienne, dans la manière dont al-Fārābī a islamisé la figure platonicienne du philosophe-roi. Jules Janssens a toujours manifesté un vif intérêt pour la métaphysique d’Avicenne et les nombreux problèmes qu’elle soulève. Damien Janos poursuit sur cette voie en examinant à nouveaux frais la question difficile de l’« analogie de l’être » et de sa modulation (tashkīk al-wujūd), tandis que Jari Kaukua reprend le long débat sur le déterminisme face à la liberté humaine dans l’éthique avicennienne. Dimitri Gutas se penche sur la notion-clé de l’ittiṣāl, la « conjonction » de l’intellect humain avec l’Intellect agent, qui a plus d’une fois prêté à équivoque dans les différentes interprétations d’Avicenne, en particulier celles de sa noétique et de son eschatologie. Un des grands défis pour les avicennisants est de déterminer les limites du corpus avicennien, en essayant d’en éliminer les écrits apocryphes. Thérèse-Anne Druart et Frank Griffel se sont attelés à cette tâche, en montrant avec des arguments doctrinaux, historiques et philologiques qu’Avicenne ne peut être l’auteur de deux ouvrages qui lui sont attribués par la tradition : le Kitāb al-Siyāsa ou Tadbīr al-manzil (Druart) et la Risāla al-‘arshiyya (Griffel). Cette dernière aurait été

10

INTRODUCTION

écrite par un avicennien postérieur à al-Ghazālī, qui avait une bonne connaissance de la théologie ash‘arite, alors dominante dans le sunnisme. Cela nous amène à la figure énigmatique d’al-Ghazālī, qui depuis longtemps fascine Jules Janssens, en particulier par la manière dont ce théologien ash‘arite a réfuté l’avicennisme tout en en adoptant des pans entiers, quitte encore à les intégrer dans une doctrine imprégnée de soufisme. Taneli Kukkonen rouvre le dossier de la fameuse crise intellectuelle qui frappa al-Ghazālī en 1095, une sorte de « court-circuit » causé par la confrontation dans l’esprit d’un même homme de trois courants de pensée apparemment incompatibles : l’avicennisme, l’ash‘arisme et le soufisme. Une attitude analogue d’amour et de haine envers la philosophie d’Avicenne caractérise cet autre théologien ash‘arite qu’était Fakhr al-Dīn al-Rāzī. Comme le montre Amos Bertolacci par une étude des manuscrits, al-Rāzī joua un rôle déterminant dans la transmission du Kitāb al-Shifā’, et cela en dépit de ses désaccords avec Avicenne. Olga Lizzini nous brosse un tableau de la première réception latine d’Avicenne au XIIe siècle. C’est le début d’une longue transmission de concepts avicenniens qui, à travers la philosophie du Moyen-Âge et de la Renaissance (Carlos Steel et Guy Guldentops), nous mène jusqu’au XXe siècle (Andrea Robiglio). À travers cette publication nous espérons pouvoir rendre à Jules Janssens un peu de tout ce qu’il nous a donné tant comme ami que comme collègue. Nous remercions vivement les auteurs qui ont contribué à ce volume, ainsi que la rédaction de la collection Bibliotheca des Recherches de Théologie et Philosophie Médiévales, en particulier Carlos Steel et Guy Guldentops. Nous tenons également à remercier le Centre De Wulf-Mansion (KU Leuven) pour son soutien financier. Daniel DE SMET / Meryem SEBTI

BIBLIOGRAPHIE DE JULES JANSSENS LIVRES : Avicenna: tussen neoplatonisme en islam, Leuven 1984 [thèse de doctorat soutenue à la KU Leuven, inédit]. An Annotated Bibliography on Ibn Sînâ (1970-1989). Including Arabic and Persian Publications, and Turkish and Russian References (Ancient and Medieval Philosophy, Series I, vol. 13), Leuven 1991. An Annotated Bibliography on Ibn Sînâ: First Supplement (1990-1994) (FIDEM, Textes et études du Moyen Âge, 12), Louvain-la-Neuve 1999. Avicenna and His Heritage. Acts of the International Colloquium Leuven Louvain-la-Neuve, September 8 - September 11, 1999, éd. J. JANSSENS – D. DE SMET (Ancient and Medieval Philosophy, Series I, vol. 28), Leuven 2002. Ibn Sīnā and His Influence on the Arabic and Latin World (Variorum Reprints, CS 843), Aldershot 2006. Avicenna Latinus, Liber primus Naturalium. Tractatus secundus. De motu et de consimilibus, édition critique par S. VAN RIET – J. JANSSENS – A. ALLARD. Introduction doctrinale par G. VERBEKE, Bruxelles 2006. Avicenne (Ibn Sīnā), Commentaire sur le livre Lambda de la Métaphysique d’Aristote (chapitres 6-10), édition critique, traduction et notes par M. GEOFFROY – J. JANSSENS – M. SEBTI (Études musulmanes, 43), Paris 2014. An Annotated Bibliography on Ibn Sînâ: Second Supplement (1995-2009) (Medieval Renaissance Texts and Studies, 439), Tempe, Arizona 2017. Avicenna Latinus. Liber primus Naturalium. Tractatus tertius. De his quae habent naturalia ex hoc quod habent quantitatem, édition critique par J. JANSSENS, Bruxelles 2017. ARTICLES : « Le Dānesh-Nāmeh d’Ibn Sīnā : un texte à revoir ? », dans : Bulletin de philosophie médiévale 28 (1986), pp. 163-177. « Ibn Sīnā’s Ideas on Ultimate Reality and Meaning. Neoplatonism and the Qur’ān as Problem-Solving Paradigms in the Avicennian System », dans : Ultimate Reality and Meaning 10 (1987), pp. 252-271. « Filosofische elementen in de mystieke leer van al-Ghazzālī », dans : Tijdschrift voor Filosofie 50 (1988), pp. 334-342. « L’idéal politique selon le Nahj al-Balāghah », dans : al-Muntaka 5 (1989-90), pp. 9-33.

12

BIBLIOGRAPHIE DE JULES JANSSENS

« Le Ma‘ārij al-quds fī madārij ma‘rifat al-nafs : un élément-clé pour le dossier Ghazzālī - Ibn Sīnā ? », dans : Archives d’histoire doctrinale et littéraire du Moyen Âge 60 (1993), pp. 27-55. « Le Livre des Religions et des Sectes d’al-Shahrastānī : une conception particulière d’historiographie de la pensée », dans : Bulletin de philosophie médiévale 35 (1993), pp. 106-112. « The Idea of God in al-Kindī », dans : Ultimate Reality and Meaning 17 (1994), pp. 4-16. « Some Elements of Avicennian Influence on Henry of Ghent’s Psychology », dans : W. VANHAMEL (éd.), Henry of Ghent. Proceedings of the International Colloquium on the Occasion of the 700th Anniversary of his Death (1293), Leuven 1996, pp. 155-169. « The Problem of Human Freedom in Ibn Sīnā », dans : P. LLORENTE (éd.), Actes del Simposi Internacional de Filosofia de l’Edat Mitjana. El pensament antropològico medieval en els àmbits islàmic, hebreu i cristià (Vic-Gerona, 11-16 abril de 1993), Vic 1996, pp. 112-118. « Creation and Emanation in Ibn Sīnā », dans : Documenti e Studi sulla Tradizione Filosofica Medievale 8 (1997), pp. 455-477. « Les Ta‘līqāt d’Ibn Sīnā, essai de structuration et de datation », dans : A. DE LIBÉRA – A. ELAMRANI-JAMAL – A. GALONNIER (éds.), Langages et Philosophie. Hommage à Jean Jolivet, Paris 1997, pp. 109-122. « Le Livre des définitions d’al-Jurjānī : un dictionnaire scientifique bien particulier », dans : Bulletin de philosophie médiévale 39 (1997), pp. 131145. « Ibn Sīnā (Avicenne) : un projet religieux de philosophie ? », dans : J. A. AERTSEN – A. SPEER (éds.), Was ist Philosophie im Mittelalter ?, Berlin / New York 1998, pp. 863-870. « L’Avicenne latin : un témoin (indirect) des commentateurs (Alexandre d’Aphrodise, Thémistius, Jean Philopon) », dans : R. BEYERS – J. BRAMS – D. SACRÉ – K. VERRYCKEN (éds.), Tradition et traduction. Les textes philosophiques et scientifiques au Moyen Âge latin. Hommage à F. Bossier, Leuven 1999, pp. 89-105. « Al-Ghazzālī’s Tahāfut : Is It Really a Rejection of Ibn Sīnā’s Philosophy ? », dans : Journal of Islamic Studies 12 (2001), pp. 1-17. « Ibrahim’in Rehberliǧinde : Hosgörü Sahibi Bir Nesil Yetiştirmek » [« In the Faith of Abraham : Educating Young People in the Spirit of Real Tolerance »], dans : M. S. AYDIN (éd.), Hazret-i Ibrahim’in İzinde, Istanbul 2001, pp. 225-240. « L’Avicenne latin : particularités d’une traduction », dans : J. JANSSENS – D. DE SMET (éds.), Avicenna and His Heritage. Acts of the International Colloquium Leuven - Louvain- la-Neuve, September 8 - September 11, 1999, Leuven 2002, pp. 113-129. « Mullā Sadrā’s Use of Ibn Sīnā’s Ta‘līqāt in the Asfār », dans : Journal of Islamic Studies 13 (2002), pp. 1-17.

BIBLIOGRAPHIE DE JULES JANSSENS

13

« L’utilisation des Ta‘līqāt d’Ibn Sīnā dans les Asfār de Mullā Sadrā Shīrāzī » dans : Mulla Sadra and Comparative Studies. Islam-West Philosophical Dialogue. The Papers Presented at the World Congress of Mulla Sadra (May, 1999, Tehran), Téhéran 2002, vol IV, pp. 245-258. « Al-Ghazzālī’s Mi’yār al-‘ilm fī fann al-mantiq : sources avicenniennes et farabiennes », dans : Archives d’histoire doctrinale et littéraire du Moyen Âge 69 (2002), pp. 39-66. « Bahmanyār ibn Marzubān : A Faithful Disciple of Ibn Sīnā ? », dans : D. C. REISMAN, with the assistance of A. H. AL-RAHIM (éds.), Before and After Avicenna. Proceedings of the First Conference of the Avicenna Study Group, Leiden / Boston 2003, pp. 177-197. « Elements of Avicennian Metaphysics in the Summa », dans : G. GULDENTOPS – C. STEEL (éds.), Henry of Ghent and the Transformation of Scholastic Thought. Studies in Memory of Jos Decorte, Leuven 2003, pp. 41-59. « Al-Ġazālī and his Use of Avicennian texts », dans : M. MAROTH (éd.), Problems in Arabic Philosophy, Piliscsaba 2003, pp. 37-49. « Avicenne et sa “paraphrase-commentaire” du livre Lambda (Kitāb al-Inṣāf ) », dans : Recherches de Théologie et Philosophie médiévales 70 (2003), pp. 401416. « Avicenna and the Qur’ān. A Survey of his Qur’ānic Commentaries », dans : MIDEO 25-26 (2004), pp. 177-192 [repris dans : M. SHAH (éd.), Tafsir. Interpreting the Qur’an, Londres 2012, vol. III, pp. 457-471]. « Al-Ghazzālī’s Political Thought : Elements of Greek Philosophical Influence », dans : E. GANNAGÉ – P. CRONE – M. AOUAD – D. GUTAS – E. SCHÜTRUMPF (éds.), The Greek Strand in Islamic Political Thought. Proceedings of the Conference Held at the Institute for Advanced Study, Princeton, 16-27 June 2003, numéro spécial des Mélanges de l’Université Saint-Joseph 57 (2004), pp. 393-410. « “Experience” (tajriba) in Classical Arabic Philosophy (al-Fārābī - Ibn Sīnā) », dans : Quaestio 4 (2004), pp. 45-62. « The Notions of Wāhib al-ṣuwar (Giver of Forms) and Wāhib al-‘aql (Bestower of Intelligence) in Ibn Sīnā », dans : M. C. PACHECO – J. F. MEIRINHOS (éds.), Intellect et Imagination dans la Philosophie Médiévale. Actes du XIe Congrès International de Philosophie Médiévale de la SIEPM (Porto, du 26 au 31 août 2002), Turnhout 2006, pp. 551-562. « Vocabulaire d’Avicenne », dans : J.-P. ZARADER (éd.), Le vocabulaire des philosophes. Suppléments I (Volume V), Paris 2006, pp. 13-64. « The Reception of Avicenna’s Physics in the Latin Middle Ages », dans : A. VROLIJK – J. P. HOGENDIJK (éds.), O ye Gentlemen: Arabic Studies on Science and Literary Culture in Honour of Remke Kruk, Leiden / Boston 2007, pp. 55-64. « Some Classics of Islamic Philosophy and their Reception in the Western World », dans : H. ÖZLAN – N. ARDIÇ – A. ARLI (éds.), Medeniyet veKlasik, Istanbul 2007, pp. 197-213.

14

BIBLIOGRAPHIE DE JULES JANSSENS

« A Remarkable Thirteenth Century Compendium of “Aristotelian” Philosophy: Ibn Sab‘īn’s Sicilian Questions », dans : Bulletin de philosophie médiévale 49 (2007), pp. 51-66. « Al-Kindī : The Founder of Philosophical Exegesis of the Qur’ān », dans : Journal of Qur’anic Studies 9 (2007), pp. 243-263 [repris dans : M. SHAH (éd.), Tafsir. Interpreting the Qur’an, Londres 2012, vol. III, pp. 436-456]. « Bahmanyār and His Revision of Ibn Sīnā’s Metaphysical Project », dans : Medioevo 32 (2007), pp. 99-117. « Al-Ghazzālī’s Mizān al-‘amal : An Ethical Summa Based on Ibn Sīnā and Rāghib al-Isfahānī », dans : A. AKASOY – W. RAVEN (éds.), Islamic Thought in the Middle Ages. Studies in Text, Transmission and Translation, in Honour of Hans Daiber, Leiden / Boston 2008, pp. 123-137. « L’âme-miroir : al-Ġazālī entre philosophie et mysticisme », dans : D. DE SMET – M. SEBTI – G. DE CALLATAŸ (éds.), Miroir et savoir. La transmission d’un thème platonicien, des Alexandrins à la philosophie arabomusulmane. Actes du Colloque International tenu à Leuven et Louvain-laNeuve les 17 et 18 novembre 2005, Leuven 2008, pp. 203-217. « Ibn Sînâ: An Important Historian of the Sciences », dans : N. BAYHAN – M. MAZAK – N. ÖZKAYA (éds.), Ulusraralası İbn Sînâ Sempozyumu. Bildirirer. 22-24 Mayıs 2008, İstanbul. / International Ibn Sina Symposium. Papers, May, 22-24, 2008, Istanbul, Istanbul 2008-9, II, pp. 83-93 ; trad. turque par O. BAŞ, ibid., II, pp. 94-103. « Abū Bakr al-Rāzī », dans : Fr. MÖRI (éd.), Orient – Occident. Racines spirituelles de l’Europe. S. l., Paris 2009, pp. 498-499 ; « Avicenne », ibid., pp. 504-505. « Al-Amidi and His Integration of Philosophy into Kalam », dans : A. ERKOL – A. ADAK – İ. BOR (éds.), Uluslararası Seyfuddîn Âmidî Sempozyumu Bildirileri / International Conference on Sayf Al-Din Al-Amidi Papers, Istanbul 2009, pp. 305-323. « Elements of Avicennian Influence in al-Fanârî’s Theory of Emanation », dans : T. YÜCEDOĞRU – O. Ş. KOLOĞLU – U. M. KILAVUZ – K. GÖMBEYAZ (éds.), Uluslararası Molla Fenârî Sempozyumu (4-6 Aralık 2009 Bursa) Bildilirer / International Symposium on Molla Fanârî (4-6 Decelver 2009 Bursa). Proceedings, Bursa 2010, pp. 315-327. « Al-Ghazzālī : The Introduction of Peripatetic Syllogistic Theory in Islamic Law (and Kalām) », dans : MIDEO 28 (2010), pp. 219-233. « Henry of Ghent and Avicenna », dans : G. WILSON (éd.), A Companion to Henry of Ghent (Brill’s Companions to the Christian tradition, 23), Leiden / Boston 2010, pp. 63-83. « Henry of Ghent and Averroes », dans : G. WILSON (éd.), A Companion to Henry of Ghent (Brill’s Companions to the Christian Tradition, 23), Leiden / Boston 2010, pp. 85-99.

BIBLIOGRAPHIE DE JULES JANSSENS

15

« Ibn Sīnā’s Impact on Fakhr ad-Dīn ar-Rāzī’s Mabāḥith al-Mashriqiyya, with Particular Regard to the Section Entitled al-Ilāhiyyāt al-maḥḍa : An Essay of Critical Evaluation », dans : Documenti e Studi sulla Tradizione Filosofica Medievale 21 (2010), pp. 259-285. « The Reception of Ibn Sīnā’s Physics in Later Islamic Thought », dans : Ilahiyyat Studies 1 (2010), pp. 15-36. « Ibn Sīnā (Avicenna), The Latin Translations of », dans : H. LAGERLUND (éd.), Encyclopedia of Medieval Philosophy. Philosophy between 500 and 1500, Dordrecht 2011, pp. 522-527 ; version revue (2020) sur https:// doi.org/10.1007/978-94-024-1665-7_232. « Al-Ġazzālī’s Maqāṣid al-Falāsifa, Latin Translation of », dans : H. LAGERLUND (éd.), Encyclopedia of Medieval Philosophy. Philosophy between 500 and 1500, Dordrecht 2011, pp. 387-390 ; version revue (2020) sur https://doi.org/10.1007/978-94-024-1665-7_181. « Ibn Bājja and Aristotle’s Political Thought », dans : V. SYROS (éd.), Well Begun is Only Half Done: Tracing Aristotle’s Political Ideas in Medieval Arabic, Syriac, Byzantine, and Jewish Sources (Medieval and Renaissance Texts and Studies, 338), Tempe, Arizona 2011, pp. 73-95. « Ibn Rushd et sa critique d’Ibn Bājja (dans le Grand commentaire sur le De Anima) », dans : A. HASNAWI (éd.), La lumière de l’intellect. La pensée scientifique et philosophique d’Averroès dans son temps. Actes du IVe colloque international de la SIHSPAI (Cordoue, 9-12 décembre 1998), Leuven 2011, pp. 405-418. « Al-Ghazālī between Philosophy (Falsafa) and Sufism (Taṣawwuf): His Complex Attitude in the Marvels of the Heart (‘Ajā’ib al-qalb) of the Ihyā’ ‘Ulūm al-Dīn », dans : The Muslim World 101 (2011), pp. 614632. « Al-Lawkarī’s Reception of Ibn Sīnā’s Ilāhiyyāt », dans : D. N. HASSE – A. BERTOLACCI (éds.), The Arabic, Hebrew and Latin Reception of Avicenna’s Metaphysics (Scientia Graeco-Arabica, 7), Berlin 2012, pp. 7-26. « Ibn Sīnā’s Ta‘līqāt : The Presence of Paraphrases of and Super-Commentaries on the Ilāhiyyāt of the Shifā’ », dans : F. OPWIS – D. REISMAN (éds.), Islamic Philosophy, Science, Culture, and Religion. Studies in Honor of Dimitri Gutas, Leiden / Boston 2012, pp. 201-222. « Al-Birr wa-l-ithm, Piety and Sin: Possible Farabian Influences on the Young Ibn Sina », dans : Ishraq 3 (2012), pp. 412-422. « Fakhr ad-Dīn ar-Rāzī on the Soul: A Critical Approach to Ibn Sīnā », dans : The Muslim World 102 (2012), pp. 562-579. « The Latin Translation of the Physics: A Useful Source for the Critical Edition of the Arabic Text ? », dans : Oriens 40 (2012), pp. 515-528. « The Physics of the Avicenna Latinus and Its Significance for the Reception of Aristotle’s Physics in the Latin West », dans : A. M. I. VAN OPPENRAAIJ,

16

BIBLIOGRAPHIE DE JULES JANSSENS

with the collaboration of R. FONTAINE (éds.), The Letter before the Spirit. The Importance of Text Editions for the Study of the Reception of Aristotle (Aristoteles Semitico-Latinus, 22), Leiden / Boston 2012, pp. 311-330. « An Essay Review of In the Age of Averroes: Arabic Philosophy in the Sixth/ Twelfth Century, edited by Peter Adamson (Warburg Institute Colloquia, 16), (London : The Warburg Institute / Turin : Nino Aragno Editore, 2011) », dans : Ilahiyat Studies 3 (2012), pp. 117-126. « The Ikhwān aṣ-Ṣafā’ on King-Prophet Solomon », dans : J. VERHEYDEN (éd.), The Figure of Solomon in Jewish, Christian and Islamic Tradition. King, Sage and Architect (Themes in Biblical Narrative. Jewish and Christian Traditions, 16), Leiden / Boston 2013, pp. 241-253. « La philosophie peut-elle contribuer au dialogue interreligieux ? Réflexions à partir de Ghazālī, Maïmonide, Thomas d’Aquin et Yaḥyā’ ibn ‘Adī », dans : B. BROECKAERT – S. VAN DEN BRANDEN – J.-J. PERENNÈS (éds.), Perspectives on Islamic Culture. Essays in Honour of Emilio G. Platti (Les Cahiers du MIDEO, 6), Louvain / Paris 2013, pp. 197-218. « Ibn Sīnā on Substance in Chapter Two of the Maqūlāt (Categories) of the Shifā’ », dans : P. FODOR – G. MAYER – M. MONOSTORI – K. SVOZÁK – L. TAKÁOS (éds.), More modoque. Die Wurzeln der europäischen Kultur und deren Rezeption im Orient und Okzident. Festschrift für Miklós Maróth zum siebzigsten Geburtstag, Budapest 2013, pp. 353-360. « Albert le Grand et sa connaissance des écrits logiques arabes : une réévaluation du dossier Grignaschi », dans : J. BRUMBERG-CHAUMONT (éd.), Ad notitiam ignoti. L’Organon dans la translatio studiorum à l’époque d’Albert le Grand (Studia Artistarum, 37), Turnhout 2013, pp. 225-257. « Al-Ġazālī: How to read the Munqiḏ ? », dans : C. BAFFIONI – R. B. FINAZZI – A. PASSONI DELL’ACQUA – E. VERGANI (éds.), Al-Ġazālī (1058-1111), La prima stampa armena, Yehūdāh Ha-Lēvī (1075-1141), La ricezione di Isacco di Ninive (Orientalia Ambrosiana, 2), Milan / Rome 2013, pp. 85-97. « What about Providence in the Best of All Possible Worlds ? Avicenna and Leibniz », dans : P. D’HOINE – G. VAN RIEL (éds.), Fate, Providence and Moral Responsibility in Ancient, Medieval and Early Modern Thought. Studies in Honour of Carlos Steel (Ancient and Medieval Philosophy, Ser. I, vol. 49), Leuven 2014, pp. 441- 454. « Éléments avicenniens dans le livre al-Maqṣad d’al-Ghazālī », dans : MIDEO 30 (2014), pp. 91- 103. « La philosophie d’Avicenne : éléments d’influence de penseurs alexandrins », dans : Ch. MÉLA – F. MÖRI, en collaboration avec S. M. AUFFREY – G. DORIVAL – A. LE BOULLUEC (éds.), Alexandrie la divine, Genève 2014, vol. II, pp. 870-875.

BIBLIOGRAPHIE DE JULES JANSSENS

17

« A Survey of Thomas’s Explicit Quotations of Avicenna in the Summa contra Gentiles », dans : American Catholic Philosophical Quarterly 88 (2014), pp. 289-308. « L’utilisation des Mubāḥathāt d’Ibn Sīnā par Mullā Ṣadrā dans ses Asfār », dans : M. A. MENSIA (éd.), Views on the Philosophy of Ibn Sīnā and Mullā Ṣadrā Shīrāzī, Tunis 2014, pp. 87-103. « Le De divisione philosophiae de Gundissalinus : Quelques remarques préliminaires à une édition critique », dans : E. CODA – C. MARTINI BONADEO (éds.), De l’Antiquité tardive au Moyen Âge. Études de logique aristotélicienne et de philosophie grecque, syriaque, arabe et latine offertes à Henri Hugonnard-Roche (Études Musulmanes, 44), Paris 2014, pp. 559-570. « Al-Fārābī : la religion comme imitation de la philosophie », dans : M. DELGADO – Ch. MÉLA – F. MÖRI (éds.), Orient-Occident. Racines spirituelles de l’Europe, Paris 2014, pp. 497-512. « Foi et raison chez Avicenne », dans : M. DELGADO – Ch. MÉLA – F. MÖRI (éds.), Orient-Occident. Racines spirituelles de l’Europe, Paris 2014, pp. 513527. « R. Marti and His References to al-Ghazālī » dans : G. TAMER (éd.), Islam and Rationality. The Impact of al-Ghazālī. Papers Collected on His 900th Anniversary. Volume 1 (Islamic Philosophy, Theology, and Science, 94), Leiden / Boston 2015, pp. 326-344. « Fakhr al-Dīn al-Rāzī’s Use of al-Ghazzālī’s Mishkāt in His Commentary on the Light Verse (Q. 24:35) », dans : F. GRIFFEL (éd.), Islam and Rationality. The Impact of al-Ghazālī. Papers Collected on His 900th Anniversary. Volume 2 (Islamic Philosophy, Theology, and Science, 98), Leiden / Boston 2016, pp. 229-252. « Metaphysics of God », dans : R. C. TAYLOR – L. X. LÓPEZ-FARJEAT (éds.), The Routledge Companion to Islamic Philosophy, Londres / New York 2016, pp. 236-247. « Ibn Sīnā. A Philosophical Mysticism or a Philosophy of Mysticism ? », dans : Mediterranea. International Journal for the Transfer of Knowledge 1 (2016), pp. 37-55. « Abū al-Barakāt al-Baghdādī and His Use of Ibn Sīnā’s al-Ḥikma al-‘Arūḍiyya (or another work closely related to it) in the Logical Part of His Kitāb al-Mu‘tabar », dans : Nazariyat. Journal for the History of Islamic Philosophy and Sciences 3/1 (2016), pp. 1-21. « Ibn Sīnā on “Experience” (tajriba) and Its Particular Significance for the Evaluation of Simple Drugs », dans : Al-Mu’tamar al-‘alamī al-awwal li-tārīkh al-‘ulūm al-taṭbīqiyya wa-l-ṭibbiyya ‘inda l-‘Arab wa-l-muslimīn (First International Conference on the History of Applied and Medical Sciences in the View of Arabs and Muslims), Riadh 1438 H. (= 2017), t. V, pp. 299-309.

18

BIBLIOGRAPHIE DE JULES JANSSENS

« The Latin Translation of the Epistle on Geography of the Ikhwān aṣ-Ṣafā’. A Few Preliminary Remarks in View of a Critical Edition », dans : Studi Maġrebini. Nuova Serie 12-13 (2017) (= Labor limae. Atti in Onore de Carmela Baffioni), pp. 367-380. « The Liber primus naturalium, i.e. the Physics of the Avicenna Latinus », dans : Documenti e Studi sulla Tradizione Filosofica Medievale 28 (2017), pp. 219-238. « A Critical Note on the Latin Translation of Aristotle’s De Anima Used by Zabarella in His Commentary on This Work », dans : Revista Española de Filosofía Medieval 24 (2017), pp. 269-276. « Ibn Sīnā’s Aristotle », dans : Mediterranea. International Journal for the Transfer of Knowledge 3 (2018), pp. 129-144. « Avicennian Elements in Faḫr al-Dīn al-Rāzī’s Discussion of Place, Void and Directions in the al-Mabāḥiṯ al-mašriqiyya », dans : D. N. HASSE – A. BERTOLACCI (éds.), The Arabic, Hebrew and Latin Reception of Avicenna’s Physics and Cosmology (Scientia Graeco-Arabica, 23), Boston / Berlin 2018, pp. 43-63. « L’utilisation du livre (pseudo-)aristotélicien Kalām fī al-Khayr al-maḥḍ dans le Budd al-‘ārif d’Ibn Sab‘īn », dans : A. ABOUCHRAA (éd.), Reality and Perspectives of Research on the History of Thought in the Islamic World. Critical Revisions on Philosophy, Sufism, and Jurisprudence. Works Dedicated to the Moroccan Thinker Dr. Abdelmajid Ashagiir, Rabat 2018, pp. 662-649. « Les commentaires d’Andrea Alpago au Compendium de anima d’Avicenne », dans : H. PASQUA (éd.), Avicenne. Ibn Sīnā [980-1037]. L’être et l’essence (= Noesis 32 [2018]), pp. 113-137. « Signification des études avicenniennes (philosophiques, scientifiques et médicales) pour la pensée contemporaine (occidentale et orientale) », dans : J.-B. BRENET – O. L. LIZZINI (éds.), La philosophie arabe à l’étude. Sens, limites et défis d’une discipline moderne / Studying Arabic Philosophy. Meaning, Limits and Challenges of a Modern Discipline, Paris 2019, pp. 675690. « Al-Ghazālī’s Use of Avicennian Texts in his Maqāṣid al-falāsifa », dans : Ishraq. Islamic philosophy yearbook 9 (2019), pp. 80-121. « The Significance of Avicenna’s Additional Remarks to the Dāneš-Nāmeh in His Ta‘līqāt, with Special Attention to the Sections on Divine Knowledge and Will », dans : M. GÜNEŞ (éd.), Avicenna, ein Universalgelehter. Eine gegenwärtige Analyse des Prinzen der Philosophen Ibn Sīnā, Göttingen 2020, pp. 57-74. « Philosophical Commentaries », dans : M. SHAH – M. ABDEL HALEEM (éds.), The Oxford Handbook of Qur’anic Studies, Oxford 2020, pp. 780793.

BIBLIOGRAPHIE DE JULES JANSSENS

19

« Bahmanyār ibn Marzubān », dans : H. LAGERLUND (éd.) Encyclopedia of Medieval Philosophy, Dordrecht [2020] : https://doi.org/10.1007/97894-024-1665-7_606 « La position de Zabarella vis-à-vis d’Averroès dans son Commentaire sur le De anima », dans : Mélanges de l’Université Saint-Joseph 68 (2019-20), pp. 105-124. « Doubles traductions et omissions : une approche critique en vue d’une édition de la traduction latine du Liber de Causis », dans : D. CALMA (éd.), Reading Proclus and the Book of Causes. Vol. 2. Translations and Acculturations, Leiden / Boston 2021, pp. 275-316. De très nombreux comptes rendus, entre autres dans Tijdschrift voor Filosofie, Revue philosophique de Louvain, Le Muséon, Journal of Ancient Philosophy, MIDEO, etc.

ACTE, PUISSANCE ET RÉVEIL DANS LE VIIIe MĪMAR DE LA THÉOLOGIE D’ARISTOTE1 Michael CHASE Abstract Faced by the ambiguity of the Greek term δύναμις in the first five chapters of Plotinus, Enneads IV, 4 (28), the Arabic Adaptor responsible for the pseudonymous Theology of Aristotle develops a highly original doctrine of the reversal of roles between power or potency (Greek δύναμις /Arabic al-quwwa) 1. Mss. de la Théologie d’Aristote consultés : Ṣ = Istanbul, AyaSofya 2457, 863/1459 ; K = al-Kāẓimiyya 269 Éditions et traductions de la Théologie d’Aristote : B = Aflūtīn ῾inda al-῾Arab / Plotinus apud arabes, nuṣūṣ ḥaqqaqa-hā wa-qaddama lahā ῾A. BADAWI, Cairo 1955. D = Die sogenannte Theologie des Aristoteles, aus arab. Hss. zum 1. Mal hrsg. von Fr. DIETERICI, Leipzig 1882, réimpr. Amsterdam 1965. L = traduction de LEWIS, voir HS1. Éditions et traductions de l’Ennéade IV de Plotin : A = Plotinus with an English translation by A. H. ARMSTRONG. t. 4, Enneads IV. 1-9, Cambridge, Mass. 1984. BD = Plotinus, Ennead IV, 3-14, 4, 29 : Problems concerning the soul, translation with an introduction and commentary by John M. DILLON and H. J. BLUMENTHAL, Las Vegas etc., 2015. Br. = Plotin, Traités 27-29, Sur les difficultés relatives à l’âme. Trois livres, présentés, traduits et annotés par Luc BRISSON, Paris 2005. HS1 = Plotini opera Tomus 2, Enneades 4-5, ediderunt Paul HENRY et Hans-Rudolf SCHWYZER ; Plotiniana Arabica ad codicum fidem anglice vertit Geoffrey LEWIS, Paris / Bruxelles 1959. HS2 = Plotini Opera 2, Enneades IV-V, ediderunt Paul HENRY et Hans-Rudolph SCHWYZER, Oxford 1977. HTB = Plotins Schriften, übersetzt von Richard HARDER, Neubearbeitung mit griechischem Lesetext und Anmerkungen fortgeführt von Rudolf BEUTLER und Willy THEILER. Band III: Die Schriften 22-29 der chronologischen Reihenfolge, 2 tomes, Hamburg 1962. I = Plotino, Enéadas III-IV, introducciones, traducciones y notas de Jesús IGAL, Madrid 1985. Rad. = Plotino, Enneadi, Porfirio, Vita di Plotino, (...) Introduzione, traduzione, note e bibliografia di Giuseppe FAGGIN (...) Revisione finale dei testi, appendici e indici di Roberto RADICE, Milano 2000. Reale = Plotino, Enneadi, Prefazione di Giovanni REALE, traduzione di Roberto RADICE, Milano 2008.

22

MICHAEL CHASE

and act, action or actuality (Greek ἐνέργεια/Arabic al-fi῾l) in the intelligible and the sensible world respectively. Whereas act is dominant in the sensible world, says the Adaptor, where power/potency must be led or bought to act, it is power perfect and complete (tāmmatun kāmilatun) that reigns in the intelligible world. This is explained by the doctrine of the barks or shells (al-qushūr), which veil both the soul and the objects of its cognition in the sensible world, a doctrine which may have been present in Porphyry. Another result of the Adaptor’s careful, insightful confrontation with Plotinus’s text is that of a separate faculty able to perceive the intelligibles even during life on earth, the possession of which is exclusively reserved for a spiritual élite described as the “people of happiness” (ahl al-sa῾āda). The text of the ThA thus preserves a key episode in the history of the interpretation of the notions of potentiality and act.

1. Introduction Il est bien connu que la Théologie d’Aristote, bien qu’elle consiste en une collection d’extraits tirés d’une traduction arabe paraphrastique des dernières trois Ennéades de Plotin (IV-VI), ne suit pas l’ordre de l’ouvrage plotinien tel qu’il nous est parvenu en grec, mais saute, de manière apparemment arbitraire, d’un passage tiré d’un traité plotinien à un autre, souvent à l’intérieur du même maymar (ou mīmar, « chapitre »). On trouve aussi le phénomène inverse : des passages qui se suivent à l’intérieur du même traité plotinien se trouvent parfois dispersés, dans la version arabe, entre divers mayāmir. C’est le cas pour le traité plotinien IV, 4 (28), que Porphyre a intitulé « Deuxième livre des apories concernant l’âme ». Les quatre premiers chapitres de ce traité sont paraphrasés au début du maymar II de la Théologie d’Aristote (pp. 29, 1-38, 12 Badawi) ; mais ce n’est que beaucoup plus tard, au maymar VIII (pp. 102, 12-108, 4 Badawi), que l’on renoue avec le reste du traité, en proposant une paraphrase des chapitres cinq à huit du traité plotinien. Tandis que l’auteur de la Théologie – que j’appellerai, en suivant la pratique de Peter Adamson, l’Adaptateur2, – reste assez proche du 2. Je reste, pour le moment, agnostique en ce qui concerne son identité, et surtout sur la question très importante de savoir si la personne responsable pour la traduction du texte grec de Plotin en arabe – c’est-à-dire, Ibn Nā῾ima al-Ḥimṣī – est autre que celui qui rédigea les différents passages de transition et d’introduction contenus dans la ThA. Je proposerai une solution possible à cette question à la fin de ce travail.

ACTE, PUISSANCE ET RÉVEIL

23

texte de Plotin pour ce qui concerne la paraphrase de cette première partie du traité plotinien, il a senti le besoin d’insérer des textes de transition après la fin de la paraphrase de IV, 4, 4 (au maymar II) et avant le début de la paraphrase de IV, 4, 5 (au maymar VIII), pour nouer ensemble ces deux unités textuelles. Il s’agit d’un phénomène assez fréquent dans la ThA, et qui reflète la volonté de l’Adaptateur de transformer sa collection d’extraits en un texte continu. Pour celui qui s’interroge sur la question de l’origine de la ThA, de tels passages de transition sont précieux. En effet, puisqu’ils s’éloignent du texte de Plotin, ces passages sont susceptibles de contenir des développements doctrinaux, des nuances ou des accents qui sont originaux à l’Adaptateur, et qui risquent donc de pouvoir nous renseigner sinon sur l’identité, au moins sur l’orientation philosophique de celui-ci. Par exemple, au maymar II, arrivé à mi-chemin de sa paraphrase du chapitre 4 de l’Ennéade IV, 4, l’Adaptateur abandonne le texte de Plotin pour se lancer (p. 37, 3-18 Badawi) dans un développement sur la supériorité de l’ignorance à la connaissance, texte où, après Pierre Thillet et d’autres, on a cru pouvoir reconnaître des traces de doctrines porphyriennes3. Puisque ce texte du maymar II a déjà fait l’objet d’études approfondies4, dans cet article je m’intéresserai plutôt au texte de transition qui, au maymar VIII de la ThA, précède la traduction arabe paraphrastique d’Ennéade IV, 4, 5. Cependant, pour permettre la compréhension de ce texte, il me faudra rappeler brièvement aussi le contenu de la paraphrase arabe d’Ennéade IV, 4, 1-4. 2. Dunamis dans Ennéade IV, 4, 1-4, ou l’ambiguïté de la puissance5 Le chapitre 4 du traité IV, 4 (28) a pour objet la question des souvenirs que garde l’âme de son séjour terrestre, aussi bien lorsque celle-ci 3. Voir M. CHASE, « Porphyry and the Theology of Aristotle”, dans : D. CALMA (éd.), Reading Proclus and the Book of Causes. Volume 2, Translations and acculturations, Leiden / Boston 2021, pp. 157-181. 4. Notamment de la part de Cr. D’ANCONA, « Il tema della docta ignorantia nel neoplatonismo arabo. Un contributo all’analisi delle fonti di Teologia di Aristotele, mimâr II », dans : G. PIAIA (éd.), Concordia Discors. Studi su Niccolò Cusano e l’umanesimo europeo offerti a Giovanni Santinello, Padova 1993, pp. 3-22. 5. Sur la dunamis chez Plotin, voir l’ouvrage important de G. AUBRY, Dieu sans la puissance. Dunamis et energeia chez Aristote et chez Plotin, Paris 2006 ; nouvelle édition revue et augmentée, 2020. Voir aussi, sur la manière dont la tradition arabe a fait face aux

24

MICHAEL CHASE

se trouve, entre deux incarnations, dans le monde intelligible, que lorsqu’elle a déjà entamé sa descente, tout d’abord jusqu’au monde céleste, puis, éventuellement, vers le monde sensible pour s’y incarner dans un corps terrestre. Au début du ch. 4, Plotin affirme que pendant son séjour dans le monde intelligible, l’âme ne voit le bien que par l’intermédiaire de l’Intellect. En effet, pendant qu’elle s’y trouve, rien ne saurait empêcher le Bien de pénétrer jusqu’à l’âme, puisqu’aucun corps ne se trouve entre les deux (IV, 4, 1, 1-4). Si l’âme dirige son attention vers les réalités inférieures, elle possède ce qu’elle souhaitait de manière analogue à sa mémoire et à son imagination (IV, 4, 1, 4-5) ; mais la mémoire, même si elle concerne ce qu’elle a de meilleur – il s’agit vraisemblablement des Formes intelligibles – n’est pas pour autant elle-même ce qu’il y a de meilleur. La mémoire est à comprendre non seulement au sens d’un souvenir conscient, mais il faut aussi prendre en compte la situation dans laquelle l’âme est disposée selon ce qu’elle a déjà vu ou expérimenté auparavant (IV, 4, 1, 7-9). En effet, les souvenirs inconscients peuvent exercer un effet encore plus fort que ceux dont on est conscient. Lorsqu’on en est conscient, il est possible de ne pas s’identifier à l’objet de son souvenir, mais si on ignore qu’on possède de tels souvenirs, on risque de s’y identifier sans le savoir, ce qui ne fait qu’aggraver, en l’approfondissant, la chute de l’âme (IV, 4, 1, 10-14). L’Adaptateur arabophone de la ThA a, semble-t-il, choisi de ne pas prendre en compte ce développement plotinien sur les souvenirs inconscients, ce qui est dommage, car il s’agit d’une observation d’une grande profondeur psychologique6. Au contraire, c’est à ce point de son argumentation qu’il se lance dans un exposé sur la supériorité de l’ignorance – interprétée, de manière très innovatrice par rapport au texte de Plotin, comme “imagination intellectuelle”7 – sur la connaissance. ambiguïtés de la doctrine aristotélicienne de la puissance, David WIRMER, Vom Denken der Natur zum Natur des Denkens. Ibn Bāǧǧa’s Theorie der Potenz als Grundlegung der Psychologie, Berlin etc. 2014 pp. 329–354. 6. Il y aurait lieu, en effet, de s’interroger sur la question de savoir si, dans ce texte, Plotin n’aurait point anticipé Freud par 1600 ans environ. Mais je laisse cette investigation à d’autres plus compétents en matière d’études psychanalytiques. 7. Al-tawahhum al-῾aqlī, ThA II, p. 37, 3 Badawi. Dans ce texte, cette notion d’imagination intellectuelle désigne la faculté grâce à laquelle l’âme, en sortant du monde intelligible, conçoit le monde sensible. Assimilée à une ignorance plus noble que la science, ce terme désigne, en outre, la modalité cognitive par laquelle l’intellect conçoit ce qui lui est

ACTE, PUISSANCE ET RÉVEIL

25

Jusqu’ici, l’argumentation de Plotin est relativement claire. Il s’agit notamment d’expliquer la chute de l’âme8. On a appris au ch. 3 que si l’âme quitte le monde intelligible, c’est qu’elle ne supporte plus son état d’identité avec l’Intellect et souhaite être autre, n’appartenant plus qu’à elle-même et se réjouissant de son autonomie9. Ce qui détermine la profondeur de la chute de l’âme, c’est le contenu de sa mémoire. D’une part, le souvenir qu’elle garde du monde intelligible la retient dans sa chute, mais de l’autre, le souvenir du monde sensible la fait se précipiter vers ce monde ci10. Cependant, si, une fois que l’âme a quitté le monde intelligible, ses souvenirs du monde sensible déterminent la profondeur de sa chute, c’est que, d’une façon ou d’une autre, elle possédait déjà ces souvenirs du monde sensible, ne fût-ce que dans un état de latence et de potentialité, pendant son séjour dans le monde intelligible. C’est, du moins, ce qui ressort d’une phrase plotinienne problématique (Ennéade IV, 4, 4, 14-15) : Ἀλλ’ εἰ ἀφισταμένη τοῦ ἐκεῖ τόπου ἀναφέρει τὰς μνήμας ὁπωσοῦν, εἶχε κἀκεῖ ἡ δύναμις· ἡ δὲ ἐνέργεια ἐκείνων ἠφάνιζε τὴν μνήμην11. Mais si, après avoir abandonné le lieu d’en haut, [l’âme] se rappelle ses souvenirs d’une façon quelconque, c’est que la puissance [les] possédait là-haut aussi, mais l’actualité en rendait invisible la mémoire. ontologiquement supérieur. Ailleurs dans la ThA (p. 72, 3-10 Badawi), cette même formule désigne la modalité cognitive selon laquelle on voit le monde non plus comme un ensemble de parties, mais comme un tout, et où les effets se perçoivent comme étant simultanés à leur cause. 8. Sur les explications multiples que proposèrent les platoniciens pour expliquer cette chute, voir par exemple A.-J. FESTUGIÈRE, La révélation d’Hermès Trismégiste. III: Les doctrines de l’âme, Paris 1953, pp. 63-96. 9. Cf. PLOTIN, Ennéades IV, 4, 3, 1-2 : Ἐξελθοῦσα δὲ ἐκεῖθεν καὶ οὐκ ἀνασχομένη τὸ ἕν, τὸ δὲ αὐτῆς ἀσπασαμένη καὶ ἕτερον ἐθελήσασα εἶναι κτλ. Passage qui, encore une fois, réclamerait l’attention des psychologues et des psychanalystes. 10. Cf. IV, 4, 3, 3-5 : Μνήμη δὲ ἡ μὲν τῶν ἐκεῖ ἔτι κατέχει μὴ πεσεῖν, ἡ δὲ τῶν ἐνταῦθα ὡδὶ φέρει. Il ressort de ce texte que l’âme, dépourvue de faculté de la mémoire pendant son séjour dans le monde intelligible (car là-haut, elle n’en avait point besoin, toutes choses lui étant présentes à la fois dans l’éternel présent qui caractérise l’Intelligible), la récupère en sortant de ce monde. C’est ce qui explique qu’elle dispose désormais de souvenirs aussi bien des splendeurs de ses expériences du monde intelligible que des péripéties qu’elle a vécues dans le monde sensible. 11. Texte de HS (in textu, mais ces éditeurs adoptent la conjecture de Goldwitzer ἢ δυνάμει dans leurs emendationes probandae, t. II, p. 389 : c’est cette conjecture qu’ont adopté les éditeurs Henry-Schwyzer, ed. minor ; Bréhier, Harder, Cilento, Armstrong, Radice ; ainsi que les traducteurs Brisson, Igal, Blumenthal/Dillon. D’autres éditeurs (Kirchoff, Müller, Volkmann) avaient proposé de biffer tout simplement les mots ἡ δύναμις.

26

MICHAEL CHASE

C’est, du moins, la traduction de ce texte qu’on doit retenir si l’on suit la leçon unanime des manuscrits. Or, cette leçon ἡ δύναμις est difficile. On ne voit trop, à première vue, de quelle δύναμις il s’agit. Il n’est pas sans intérêt à ce propos d’examiner les occurrences de ce terme qui précèdent, dans le traité IV, 4, ce texte du chapitre 4. (i) Au chapitre 1 du traité IV, 4, il était question d’une puissance (δύναμις) par laquelle l’âme regarde les choses intelligibles ; puissance qui est une, mais qui n’en est pas moins susceptible de devenir multiple lorsqu’elle se trouve dans autre chose12. (ii) À la ligne suivante, dans un contexte textuel qui risque d’avoir subi des perturbations13, il est question d’actes ou d’activités qui se trouvent ensemble dans une puissance qui reste dans la stabilité (δυνάμει ἑστώσῃ14). Enfin, (iii) au chapitre 2, il est affirmé que lorsque, dans le monde sensible, nous nous adonnons à l’activité intellectuelle, nous ne dirigeons pas notre activité vers nous-mêmes. Au contraire, à de tels moments, nous nous contentons de nous posséder nous-mêmes (ἀλλ’ ἔχει μὲν ἑαυτόν), tandis que notre activité est dirigée vers l’objet de notre contemplation15. En effet, dans de telles circonstances nous devenons cet objet, en nous offrant nous-mêmes à lui comme de la matière, et 12. Ἢ ἡ δύναμις ἡ μία οὕτως ἦν μία, ὡς πολλὰ ἐν ἄλλῳ IV, 4, 1, 33-34 : « C’est que cette puissance qui est une au sens où elle peut devenir multiple en autre chose » Br. ; « jenes eine (Denk-)Vermögen ist in der Weise eines, dass es in einem andern zur Vielheit wird » H/T/B ; « The one power is one in such a way that it becomes many in something else » A. ; « The answer is that the power, which is one, is one in such a way as to be many things when it is in something else » BD ; « esa potencia que es una » I. 13. Locus nondum sanatus selon Henry-Schywzer, suivi par Armstrong. 14. « Une puissance qui ne change pas » Br. ; « In a power which remains unchanged » A. ; « But they [sc. the activities] are always all there in a potency which is stable » BD ; « In dem Vermögen, das in sich steht » HTB ; « Una potencia que permanece en quietud » I. 15. Notons bien le contraste : lorsqu’on concentre son attention sur un objet, on se possède, soi-même – c’est-à-dire, de manière inconsciente –, mais on devient identique à l’objet externe sur lequel porte l’attention. En effet, selon la doctrine aristotélicienne et surtout son développement par Alexandre d’Aphrodise, dans l’acte d’intellection des choses immatérielles, l’intellect s’identifie à l’objet de son intellection. Voir, sur cette doctrine extraordinairement importante de l’identification du penseur et de l’objet de pensée, le grand livre, trop peu lu, de Ph. MERLAN, Monopsychism, Mysticism, Metaconsciousness. Problems of the Soul in the Neoaristotelian and Neoplatonic Tradition, La Haye 1963, ainsi que P. HADOT, « La conception plotinienne de l’identité entre l’intellect et son objet. Plotin et le De anima d’Aristote », dans : G. ROMEYER-DHERBY (éd.), Corps et Âme. Sur le De anima d’Aristote, Paris 1996, pp. 367-376 ; Stephen MENN, « Plotinus on the Identity of Knowledge with its Object », dans : Apeiron 34/3 (2001), pp. 233-246.

ACTE, PUISSANCE ET RÉVEIL

27

en nous laissant former selon l’objet de notre vision, n’étant alors nous-mêmes qu’en puissance (καὶ δυνάμει ὢν τότε αὐτός IV, 4, 2, 8). On voit bien que les trois occurrences du mot δύναμις qui précèdent notre texte tiré des Ennéades IV, 4, 4, présentent une ambigüité. Dans le cas de la première, il s’agit clairement d’une δύναμις (substantif au nominatif !) envisagée comme une faculté analogue à la vision, qui permet à l’âme, au cours de son séjour dans le monde intelligible, de regarder les Formes16. De même, à supposer que le texte soit en ordre, le sens du substantif δύναμις dans la deuxième de nos occurrences doit être, là encore, celui d’une puissance ou d’une faculté. Il en va différemment pour la troisième occurrence : dans la formule δυνάμει ὢν τότε αὐτός, il s’agit d’une utilisation du mot δύναμις dans son acception aristotélicienne de « puissance » au sens de « potentialité » : au datif, δυνάμει veut donc dire ici « in potentia », « en puissance ». Dans la partie de l’Ennéade IV, 4 qui précède notre texte du chapitre 4, donc, le sens du mot δύναμις oscille entre « puissance » au sens de « faculté », d’une part, et d’autre part « puissance » au sens de « potentialité ». La suite et fin du chap. 4 d’Ennéade IV, 4 est, par contre, claire à ce sujet : il y est bien question d’une puissance (δύναμις) au sens de faculté active. En effet, y lit-on, dans le monde intelligible, l’acte ou l’actualité (ἐνέργεια IV, 4, 4, 16) des réalités intelligibles effaçait la mémoire. Les souvenirs ne sont pas à envisager comme des empreintes, hypothèse qui mènerait à des conséquences absurdes. C’est à ce point qu’on se heurte à une autre phrase difficile (IV, 4, 4, 18) : ἀλλ’ ἡ δύναμις ἦν ἡ ἀφεθεῖσα ὕστερον εἰς ἐνέργειαν17. Mais c’était la puissance, celle qui, par la suite, fut déclenchée vers l’acte.

16. Ἀλλ’ὅταν εἰς ἓν βλέπῃ (...) Ἢ ἡ δύναμις ἡ μία οὕτως ἦν μία κτλ. 17. H.-S. ne signalent aucune variante dans les mss. Quelques traductions : « une puissance [intellectuelle] qui a passé ensuite à l’état d’acte » Bouillet ; « Mais [les souvenirs s’apparentaient à] une puissance qui plus tard allait passer à l’acte » Br. ; « but the potentiality was there which was later let loose into actuality » A ; « lo que había era la potencia que posteriormente se tradujo en acto » I ; « come potenza che più tardi doveva pasare all’ atto » Rad. ; « una potenza che in seguito, al momento del distacco, si sarebbe attualizzata » Reale ; « sondern die Potenz bestand weiter, und sie wurde später zur Aktualität freigelassen » HTB ; « But it is rather the case of a potentiality which was later subsumed into the actuality » BD. J’avoue ne point comprendre cette traduction de BD : que veut dire « to be subsumed into actuality » ?

28

MICHAEL CHASE

Quelle que soit la manière dont on traduit le mot ἀφεθεῖσα18, il s’agit clairement ici d’une sorte de puissance (= potentialité au sens aristotélicien) qui ne passera à l’acte que par la suite. Le chapitre 4 conclut en faisant remarquer que lorsque s’arrête l’activité (ἐνέργεια) qui caractérise le monde intelligible, l’âme voit ce qu’elle avait vu avant d’entrer dans ce monde-là, c’est-à dire, dans le monde intelligible19. C’est donc, comme nous l’avions vu, cette ἐνέργεια qui, dans le monde intelligible, mettait l’âme dans l’impossibilité de se souvenir de quoi que ce soit. Il me semble que la seule interprétation possible de ce texte est la suivante. En sortant du monde intelligible, l’âme récupère les souvenirs du monde sensible qu’elle avait perdus20, ou qui avaient été rendus imperceptibles, lorsqu’elle se trouvait dans le monde intelligible, car l’acte des réalités intelligibles les avait maintenus dans un état d’effacement, ou du moins, cet acte faisait en sorte que ces souvenirs restent imperceptibles. En effet, dans le monde intelligible l’âme ne s’adonne qu’à la contemplation intuitive, contemplation qui l’absorbe dans sa totalité, ne laissant de la sorte plus aucune place pour l’appréhension de soi, encore moins pour d’éventuels souvenirs du monde sensible. D’ailleurs, puisque le monde intelligible est caractérisé par l’absence du temps, tout y est présent ensemble à l’âme, de sorte qu’elle n’a aucun besoin de mémoire. Cependant, le fait qu’elle récupère ses souvenirs à la sortie du monde intelligible prouve qu’elle avait gardé ces souvenirs – dans un sens, et selon une modalité, qui restent à préciser – même pendant son séjour dans le monde intelligible. Dans ce monde supérieur, c’est donc la puissance (ἡ δύναμις) qui les garde, puissance ou faculté 18. Participe passif aoriste au féminin singulier du verbe ἀφίημι, « laisser aller, lancer, jeter, projeter, émettre, renvoyer » (Bailly). 19. πρὶν ἐκεῖ γενέσθαι IV, 4, 4, 20 : « ante quam illic esset » Ficin ; « avant de venir ici-bas » Br. ; « prima di entrare lassù » Radice ; « prima d’essere lassù » Reale ; « before it came to be there » BD ; « antes de estar allá » I ; « before it came to be in that world » A ; « was sie vor ihrem Eintritt in die obere Welt gesehen hatte » HTB. On voit que L. Brisson est le seul traducteur qui interprète le mot ἐκεῖ dans le sens « ici-bas ». Je ne puis partager cette interprétation. Il s’agit bien ici de la récupération des souvenirs du monde sensible, perdus ou rendus imperceptibles par l’acte intellectif, au cours du séjour de l’âme dans le monde intelligible, mais retrouvés dès que l’âme ressort du monde intelligible. 20. Interprétation qui se trouve confirmée sans ambages par la suite du texte de Plotin : voir Enn. IV, 4, 5, 11, 13: Ἡ τοίνυν μνήμη ἐκ τοῦ λόγου φαίνεται ἄρχεσθαι ἀπ’ οὐρανοῦ, ἤδη τῆς ψυχῆς τοὺς ἐκεῖ τόπου καταλειπούσης, avec le commentaire ad loc. de Theiler : « Die Erinnerung nicht im Jenseits, sondern vom Himmel an abwärts ».

ACTE, PUISSANCE ET RÉVEIL

29

qui, une fois qu’aura cessé l’activité du monde intelligible, lorsque l’âme sort du monde intelligible, va être réactualisée, permettant de la sorte à l’âme, résidant désormais dans le monde sensible, de voir ce qu’elle avait vu avant d’entrer dans le monde intelligible, c’est-à-dire, de récupérer les souvenirs de son dernier séjour terrestre. On voit donc que l’ensemble du texte des Ennéades IV, 4, 1-4 est caractérisé par une ambigüité fondamentale, largement passée sous silence par les traducteurs, qui se contentent le plus souvent de traduire le même mot grec de différentes manières : l’ambigüité entre la puissance (δύναμις), entendue d’une part au sens de faculté active, et d’autre part au sens de potentialité21. 3. De retour à la Théologie d’Aristote: l’Excursus sur la puissance Revenons à présent à l’interprétation de ces textes reflétée par la paraphrase de la Théologie d’Aristote, et que l’on peut lire (p. 37, 18sq. Badawi) après l’excursus que nous avons évoqué plus haut sur la 21. Lorsque Plotin parle ailleurs de la dunamis de l’âme, c’est normalement le sens de « faculté » qui domine. Dans le traité IV, 3, 15, Plotin décrit la descente des âmes. Celles-ci descendent d’abord dans le monde céleste, où elles assument un corps ; pour se déplacer ensuite dans le ciel, en assumant un corps terrestre, selon le degré dont elles se sont étendues (εἰς ὅσον ἂν εἰς μῆκος ἐκταθῶσι, IV, 3, 15, 3-4). Au cours de cette descente des corps célestes aux corps terrestres, certaines âmes passent d’un corps à un autre : ce sont celles dont la puissance (ἡ δύναμις IV, 3, 15, 5) n’était pas suffisante pour les faire remonter d’ici-bas, à cause de leur oubli et du poids qui les faisait descendre. Dans Ennéade V, 9, 1, Plotin classe les types d’âmes au moyen de l’allégorie des trois types d’oiseaux. La première classe a assumé une telle quantité de terre qu’elle en est alourdie et reste incapable de voler, bien qu’on lui ait accordé des ailes. La deuxième classe, poussée vers le plus beau par l’élément le plus fort de leur âme (κινοῦντος αὐτοὺ πρὸς τὸ κάλλιον (...) τοῦ τῆς ψυχῆς κρείττονος, allusion probable à la force motrice de l’Éros dans le Banquet de Platon) parvient à s’élever légèrement au-dessus de la terre, mais, incapables de voir ce qui est en haut, ces âmes sont ramenées en bas, où elles se consacrent à la pratique des vertus. La troisième classe est celle des hommes divins : grâce à leur puissance (δυνάμει) et à la qualité pénétrante de leur vue, de tels hommes sont capables de voir les splendeurs brillantes du monde intelligible. Ils peuvent donc s’élever jusqu’à ce monde, où ils restent, jouissant de cette place de vérité qui leur est propre, à l’instar d’un homme qui, après de longues errances, retourne dans sa patrie bien réglée. L’inspiration ultime de ces textes est bien sûr Platon, qui, dans l’introduction de son célèbre récit dans le Phèdre du mythe des chars des dieux et du voyage prénatal des âmes dans leur suite à travers le lieu supra-céleste, parle de la « puissance innée » (σύμφυτος δύναμις, 246a5) de chaque char et de chaque cocher, ainsi que de la « puissance de l’aile de l’âme » (πτεροῦ δύναμις), qui « tend par nature à entraîner vers le haut ce qui est lourd » (πέφυκεν (...) τὸ ἐμβριθὲς ἄγειν ἄνω, 246d5-6).

30

MICHAEL CHASE

supériorité de l’ignorance à la connaissance. Lorsque l’âme quitte ce monde pour entrer dans le monde intelligible, y lit-on, elle ne se souvient de rien de ce qui concerne le monde sensible. En effet, elle ne peut plus recevoir, pendant qu’elle se trouve dans le monde intelligible, les impressions (āthār) qui constituent le fondement de la perception dans le monde sensible. On ne saurait en définitif admettre que l’âme, dans le monde intelligible, reçoive des impressions dans son imagination ou sa faculté de représentation (arabe wahm = grec φαντασία), impressions auxquelles elle s’assimilerait, car cela impliquerait une identité entre l’état de l’âme dans le monde sensible et le monde intelligible22. Sur ce, l’Adaptateur interrompt sa paraphrase du traité IV, 4, pour s’embarquer (p. 38, 13-44 fin Badawi) dans une paraphrase d’Ennéade IV, 3, 19-20. Comme nous l’avons vu, il ne renouera sa paraphrase du traité IV, 4 qu’au huitième maymar. Avant cette paraphrase, cependant, nous trouvons un texte de transition constitué par une sorte d’excursus sur le statut de l’acte et de la puissance, d’une part dans le monde sensible, et de l’autre, dans le monde intelligible. Tournons-nous donc vers l’interprétation de cet excursus, destiné, comme j’essaierai de le montrer, à rendre compte de l’ambiguïté que nous venons de constater dans le texte de Plotin entre les différents sens du mot grec δύναμις = quwwa en arabe. 4. La doctrine des écorces (qushūr) Parmi toutes les doctrines que l’on rencontre dans le passage du maymar VIII, la plus originale est sans doute celle de la puissance et de l’acte, que nous trouvons dans un passage qui n’a pas de correspondant chez Plotin (p. 99, 9sqq. Badawi) et relève donc de la plume de l’Adaptateur. En effet, selon celui-ci, en ce qui concerne la puissance, l’on se trouve en présence d’un renversement de perspectives entre le monde intelligible et le monde sensible. Tandis que l’acte, l’actualité ou l’activité (fi῾l) est supérieur à la puissance (quwwa) dans 22. Sur l’implication mutuelle entre mémoire, imagination, et assimilation, voir par exemple ThA, p. 35, 16 Badawi. L’identification entre mémoire et imagination se trouve déjà chez PLOTIN (Enn. IV, 3, 29, 31 ; IV, 4, 3, 1), mais l’emphase sur la notion selon laquelle l’imagination implique que l’on s’assimile à ce qu’on imagine (ThA, p. 35, 14-16 ; 36, 3-6 Badawi) semble constituer un apport original de la part de l’Adaptateur.

ACTE, PUISSANCE ET RÉVEIL

31

ce monde sensible, dans le monde intelligible, c’est l’inverse : la puissance prime sur l’acte. En effet, la puissance qui caractérise le monde intelligible est parfaite (tāmmatun kāmilatun p. 99, 12 Badawi), à la différence de celle qui a besoin d’être amenée à l’acte par autre chose, c’est-à-dire dans le monde sensible. Il s’agit, dans le monde intelligible, d’une puissance au sens de « faculté », capable de percevoir les réalités spirituelles d’une manière analogue au fonctionnement de la faculté de la vision dans le monde sensible. L’infériorité de la puissance dans le monde sensible, par contre, consiste dans le fait qu’elle a besoin de sortir à l’acte (takhruju ilā l-fi῾l p. 99, 14 Badawi)23 pour percevoir les choses sensibles. Or, ce besoin de l’acte (fi῾l) qui caractérise le monde sensible est expliqué d’une manière qui, par rapport au texte de Plotin, est encore une fois d’une grande originalité. La cause en est que les choses sensibles se trouvent avoir été recouvertes d’enveloppes, de coquilles, ou d’écorces (qushūr, p. 99, 14 B.) qui en voilent la substance et les puissances, et que l’âme doit déchirer (kharq, p. 99, 15 B) pour pouvoir les percevoir. Point n’est besoin d’un tel acte de déchirement dans le monde intelligible, où les substances se trouvent dans un état de nudité ou d’abstraction (mujarrad, p. 99, 16 B) et de dévoilement (makshūf, ibid.). Cette notion des écorces, attestée ailleurs dans les versions courte et longue de la ThA24, est destinée à un long avenir. Elle se retrouve dans un autre ouvrage plus ou moins contemporain de la ThA : la Doxographie du Pseudo-Ammonios25. Comme l’a souligné Charles Genequand26, on la rencontre à plusieurs reprises dans les Muqābasāt d’al-Tawḥīdī ; mais elle trouve aussi sa place dans la version arabe des 23. Peut-être faut-il vocaliser tukhraja (3e personne singulier passif de la IVe forme) : elle a besoin qu’on la fasse sortir à l’acte ; cf. Enn. IV, 4, 4, 18 : ἡ δύναμις ἦν ἡ ἀφεθεῖσα ὕστερον εἰς ἐνέργειαν. 24. Version brève (= Vulgate) : cf. pp. 7, 1 ; 32, 3 Badawi. Version longue : Fenton 10 (= la translittération arabe de l’édition éclectique encore inédite, établie à partir de sept mss. arabes en caractères hébreux), fol. 13b, cité dans P. FENTON, « The Arabic and Hebrew Versions of the Theology of Aristotle », dans : J. KRAYE - W. F. RYAN - C. B. SCHMITT (éds.), Pseudo-Aristotle in the Middle Ages : The Theology and other Texts, Londres 1986, pp. 241264, en particulier p. 251. 25. Voir U. RUDOLPH, Die Doxographie des Pseudo-Ammonios. Ein Beitrag zur neuplatonischen Überlieferung im Islam, Stuttgart / Wiesbaden 1989, p. 156. 26. Ch. GENEQUAND, « La mémoire de l’âme. Porphyre et la Théologie d’Aristote », dans : Bulletin des Études Orientales 48 (1996), pp. 103-113, en particulier. p. 111, citant al-Tawḥidī, Muqābasāt, éd. ḤUSAYN, Baghdad 1970, pp. 130, 135, 144, 313.

32

MICHAEL CHASE

Parva Naturalia27, dans les Taṣawwurāt de l’auteur ismaélien Naṣīr al-Dīn Ṭusī28, et elle joue un rôle fondamental dans la pensée d’Ibn ῾Arabī29. Dans la philosophie juive, l’opposition qishr/lubb se retrouve chez Isaac Israeli30, Ibn Gabirol31, Ibn Ezra32, le penseur yéménite Ḥōter ben Shlomo (XVe s.)33, et le Zohar34. De manière intéressante, cette notion des écorces est attribuée à Empédocle par al-Shahrastānī : La doctrine d’Empédocle comporte un autre développement. Il dit : l’âme végétative est une écorce (qishr) pour l’âme bestiale et animale, l’âme animale est une écorce pour l’âme rationnelle (manṭiqiyya), l’âme rationnelle est une écorce pour l’âme intellective (῾aqliyya) ; tout ce qui est plus bas est une écorce pour ce qui est plus haut et ce qui est plus haut en est la pulpe (lubb). Parfois, il emploie au sens d’écorce et de pulpe les mots corps et esprit, de sorte qu’il fait de l’âme végétative un corps pour l’âme animale qui en est l’esprit, et ainsi de suite jusqu’à l’intellect35.

Franz Altheim s’était servi de ce témoignage pour appuyer sa thèse, assez vraisemblable, selon laquelle ce compte rendu sur la doctrine 27. Ms Rampur 1752 fol. 15b, cité dans R. HANSBERGER, « Representation of which reality ? ‘Spiritual forms’ and ‘ma῾ānī’ in the Arabic adaptation of Aristotle’s Parva Naturalia », dans : B. BYDÉN - F. RADOVIC (eds.), The Parva Naturalia in Greek, Arabic and Latin Aristotelianism. Supplementing the Science of the Soul, Cham 2018, pp. 99-122, en particulier p. 106. 28. H. CORBIN, Temps cyclique et gnose ismaélienne, Paris 1982, p. 66. 29. Pour Ibn Arabī, qui intitule l’un de ses traités Al-qishr wa-l-lubb, le macrocosme est l’écorce (qishr), dont la moelle (lubb) est le microcosme : voir G. T. ELMORE, Islamic Sainthood in the Fullness of Time : Ibn al-῾Arabī’s Book of the Fabulous Gryphon, Leiden 1999, p. 244. 30. Livre des éléments, p. 56, cité par A. ALTMANN - S. STERN, Isaac Israeli : A Neoplatonic Philosopher of the Early Tenth Century, Oxford 1958, p. 184. 31. Voir S. PESSIN, Ibn Gabirol’s Theology of Desire : Matter and Method in Jewish Medieval Neoplatonism, Cambridge, 2013, p. 49 et saepe. 32. Book of discussions 77b, cité par M. Z. COHEN, « Words of Eloquence : Rhetoric and Poetics in Jewish Bible Exegesis in its Muslim and Christian Contexts », dans : M. Z. COHEN - A. BERLIN (eds.), Interpreting Scriptures in Judaism, Christianity and Islam : Overlapping Inquiries, Cambridge 2016, p. 271. 33. Cité dans Y. T. LANGERMANN, « The Debate Between the Philosopher and the Mutakallim », dans : Proceedings of the American Academy for Jewish Research (1994), pp. 189-240, en particulier § 82, p. 224. 34. Zohar, I, 19b-20a, cité dans S. KARPPE, Étude sur les origines et la nature du Zohar, Paris 1901, pp. 361-362. 35. AL-SHAHRASTĀNĪ, Kitāb al-milal wa-l-niḥal p. 828 éd. BADRĀN = t. II, p. 196 JOLIVET-MONNOT.

ACTE, PUISSANCE ET RÉVEIL

33

d’Empédocle proviendrait de l’Histoire philosophique de Porphyre36. En effet, on peut comparer à ce développement tiré d’al-Shahrastānī un passage de l’Ad Marcellam de Porphyre, où on lit que « l’Intellect (...) tire une nourriture pour l’âme, qui est comme pour lui un corps. En effet, il faut tenir que le corps de l’intellect est l’âme raisonnable... »37. 5. L’interprétation arabe d’Ennéade IV, 4, 5 : pas pire que les autres ? Revenons à notre passage du maymar VIII de la ThA, qui reprend le début d’Ennéade IV, 4, 5, texte qui est d’une interprétation très délicate38. L’Adaptateur donne à présent une autre explication du fait que l’âme a besoin de l’acte ou de l’actualité (fi῾l) pour exercer la perception dans le monde sensible, tandis qu’elle n’en a point besoin dans le monde intelligible : c’est que, dans le monde intelligible, tout est simple (basīṭ), de sorte que l’âme se perçoit elle-même et les choses simples par sa puissance simple (ThA, p. 100, 3sqq. Badawi). Dans le monde sensible, comme nous venons de le voir, l’âme se voit recouverte d’une multiplicité d’écorces39, ce qui fait qu’elle a besoin d’un effort (ta῾ab) pour percevoir les réalités intelligibles. Or cet effort est un acte (wa-l-ta῾ab fi῾lun, p. 100, 5 Badawi), et l’acte est un composé (murakkab) ; un tel acte est donc incapable de percevoir à fond les réalités intelligibles. Il s’ensuit que l’âme ne perçoit les réalités intelligibles pendant qu’elle se trouve au monde sensible, que par un acte qu’elle acquiert ici-bas et non pas par sa puissance (p. 100, 6-7 B). Si l’âme a du mal, dans le monde sensible, à se souvenir de ce qu’elle avait vu dans le monde intelligible, c’est que l’acte s’y empare (yastaghriqu) de l’ensemble de sa puissance40, l’empêchant de la sorte de percevoir ce qu’elle avait perçu. On voit que l’ambiguïté des termes « puissance » (ar. quwwa) et « acte » (ar. fi῾l) chez Plotin ont mené l’Adaptateur à donner des explications diverses de la difficulté que rencontre l’âme lorsqu’il s’agit de 36. F. ALTHEIM - R. STIEL, Porphyrios und Empedocles, Tübingen 1954, pp. 28-29. 37. Texte cité par A-Ph. SEGONDS, « Les fragments de l’Histoire de la philosophie », dans son Appendice à Porphyre, Vie de Pythagore, Lettre à Marcella, texte établi et traduit par É. DES PLACES, Paris 1982, p. 176. 38. On en voudra pour preuve les versions extrêmement diverses que les traducteurs modernes ont proposé des seules lignes IV, 4, 5, 2-3 ; voir infra. 39. ThA, p. 100, 4-5 B : li-kathrati l-qushūr allatī labisathā. 40. Cf. Enn. IV, 4, 4, 15 : ἡ δὲ ἐνέργεια ἐκείνων ἠφάνιζε τὴν μνήμην.

34

MICHAEL CHASE

percevoir les réalités intelligibles pendant son séjour dans le monde sensible. D’un côté, l’âme a besoin d’un acte (fi῾l au sens d’action, d’activité) pour déchirer les écorces que recouvrent ici-bas aussi bien l’âme elle-même que les choses sensibles ; acte qui, étant un composé, n’est pas capable d’une perception exacte de ces réalités intelligibles. De l’autre côté, c’est le fait que dans le monde sensible l’acte (fi῾l au sens d’actualité) submerge la puissance (quwwa) qui explique cette même difficulté. C’est donc à cause des écorces (qushūr) – aussi bien celles qui voilent les choses sensibles que celles qui recouvrent l’âme elle-même – que l’âme dans le monde sensible a besoin de l’acte, tant pour déchirer les écorces des choses sensibles dans le but de pénétrer jusqu’à leur substance, que pour pénétrer les écorces qui recouvrent l’âme elle-même, l’empêchant de la sorte de percevoir les réalités intelligibles. Nous avons affaire, semble-t-il, à un autre développement originel par rapport au texte plotinien. En effet, selon Plotin, c’est dans le monde intelligible que l’acte (ou plutôt l’actualité, gr. ἐνέργεια) explique que la puissance (gr. δύναμις) de l’âme humaine y soit mise en échec, en rendant inactive, ne fût-ce que de manière temporaire, la faculté de mémoire de celle-ci. Contre la thèse selon laquelle la puissance est supérieure à l’acte, du moins dans le monde intelligible, l’interlocuteur fictif mis en scène par l’Adaptateur pose une série d’apories (ThA, p. 100, 9sqq. B.). Lorsqu’on commence par percevoir quelque chose par la puissance (bi-l-quwwati)41, dit-il, pour passer ensuite à la percevoir par l’acte (bi-l-fi῾li, ibid.), la perception en acte représente quelque chose de plus stable et de plus forte, car l’acte est l’achèvement de la puissance (tamām p. 100, 10)42. En réponse à cette objection, on distingue deux cas de figure. Il est vrai que lorsque la perception s’effectue par la réception d’une impression (athar) – c’est le cas qui correspond à la perception dans le monde sensible –, la puissance, envisagée ici comme faculté psychique, commence par faire un dessin (rasm) de l’impression (athar) de la chose43. 41. Un « peer-reviewer » anonyme, que je remercie chaleureusement, soutient que dans ce contexte bi-l-quwwati signifie non pas « par la puissance » mais « en puissance ». C’est possible, mais tout compte fait, je maintiens mon interprétation. 42. Je lis li-anna l-fi‘l innamā huwa tamāmun al-quwwati avec le ms. Ṣ. 43. Je lis ka-annahā taf῾alu rasma athari l-shay’ avec ms. K. L’on comparera la traduction de Lewis : « For the potentiality will be as though acting and perfecting the impression of the thing. Actuality makes that impression perfect... ».

ACTE, PUISSANCE ET RÉVEIL

35

Ensuite, l’acte perfectionne (atamma) cette impression ; c’est dans ce sens que l’acte perfectionne la puissance (mutammim al-quwwati)44. Cependant, si la perception s’effectue sans réception d’une impression – cas qui correspond, nous l’avons vu, à la situation de l’âme pendant son séjour dans le monde intelligible –, alors la seule puissance suffit pour que la perception ait lieu. Ce texte appelle plusieurs remarques. Tout d’abord, nous y retrouvons l’ambiguïté entre divers sens des termes « acte » (fi῾l) et « puissance » (quwwa) que nous avions constatée chez Plotin. Dans sa remarque, l’interlocuteur utilise les formules adraka l-shay’a bi-lquwwati et adraka l-shay’a bi-l-fi῾li au sens, respectivement, de « percevoir une chose potentiellement » — l’équivalent grec serait δυνάμει – et « percevoir une chose en acte » – en grec, ἐνεργείᾳ. Selon l’Adaptateur, c’est dans ce dernier sens que, selon la doctrine philosophique aristotélicienne, l’acte (ἐνέργεια) est effectivement l’achèvement ou la perfection de la puissance (δύναμις)45 ; ce qui n’empêche pas notre auteur, dans sa réponse à cette objection, de revenir au sens de quwwa qui correspond à notre notion de « faculté ». Quoi qu’il en soit, nous trouvons dans cette réponse46 l’esquisse d’une doctrine psychologique assez intéressante. Il y a deux sortes de perception : celle qui se fait par réception d’une trace, d’une affection, ou d’une impression sensible (athar, 44. Nous trouvons une doctrine très semblable chez BOÈCE (In perihermeneias, 2e éd., pp. 28-29 Meiser) : « Les sens et l’imagination sont des sortes de premières figures, sur lesquelles, comme sur une espèce d’arrière-fond, l’intelligence survient en y prenant appui. En effet, comme les peintres ont coutume de tracer des corps par des lignes, et de les poser comme substrat là où ils veulent représenter l’apparence de quelqu’un, de la même façon les sens et l’imagination sont posés naturellement comme substrats dans la perception de l’âme. Car lorsqu’une chose tombe sous les sens ou sous la pensée, il est nécessaire qu’en naisse d’abord une sorte d’imagination et que par la suite survienne une compréhension plus complète, qui développe toutes celles de ses parties qui se trouvaient comprises dans l’imagination à l’état de confusion ». Comme on a essayé de montrer dans le cas de Marius Victorinus et d’Avicenne, lorsque nous avons affaire à de tels parallèles entre la littérature philosophique latine de l’Antiquité tardive, d’une part, et de l’autre des textes faisant partie des Neoplatonica arabica – toute possibilité d’une lecture directe des textes latins par des auteurs arabophones étant exclue – il faut songer à une source commune, qu’on a proposé de trouver dans la personne de Porphyre. Voir M. CHASE, « Essence and Existence in Marius Victorinus and in Avicenna », à paraître dans St. COOPER – V. NĚMEC (éds.), Marius Victorinus : Pagan Rhetor, (Neo-)Platonist Philosopher, and Christian Theologian, Atlanta. 45. Voir, par exemple, P. LETTINCK, « Aristotle’s Physical Works and the Arabic Tradition”, dans : A. ALWISHAH - J. HAYES (éds.), Aristotle and the Arabic Tradition, Cambridge 2015, pp. 105-120. 46. Du moins en apparence, car l’état du texte n’est pas sûr.

36

MICHAEL CHASE

terme qui correspond souvent dans la ThA au grec πάθος), et celle qui ne passe pas par une telle réception. Dans le premier cas, c’est la puissance (quwwa au sens de « faculté ») qui produit une sorte de trace ou d’« esquisse » (rasm) de l’impression sensible, esquisse qui est parachevée ou menée à la perfection (atamma) par l’acte (fi῾l). Donc, lorsque, comme c’est le cas dans le monde sensible, la perception se fait par la réception d’une impression, l’objecteur a raison : ici, c’est effectivement l’acte qui parachève la puissance. Cependant, lorsque, comme dans le monde intelligible, il n’y a aucune réception d’une impression, la puissance est autosuffisante : c’est-à-dire qu’elle est capable d’effectuer la perception par elle-même47. Cette distinction entre deux types de perception représente, semblet-il, une tentative hautement originale pour comprendre le début d’Ennéade IV, 4, 5, texte qui continue à poser de gros problèmes aux traducteurs et aux commentateurs modernes48. Plotin y commence par poser la question suivante à propos de la puissance (δύναμις), dont la nature précise dans le texte des Ennéades IV, 4-14, nous l’avons vu, n’est pas facile à préciser. Cette puissance, demande Plotin à présent, qui est responsable de la mémoire, est-ce la même puissance qui, maintenant (c’est-à-dire, dans le monde sensible), mène ces choses-là (c’est-à-dire, vraisemblablement, nos souvenirs du monde intelligible), à l’actualité ?49 La réponse plotinienne est des plus énigmatiques: ἢ εἰ μὲν μὴ αὐτὰ ἑωρῶμεν, μνήμῃ, εἰ δ᾿ αὐτά, ᾧ κἀκεῖ ἑωρῶμεν. 47. Il n’est peut-être pas sans intérêt de noter que ce contraste entre la perception sensible, qui passe par la réception d’une impression, et l’intuition intellectuelle, libre de toute réception d’impressions, est caractéristique de la pensée d’Évagre du Pont. Voir par exemple ÉVAGRE, Chapitres gnostiques I, 37, avec le commentaire de I. L. E. RAMELLI dans : Evagrius’ Kephalaia Gnostika. A new translation of the unreformed text from the Syriac, Atlanta 2015, p. 33. Sur les ressemblances entre la doctrine psychologique d’Évagre et celle de Porphyre, voir par exemple, A. PIRTEA, « The Origin of Passions in Neoplatonic and Early Christian Thought : Porphyry of Tyre and Evagrius Ponticus », dans : P. G. PAVLOS – L.F. JANBY – E. K. EMILSSON – T.T. TOLLEFSEN (éds.), Platonism and Christian Thought in Late Antiquity, Londres 2019, pp. 258-274 (je remercie l’auteur de m’avoir communiqué cette importante étude). 48. On se contentera ici d’un simple échantillon. Pour rendre la seule particule grecque ἢ dans ce texte, les traducteurs ont proposé : « Según » (I), « No » (Rad.), « Certamente » (Reale), « Non » (Br.), « Nun » (H/T/B), « Well » (B/D.) La meilleure solution est sans doute celle d’Armstrong, qui ne traduit tout simplement pas la particule en question. Nous avons choisi de la traduire « Eh bien ». 49. PLOTIN, Ennéades V, 4, 1-2 : Τί οὖν; κἀκεῖνα νῦν αὐτὴ ἡ δύναμις, καθ᾿ ἣν τὸ μνημονεύειν, εἰς ἐνέργειαν ἄγει;

ACTE, PUISSANCE ET RÉVEIL

37

Eh bien, si nous ne les avons pas vues elles-mêmes, c’est par la mémoire ; si elles-mêmes, c’est grâce à la même chose par laquelle nous avions vu là-haut.

L’interlocuteur revient maintenant à la charge : si la puissance ne perçoit les choses intelligibles que par un acte, affirme-t-il, et que l’acte détruit la puissance (al-fi῾lu mufsidun li-l-quwwati, p. 100, 16-17), alors la puissance de l’âme grâce à laquelle elle avait perçu correctement les choses intelligibles, a péri. On voit bien l’ambigüité qui est à l’arrière-plan de cette objection, ambigüité qui relève de la confusion entre le sens des termes techniques de « puissance » (quwwa) et d’« acte » (fi῾l). Dans cette objection, l’interlocuteur envisage la puissance en même temps d’une part comme la faculté, analogue à la vision dans le monde sensible, par laquelle l’âme perçoit les formes lorsqu’elle se trouve dans le monde intelligible ; et d’autre part comme la potentialité au sens aristotélicien. L’acte, pour sa part, est conçu aussi bien comme action – nécessairement composée, à l’instar de l’effort que doit mobiliser l’âme pour exercer la perception dans le monde sensible – et comme actualité, qui, selon une certaine interprétation de la doctrine aristotélicienne50, vient non seulement compléter la puissance, mais aussi la détruire. On répond à cette objection (ThA VIII, p. 100, 17sq. B) en affirmant que, malgré les apparences, la puissance de l’âme n’est pas détruite lorsque l’acte rentre en elle, mais elle est tout simplement dispersée51. La preuve, c’est le fait que lorsque l’âme – vraisemblablement au cours de son existence terrestre – s’occupe des choses intelligibles, elle n’a plus recours à l’acte, ni n’a besoin de la pensée discursive (ar. tafakkur = grec dianoia) pour percevoir les intelligibles. Dans ces 50. En fait, pour Aristote, l’activité/actualité (ἐνέργεια) complète et met fin aux mouvements (κινήσεις) : c’est ainsi que lorsque je marche et que j’arrive à ma destination, mon activité de marche s’arrête et se voit donc détruite. Cependant, il existe un autre type d’acte, que le Stagirite qualifie précisément d’ ἐνέργεια (activité/actualité), qui, elle, atteint son but à chaque instant et n’est pas détruite en ce faisant : que l’on pense aux actes que sont voir, penser, vivre bien, être heureux. Voir, par exemple, Métaphysique Theta 6, 1048b18-36, avec la discussion dans M. CHASE, « Discussions on the Eternity of the world in Late Antiquity », dans : ΣΧΟΛΗ, A Journal of the Centre for Ancient Philosophy and the Classical Tradition (Novosibirsk), 5/2 (2011), pp. 111-173. 51. Je suis ici le texte de Badawi : lam tafsid al-quwwatu, lākinnahā tubḥathu ῾an al-nafs ῾inda dukhūl al-fi῾l ῾alayhā faqaṭ. La traduction de Lewis est légèrement différente : « Potentiality is not destroyed but is merely quiescent in the soul when actuality enters her ».

38

MICHAEL CHASE

conditions, la puissance – ici envisagée à nouveau comme une faculté qui équivaut dans le monde intelligible à la vision dans le monde sensible – lui revient52. Cependant, plutôt que de dire que l’âme acquiert une nouvelle puissance ou faculté, il faut comprendre que la même puissance se réveille (nahaḍat), car elle n’avait point abandonné l’âme53. C’est grâce à ce réveil de la puissance que l’âme est capable de voir les formes intelligibles qu’elle avait contemplées avant de descendre dans le monde sensible. La puissance stable (al-quwwa al-thābita), quant à elle, se trouve dans les substances qui s’appliquent aux choses par une intuition correcte54, sans délibération ni pensée discursive, en les observant de leurs propres yeux. La référence est sans doute au mode de pensée de l’Intellect et de l’âme qui s’y assimile dans le monde intelligible. On retiendra de ce texte assez difficile deux éléments principaux. Tout d’abord, il y a la distinction entre deux modes de pensée : d’une part, la pensée discursive (al-rawiyya wa-l-tafakkur ; double traduction d’un terme grec, par exemple διάνοια ?), qui caractérise l’activité cognitive de l’âme lorsqu’elle se trouve dans le monde sensible ; de l’autre, ce qu’on pourrait désigner comme la pensée intuitive (taqa῾u [...] wuqū῾an ṣaḥīḥan, p. 101, 3, mots qui pourraient rendre les termes grecs ἐπιβάλλειν/ ἐπιβολή), mode de pensée de l’âme pendant qu’elle se trouve dans le monde intelligible, qui se caractérise précisément par l’absence de toute pensée discursive (bi-ghayri rawiyyati wa lā fikr, p. 101, 4 B). Il s’agit d’un thème bien attesté chez Plotin et dans tout le néoplatonisme post-plotinien55. 52. Raja῾at tilka l-quwwa ilayhā, ThA, p. 100, 19 B. Cela correspond au texte de PLOTIN, Enn. IV, 4, 4, 19-20 : Παυσαμένης οὖν τῆς ἐν τῷ νοητῷ ἐνεργείας, εἶδεν ἃ πρότερον ἡ ψυχή, πρὶν ἐκεῖ γενέσθαι, ἰδοῦσα ἦν. Cependant, la divergence n’en est que plus frappante : pour Plotin, c’est dans le monde intelligible que l’acte (ἐνέργεια) cache la puissance/ faculté de mémoire de l’âme ; pour l’Adaptateur, c’est lorsque l’âme, dans le monde sensible, cesse d’avoir recours à l’acte en ce qui concerne les intelligibles – c’est-à-dire, lorsqu’elle ne se sert plus de la pensée discursive (tafakkur p. 100, 19 B.) – que cette puissance – celle qui lui permet de percevoir les intelligibles – lui revient. 53. li-annahā lam tufāriq al-nafsa, ThA, p. 100, 19 B. Il s’agit de la version néoplatonicienne de la doctrine platonicienne de l’anamnèse. 54. Je lis fī al-jawāhir allatī taqa῾u fī al-ashyā᾿ wuqū῾an ṣaḥīḥan avec Dieterici. Cette interprétation de la puissance grâce à laquelle l’âme perçoit les intelligibles comme « intuition » (wuqū‘), correspond à l’interprétation du texte plotinien tant de J. Igal (p. 380, n. 26 de la traduction chez Gredos) que de L. Brisson (p. 247 n. 40 de la traduction chez Flammarion). 55. Voir chez PLOTIN, par exemple, Ennéades V 3 [49] 17, 24 ; VI 6 [34] 4, 16 ; III 8 [30] 9, 21-22.

ACTE, PUISSANCE ET RÉVEIL

39

Le deuxième thème important est la doctrine selon laquelle l’âme possède, même lorsqu’elle est venue s’incarner dans un corps humain, une faculté (ar. quwwa = grec δύναμις) qui lui permet d’intelliger les Formes du monde intelligible. Même si cette faculté ou cette puissance s’est endormie lors de l’entrée de l’âme dans le monde sensible – ou plutôt s’est vue étouffer par l’acte or l’actualité (ar. fi῾l = grec ἐνέργεια) qui y règne – elle n’abandonne jamais complètement l’âme (lā tufāriqu, p. 101, 11), mais elle est toujours susceptible d’être réveillée (nahaḍat, p. 100, 19), permettant ainsi à l’âme de percevoir à nouveau, sur terre, ce qu’elle avait vu dans le monde intelligible. Il s’agit là, bien entendu, du thème de la réminiscence platonicienne, tel qu’attesté dans le Ménon, mais aussi dans le Phédon et le Théétète, et qui fait l’objet du chapitre 5 de l’Ennéade IV, 4 de Plotin. Sur ce, l’Interlocuteur revient à la charge (ThA VIII, p 101, 5sqq. Badawi) avec une autre objection, en posant la question de savoir comment l’âme peut connaître et se souvenir des Formes intelligibles pendant qu’elle se trouve dans le monde sensible. On présente cette objection sous forme de dilemme : si l’âme, au cours de son existence terrestre, connaît (ta῾lamu) ou perçoit (tudriku) les Formes, c’est qu’elle le fait soit (a) par la même puissance (quwwa) par laquelle elle connaissait et percevait ces Formes dans le monde intelligible, soit (b) par un acte (fi῾l) autre que cette puissance. Or, l’alternative (a) est exclue, car on ne peut admettre que l’âme jouisse de la même perception des Formes dans l’état de nudité et de pureté qui est le sien pendant son séjour dans le monde intelligible, et lorsqu’elle se trouve contaminée par le corps pendant son séjour terrestre. Mais l’alternative (b) n’est pas non plus tenable : en effet, si c’est par un acte, plutôt que par une puissance, que l’âme perçoit dès ici-bas les réalités intelligibles, elle le ferait sans passer par une puissance perceptive (quwwa darrāka) ; or, c’est impossible, car tout ce qui perçoit ne le fait que par la puissance perceptive qui lui est connaturelle (l’adjectif gharīzī rend ailleurs dans la ThA le grec σύμφυτος), puissance qui n’abandonne l’agent perceptif qu’au moment de sa mort. La conclusion est pourtant claire : l’âme, selon l’Interlocuteur, ne saurait ni connaître ni percevoir les réalités intelligibles pendant qu’elle se trouve dans le monde sensible. La réponse à cette objection consiste dans une sorte de solution de compromis : d’une part, affirme-t-on, c’est bien par la même puissance que l’âme contemple les Formes intelligibles lorsqu’elle se trouve dans

40

MICHAEL CHASE

le monde intelligible et dans le monde sensible56. Pourtant, la descente de l’âme dans le monde sensible l’oblige à avoir recours à « autre chose » (shay’ ākhar) pour contempler ces Formes. C’est donc la puissance (quwwa) elle-même, régnant sur le monde intelligible, qui, en réponse au besoin de l’âme, manifeste et rend actif l’acte57. On a là encore un exemple du renversement des rapports entre puissance et acte dans le monde intelligible et le monde sensible. Tandis que dans le monde sensible, c’est l’acte qui – selon la bonne doctrine péripatéticienne – complète la puissance et la mène à sa fin, dans le monde intelligible, c’est la puissance qui manifeste et complète l’acte. Face à l’impuissance de l’âme pendant son séjour dans le monde sensible, qui, à cause de son état déchu, se trouve dans l’incapacité de percevoir le monde intelligible, la puissance mobilise l’acte, en le manifestant et en le menant à l’actualité, acte qui va permettre à l’âme de renouer connaissance avec les Formes intelligibles, même pendant son existence terrestre. C’est donc effectivement, poursuit l’Adaptateur (p. 102, 1sq. B), grâce à une seule et même puissance que l’âme voit les Formes intelligibles, aussi bien dans le monde intelligible que dans le monde sensible. En effet, l’acte auquel l’âme a recours pour percevoir les formes intelligibles pendant son séjour dans le monde sensible n’est que le redressement ou l’éveil (nuhūḍ) de cette puissance dont elle se servait dans le monde intelligible ; mais si elle a besoin d’être réveillée, c’est que la perception des Formes lui est beaucoup plus difficile dans le monde sensible. Nous avons vu que la cause de cette difficulté, ce sont les écorces (qushūr) obscurcissantes qui enveloppent non seulement les choses dans le monde sensible, mais aussi l’âme pendant son séjour terrestre. Arrivé à ce point de son exposé (p. 102, 3sqq. B.), l’Adaptateur ajoute une phrase qui est, me semble-t-il, d’une importance capitale : cette puissance, qui permet la contemplation des Formes depuis le monde sensible, ne se réveille pas chez tout le monde, mais uniquement chez l’élite du peuple (fī khawāṣṣ al-nās) et celui qui appartient aux gens du bonheur (man kāna min ahl al-sa῾āda)58. Avec la doctrine 56. Cf. Enn. IV, 4, 5, 1-3, cité supra. 57. fa-aẓharat al-quwwatu l fi῾la wa-sayyarathu ῾ammālan ThA, p. 101, 14 B. 58. L’on comparera la doctrine plotinienne des trois genres d’homme, déjà évoqué supra n. 21.

ACTE, PUISSANCE ET RÉVEIL

41

d’une élite qui seule, grâce à sa capacité de contemplation, a le privilège d’avoir accès aux Formes, on a déjà le fondement de toute une philosophie politique, telle qu’on la retrouvera chez un al-Fārābī, par exemple. Nous avons affaire, ici, à un autre cas où l’Adaptateur se révèle comme un excellent lecteur de Plotin. Comme nous l’avons vu, le début du ch. 5 du traité IV, 4, 28 est extrêmement difficile, même s’agissant de Plotin. On commence par poser la question de savoir si c’est par la même puissance (δύναμις = quwwa) par laquelle s’effectue la mémoire que les intelligibles sont amenés à l’acte (ἐνέργεια = fi῾l). Nous avons pu constater que la réponse de Plotin, telle que transmise par les manuscrits grecs qui nous sont parvenus, n’est pas facile à déchiffrer. Vient ensuite une autre phrase clé (Enn. IV, 4, 5, 3-4) : Ἐγείρεται γὰρ τοῦτο οἷς ἐγείρεται. En effet, elle se réveille chez ceux chez qui elle s’éveille.

Ce qui pose problème, ici, c’est surtout le pronom relatif οἷς de la ligne 459. La majorité des traducteurs modernes a compris ce mot comme un datif instrumental : c’est ainsi que L. Brisson traduit cette phrase « Cette faculté est éveillée par les choses qui l’éveillent »60. Le problème posé par cette interprétation, parfaitement acceptable sur le plan grammatical, c’est qu’elle attribue à Plotin une affirmation tautologique. Bien sûr que cette faculté est réveillée par ce qui la réveille – par quoi d’autre pourrait-elle être réveillée ? – mais pourquoi Plotin aurait-il cru utile de le dire ? Du point de vue du sens, l’interprétation de Harder, qui comprend le pronom relatif au datif pluriel οἷς comme datif d’intérêt, et comprend l’antécédent de ce pronom non pas comme un neutre pluriel, mais comme un masculin pluriel, semble préférable. Selon cette interprétation, il faudrait traduire : « cette puissance s’éveille chez ceux chez qui elle s’éveille »61 ; autrement dit, cette 59. Leçon de la plupart des mss. ; B. lit αἷς. 60. Cf. Armstrong : « this is awakened by that which awakens it » ; BD : « this is aroused by the things by which it is aroused » ; Reale : « Quest’ultima è risvegliata dagli oggetti che la eccitano ». 61. « Denn dieses erwacht, bei denen es aufwacht » HTB. Même interprétation chez Igal : « pues ésa es la potencia que se despierta en quienes se despierta ». La traduction de Radice : « Questo potere si ridesta contemporaneamente agi oggetti che lo destano », semble présupposer la conjecture de Harder εγείρεται à la ligne trois, même si le texte grec reproduit par Radice n’accueille pas cette conjecture.

42

MICHAEL CHASE

puissance s’éveille chez certains ; cependant, chez d’autres, elle ne s’éveille point. C’est, je crois, l’interprétation de cette ligne, ou plutôt d’un seul mot, le pronom relatif οἷς, qui a inspiré à l’Adaptateur sa doctrine — si lourde de conséquences, car elle implique un élitisme prononcé — selon laquelle le réveil de la dunamis est réservé à l’élite du peuple, qu’il qualifie de « gens du bonheur ». 6. Conclusion Faisons le bilan de notre étude de cet excursus ou passage de transition qui figure au VIIIe mīmar de la ThA. Face à l’ambiguïté, pour ne pas dire la confusion, qui caractérise l’utilisation par Plotin du terme technique δύναμις au cours des cinq premiers chapitres du traité IV, 4 (28) des Ennéades, l’Adaptateur développe, dans son excursus, la doctrine très originale du renversement des rôles de la puissance et de l’acte dans le monde intelligible et le monde sensible. Tandis que c’est l’acte qui domine le monde sensible, c’est la puissance qui règne dans le monde intelligible. Pour appuyer sa thèse, l’Adaptateur a recours à la doctrine des écorces (qushūr), doctrine qui n’apparaît pas chez Plotin mais qui a pu être présente chez son disciple Porphyre. Enfin, une lecture très fine du texte difficile de Plotin semble avoir donné naissance à l’idée selon laquelle la faculté de percevoir les réalités depuis le monde sensible est réservée à une élite qualifiée de « gens du bonheur ». Il est très difficile de tirer de ce texte des conclusions définitives en ce qui concerne la question de l’éventuelle identité de l’Adaptateur. Fin lecteur de Plotin, l’Adaptateur semble parfois un peu embarrassé par l’ambigüité de l’usage plotinien des termes de dunamis et d’energeia ; mais c’est une confusion qui est également présente – nous en avons été les témoins au cours de ce travail – chez les exégètes modernes. L’Adaptateur a fait un vaillant effort pour expliquer un texte plotinien qui présente de très graves difficultés de compréhension. Tout compte fait, puisque la notion des qushūr, qui joue un rôle si important dans notre texte, apparaît déjà dans une partie de la ThA qui provient très probablement d’al-Kindī, et paraît dans la Doxographie du PseudoAmmonios, dont l’origine pourrait remonter au Cercle d’al-Kindī, on pourrait suggérer, avec prudence, que notre texte de l’excursus provient de cette même source. Ce serait, dans cette hypothèse, al-Kindī luimême qui, en tant qu’« éditeur » de la ThA, aurait rédigé les passages

ACTE, PUISSANCE ET RÉVEIL

43

de transition qui relient ensemble les différents extraits des Ennéades qui constituent la Théologie d’Aristote62. Cependant, il me semble probable qu’al-Kindī, si c’est bien lui qui a rédigé notre texte, se sera appuyé pour ce faire sur des matériaux doctrinaux d’origine néoplatonicienne, par exemple les hypomnêmata aux Ennéades de Porphyre. Enfin, c’est probablement chez Porphyre que l’Adaptateur aura trouvé la doctrine des qushūr, mais il est difficile, dans l’état actuel de nos connaissances, de dire si c’est Porphyre qui est à l’origine de la doctrine, également originale, de la puissance et de l’acte que nous trouvons dans notre excursus, ou si c’est al-Kindī qui l’aura élaborée à partir de ses propres connaissances de la philosophie aristotélicienne et surtout néoplatonicienne. Quoi qu’il en soit, notre texte représente une étape intéressante dans l’histoire de l’interprétation de la puissance et de l’acte, telle que mise en lumière de manière brillante par Gwenaëlle Aubry. En effet, notre texte témoigne du développement de la notion de dunamis de son sens aristotélicien de « potentialité » à celui, plotinien, de « puissance » au sens de pouvoir, ce qui permet au fondateur du néoplatonisme de parler du premier principe comme « puissance de toutes choses » (δύναμις πάντων)63. Dans notre texte arabe, l’Adaptateur, quant à lui, essaie de résoudre la confusion entre ces deux sens, qu’il retrouve chez Plotin, en attribuant le sens de « puissance » au monde intelligible, tandis que le sens de « potentialité » se voit réservé au monde sensible. Il s’agit d’une solution typiquement néoplatonicienne et surtout porphyrienne : préserver la cohérence d’une doctrine aux éléments conflictuels en les attribuant à divers niveaux ontologiques64. Michael CHASE CNRS Centre Jean Pépin-UMR 8320-ENS-PSL 7 rue Guy Môquet F-94800 Villejuif [email protected]

62. C’est ce qu’a montré Cr. D’ANCONA, « Pseudo-Theology of Aristotle, Chapter I : Structure and Composition », dans : Oriens 36 (2001), pp. 78-112 63. PLOTIN, Ennéades III, 8, 10, 1 ; V, 1, 7, 9. Voir l’ouvrage de AUBRY, cité supra n. 5 64. Voir par exemple M. CHASE, « Porphyre et les Catégories », texte inédit disponible à l’adresse suivante : https://www.academia.edu/24202954/Porphyre_et_le_Cat%C3%A9gories

LE COMMENTAIRE D’AVICENNE À LA THÉOLOGIE D’ARISTOTE : ÉTUDE PRÉLIMINAIRE À L’HISTOIRE D’UNE TRANSMISSION COMPLEXE* Meryem SEBTI Abstract The transmission of Avicenna’s commentary on the Theology of Aristotle is complex, despite the existence of a manuscript of it copied by the hand of a third generation disciple of Avicenna. This transmission took place through two distinct recensions – a long one and a short one –, which have not been differentiated in A. Badawi’s edition. The present study aims to present the manuscript tradition of the two recensions and to study their doctrinal differences. Indeed, their comparison reveals that a doctrinal choice was made by the person or persons responsible for the short recension.

La Théologie d’Aristote est un texte d’une importance majeure dans l’histoire de la réception du corpus grec dans le monde arabo-musulman. Texte élaboré à partir des trois dernières Ennéades de Plotin, il a été attribué à Aristote1. Le commentaire d’Avicenne fait partie, comme le Commentaire à Métaphysique Lambda, du Kitāb al-inṣāf wa-l-intiṣāf, * La rédaction de cet article n’a été possible que grâce à un travail commun préalable entrepris avec Marc Geoffroy, qui nous a quitté en 2018, Jules Janssens et Michael Chase. Nous poursuivons tous trois ce travail, en hommage à notre ami Marc Geoffroy. Notre intention est d’établir une édition critique du Tafsīr Kitāb Uthūlūjiyā min al-Inṣāf ‘an al-Shaykh al-Ra’īs Ibn Sīnā. Le présent article ne prend en compte que les mss suivants : Le Caire Dār al-kutub, Ḥikma 6 ; Le Caire, Dār al-kutub, Ḥikma 102 ; Oxford Bodleian Marsh 536 ; Téhéran, Majlīs 1845 ; Tashkent, 2385. Pour notre édition critique du Tafsīr, nous prenons en compte d’autres mss : Malik 6169, 3, (folios 41v-56r) daté Jumāda al-awwal 1105 H.Q ; ‘Ulūm pizishkī 190, 1, folios 1-13 (non daté) le début manque ; Millī 276 folios 15(16?)v-42v (?) (La numérotation des folios n’est pas très lisible). Le travail présenté dans cet article repose sur l’étude du seul Mīmar un. Ses résultats sont donc appelés à être affinés après l’étude des autres chapitres ainsi que celle des autres manuscrits. Je remercie le projet PhASIF grâce auquel nous avons obtenu certains des mss qui servent de base à cette étude. 1. Sur la nature de ce texte, voir M. AOUAD, « La théologie d’Aristote et autres textes du Plotinus arabus », dans : R. GOULET (éd.) Dictionnaire des philosophes antiques, vol. 1, Paris 1989, pp. 541-593. Voir aussi pour le premier Mīmar, auquel se réfère notre étude,

46

MERYEM SEBTI

ouvrage perdu du vivant même d’Avicenne2. Les deux textes présentent d’ailleurs des similitudes de style : syntaxe complexe, emploi fréquent des pronoms affixes qui rendent difficile l’identification des référents. Le Commentaire à La Théologie d’Aristote a été transmis sous deux recensions distinctes : une version longue, le Tafsīr Kitāb Uthūlūjiyā min al-Inṣāf ‘an al-Shaykh al-Ra’īs Ibn Sīnā et une version courte, Fī Sharḥ Uthūlūjiyā min Kitāb al-Inṣāf ‘an al-Shaykh al-Ra’īs Ibn Sīnā. L’édition existante du Commentaire d’Avicenne à la Théologie3 est très problématique puisqu’elle ne différencie pas les deux recensions de ce même texte qui, dans le manuscrit Ḥikma 6, sont clairement séparées : elle pioche dans l’une ou l’autre sans les distinguer ni les collationner. Elle est même source de confusion puisque le titre de l’ouvrage indiqué dans la page de garde est : Sharḥ Kitāb Uthūlūjiyā al-mansūb ilā Arisṭāṭālīs4 (c’est aussi le titre inscrit dans la table des matières). Or, à la page suivante, le titre du texte est : Tafsīr Kitāb Uthūlūjiyā min al-Inṣāf ‘an al-Shaykh al-Ra’īs Ibn Sīnā. Tel qu’il est édité, le texte est inutilisable pour servir de base à une étude doctrinale sérieuse. Voici la répartition respective des deux versions dans la tradition manuscrite : Fī Sharḥ Uthūlūjiyā min Kitāb al-Inṣāf ‘an al- Shaykh al-Ra’īs Ibn Sīnā (version courte) : – Le Caire Dār al-kutub, Ḥikma 6, f. 142r14 -146r6 – 1e moitié du XIIe s. Copies modernes : – Le Caire, Dār al-kutub 216, f. 1-225. C. D’ANCONA, « Pseudo-Theology of Aristotle, Chapter 1 : Structure and composition », dans : Oriens 36 (2001), pp. 78-112. 2. D. GUTAS, Avicenna and the Aristotelian Tradition, Leyde 2014 (2e éd.), pp. 144-155. 3. Sharḥ Kitāb Uthūlūjiyā al-mansūb ilā Arisṭāṭālīs, dans : A. BADAWI (éd.), Ariṣṭū ‘inda-l-‘Arab, Koweit (2e éd.), 1978, pp. 37-116. 4. Sharḥ Kitāb Uthūlūjiyā al-mansūb ilā Arisṭāṭālīs, dans : Ariṣṭū ‘inda-l-‘Arab, p. 36. 5. Ce manuscrit est aisément datable (1335H =1916-17). Il s’agit en effet d’une copie du ms Ḥikma 6 commandée par les autorités de la bibliothèque khédivale cette même année, voir D. GUTAS, « Texts from Avicenna’s Library in a Copy by Abd al-Razzâq al-Sighnâkhî », dans : Manuscripts of the Middle East 2 (1987), pp. 8-17 ; version révisée dans D. GUTAS, Orientations of Avicenna’s Philosophy. Essays on his Life, Method, Heritage (CSS 1050), Farnham, Surrey / Burlington 2014 (Paperback Routledge, 2020), III, p. 12.

LE COMMENTAIRE D’AVICENNE À LA THÉOLOGIE D’ARISTOTE

47

– Le Caire, Dār al-kutub, Taymūr Ḥikma 86, f. 111-1266. – Brousse, daté de 675H7, f. 123v-128r. Le Tafsīr Kitāb Uthūlūjiyā min al-Inṣāf ‘an al-Shaykh al-Ra’īs Ibn Sīnā (version longue) : – Le Caire, Dār al-kutub, Ḥikma 6, f. 146r7-153v. Copies modernes : – Le Caire, Dār al-kutub 215 est un apographe moderne de Ḥikma 68. – Le Caire, Dār al-kutub 86, Taymūr Ḥikma 86, f. 126-1579. – Le Caire, Dār al-kutub, Ḥikma 102, f. 2-20. Daté de 1684. – Oxford, Bodleian, Marsh 536, f. 69v-84v. Non daté et composite. – Tashkent, 2385, f. 105 v-113r. – Téhéran, Ilāhiyyāt, f. 2v-11r, daté Muḥarram 1087 H.Q./1708. – Téhéran, Majlis 1845 (ancien Tangābunī 335, 2), f. 29-52, daté Muḥarram 1088 H.Q./1709. – Téhéran, Malik 6169, 3 (folio 41v-56r) daté Jumādā al-awwal 1105 H.Q. – Téhéran, ‘Ulūm pizishkī 190, 1, folio 1-13 (non daté), le début manque. – Téhéran, Millī 276 folio 15 (16?)v-42v (?) non daté10. – Ankara, İsmail Saib 4605 ? (copié 696 H.)11. Dimitri Gutas, dans son article pionner, a eu le mérite de signaler l’importance du manuscrit Ḥikma 6 appartenant à la collection Muṣṭafā Il existe, de plus, une copie manuscrite et une photocopie, toutes deux datant du milieu du XXe siècle, voir ibid., p. 11. 6. Pour la datation tardive de ce manuscrit, voir GUTAS, Orientations, III, pp. 13-14. 7. Voir Y. MICHOT, « Un important recueil avicennien du VIIe/XIIIe s. : La Majmû‘a Hüseyin Çelebi 1194 de Brousse », dans : Bulletin de Philosophie médiévale 33 (1991), pp. 122-123. 8. Voir supra, n. 4. Voir aussi P. KRAUS, « Plotin chez les Arabes. Remarques sur un nouveau fragment de la paraphrase arabe des Ennéades », dans : Bulletin de l’Institut d’Égypte 23 (1940-41), pp. 263-295, en particulier p. 274, n. 2. 9. Voir supra, n. 5. 10. La foliation n’est pas claire sur la copie dont nous disposons. Une consultation de l’original s’impose. 11. Il en existerait un microfilm à Téhéran, mais nous n’y avons pas encore eu accès (information présente dans la version revue de l’article de GUTAS, « Text » dans Id., Orientations, III, p. 18). Selon la numérotation actuelle, le manuscrit İsmail Saib 4605 ne contient pas le commentaire d’Avicenne – information obtenue grâce à l’amabilité de M. Aouad et son équipe dans le cadre du projet PhASIF.

48

MERYEM SEBTI

Fāḍil à Dār al-Kutub au Caire. Cette majmū‘a contient des œuvres importantes d’Avicenne, ainsi que des traductions arabes d’Aristote et de Thémistius. L’importance de ce manuscrit, ainsi que l’indique Dimitri Gutas, est principalement liée au fait qu’il a été copié dans la première moitié du douzième siècle12 par un disciple de la troisième génération d’Avicenne, ‘Abd-al-Razzāq al-Ṣighnākhī13 à partir de textes qui se trouvaient dans la bibliothèque d’Avicenne. Ce manuscrit contient les deux recensions de l’ouvrage d’Avicenne, le Sharḥ et le Tafsīr. Dimitri Gutas est aussi le premier à avoir mentionné l’existence des deux versions dans les différents manuscrits et à avoir signalé la foliotation des deux versions. Pour bien comprendre le problème posé par l’édition de Badawī, il faut également se référer à l’article de Gutas – qui explique en détail ce qui, du Tafsīr, a été édité par le savant égyptien. Badawī n’a publié qu’une partie du Sharḥ (qui va du folio 142r l. 15 au folio 146r l. 6 dans le ms. Ḥikma 6). Il s’agit de la partie qui, dans le manuscrit, va du folio 143v l. 19 au folio 146r l. 6 et qui, dans l’édition de Badawī, va de la page 59, l. 13 à la page 66, l. 4. Ainsi que l’explique Dimitri Gutas, les passages allant dans le manuscrit des folios 142r l. 15 au folio 143v l. 19 sont omis par Badawī parce qu’ils offrent une recension différente de certaines parties du texte inclus dans le manuscrit sous le titre Tafsīr Kitāb Uthūlūjiyā. Dimitri Gutas donne en détail le recoupement des sections du Tafsīr Kitāb Uthūlūjiyā, publiées dans Ariṣṭū ‘inda-l-‘Arab. Le lien entre la recension courte et la recension longue n’a jusqu’à présent jamais été étudié. On pourrait penser que la version courte – le Sharḥ – est un résumé fait par al-Ṣighnākhī après avoir copié la version longue. Cependant, l’examen de la tradition manuscrite semble indiquer que ce n’est pas le cas. L’étude de cette tradition nous permet de savoir deux choses importantes. Tout d’abord, qu’al-Ṣighnākhī n’est pas l’auteur de la version courte. Elle existait déjà au moment où il copie le ms. Elle est donc ancienne et date d’avant 1150. En effet, nous constatons que la version courte comporte quelques leçons originales relevées au cours de cette collation du premier Mīmar. Cela pourrait signifier qu’elle n’a pas été copiée à partir de la copie manuscrite 12. GUTAS, « Texts », dans Id., Orientations, III, p. 10. 13. Il a été l’élève d’Abū-l-‘Abbās al-Lawkarī comme cela a été rapporté par al-Bayhaqī ; cf. GUTAS, « Texts », pp. 8-9.

LE COMMENTAIRE D’AVICENNE À LA THÉOLOGIE D’ARISTOTE

49

d’al-Ṣighnākhī dans Ḥikma 6. Ensuite, l’examen des manuscrits contenant la version longue nous permet d’affirmer que la copie de la version longue d’al-Ṣighnākhī (Ḥikma 6) repose déjà sur une tradition manuscrite complexe et non sur un autographe comme pourrait le laisser penser sa position de « disciple de la troisième génération » d’Avicenne et héritier de sa bibliothèque. À ce stade de la collation (qui ne couvre que le premier Mīmar), quatre cas nous permettent d’affirmer que la tradition manuscrite n’a pas pour origine la copie d’al-Ṣighnākhī. Cette tradition remonte donc avant 1150 et se révèle être déjà complexe au moment où il réalise sa propre copie des deux versions du commentaire d’Avicenne. Examinons maintenant les exemples qui permettent de soulever la question de savoir si la version courte, le Fī Sharḥ Uthūlūjiyā, a pour source directe la version longue ou non. Exemple n° 1 : Manuscrit Ḥikma 6, Folio 142, ligne 18 du Sharḥ : Ms Bursa folio 123, ligne 17 du Sharḥ : ‫مجردة عن المادة معقولة لما يكون لها ويجوز‬ ‫وإنما كان عاقلا لذاته لأنها‬ ّ ‫ فكيف حالها مع نفسها ؟ فإذا عقلت ذاتها فالمعقولات التي تتلو‬.‫اقترانها به‬ 14 .‫ذاتها بلا واسطة تكون ممكنة له أن يعقلها‬ Manuscrit Ḥikma 6, Folio 146, ligne 18 Tafsīr : ‫مجردة عن المادة معقولة لما يكون لها ويجوز‬ ‫وإنما كان عاقلا لذاته لأن ذاته‬ ّ ‫ فكيف حالها مع نفسها ؟ فإذا عقلت ذاتها فالمعقولات التي تتلو‬.‫اقترانه بها‬ .‫ذاتها بلا واسطة تكون ممكنة له أن تعقلها‬ Exemple n° 2 : Ms Ḥikma 6, Folio 142v, ligne 3 du Sharḥ : ‫نعم ما حكم بان فعل مبدا الحق المعقول والشيء الموجود في المحسوس‬ .‫وهذا الموجود الجرماني واحدا وهو الحق الأول‬ 14. Correspond à l’édition Badawī du Tafsīr, pp. 37, l. 16-18 – 38, l. 1-2.

50

MERYEM SEBTI

La leçon de Bursa (folio 123 v, ligne 26) n’est pas tout à fait claire et peut être lue tout aussi bien comme ‫ فعل‬que comme ‫ جعل‬en raison de la ressemblance des deux ductus. Le Tafsīr a : ‫نعم ما حكم بان جعل مبدا الحق المعقول والشيء الموجود في المحسوس‬ .‫وهذا الموجود الجرماني واحدا وهو الحق الأول‬ Ḥikma 6 folio 148r, ligne 20 ; Le Caire, Dār al-kutub, Taymūr Ḥikma 102, folio 7, ligne 13 ; Oxford Bodleian, Marsh 536, f. 14, ligne 1. Comme on peut le constater, la leçon du Tafsīr est correcte, alors que celle du Sharḥ est fautive. Exemple n° 3 : Ms Ḥikma 6, Folio 142v, ligne 9 Sharḥ on lit (Bursa, folio 124r, ligne 9) : ‫الذي يلزمه باعتبار ذاته الإمكان وقوة ما بوجه اخر‬ Dans le Tafsīr on lit : ‫الذي يلزمه باعتبار ذاته الإمكان وهو قوة ما بوجه اخر‬ Manuscrit Ḥikma 6, Folio 148v, ligne 5 ; Le Caire, Dār al-kutub, Taymūr Ḥikma 102 folio 7, lignes 21-22 ; Oxford Bodleian, Marsh 536, folio 64, ligne 13. Ici encore, la leçon du Tafsīr est juste et celle du Sharḥ pose problème. Exemple n° 4 : Dans le Sharḥ (Ḥikma 6, Folio 142v, ligne 11 et Bursa, folio 124r, ligne 11) on a : ‫أقول ان صدور الفعل عن الحق الأول انما يتأخر عن المبدأ الأول لا بزمان‬ .‫بل بحسب الذات‬ Pour le Tafsīr, Ḥikma 6, Folio 148v, ligne 7 ; Le Caire, Dār al-kutub, Taymūr Ḥikma 102 folio 8, ligne 2 ; Oxford Bodleian, Marsh 536, f. 14, ligne 17 :

LE COMMENTAIRE D’AVICENNE À LA THÉOLOGIE D’ARISTOTE

51

‫أقول ان صدور الفعل عن الحق الأول انما يتأخر عن المبدأ الأول لا بزمان‬ .‫بل بحسب الذات على ما صحح في الكتب‬ On constate qu’il manque dans le Sharḥ une précision qui se trouve dans le Tafsīr. Ces quatre exemples ne permettent pas d’affirmer de façon catégorique que le Sharḥ n’a pas été copié à partir de la version longue. Les différences de leçons pourraient être dues à des erreurs de copistes ou à une omission (pour ce qui est du dernier exemple). Ils permettent cependant de légitimement douter du fait que le Sharḥ. a été copié à partir du Tafsīr. Cette hypothèse devra être examinée avec le plus grand soin lors de la collation du reste du texte. L’examen de la tradition manuscrite du Tafsīr L’examen de la tradition manuscrite permet de penser que la copie du Tafsīr d’al-Ṣighnākhī (Ḥikma 6) repose déjà sur une tradition manuscrite complexe et non sur un autographe. Je me limite dans toute cette étude à la collation de 5 manuscrits (Ḥikma 6 ; Marsh ; Ḥikma 102 ; Majlis ; Tashkent). Un cas relevé lors de la collation montre qu’une erreur évidente dans Ḥikma 6 ne se trouve pas dans le reste de la tradition manuscrite. (1) On lit dans Ḥikma 6 (folio 146r, ligne 11) : ‫وان كانت النفس مع ذالك لا تموت ولكن معناه ان النفس لم وجب لها ان لا‬ ‫تكون مجردة عن البدن‬ Ce ‫ وجب‬est une faute manifeste que l’on ne trouve que dans Ḥikma 6. Marsh (folio 69 v, ligne 6) a : ‫وان كانت النفس مع ذالك لا تموت ولكن معناه ان النفس لم تجب لها ان لا‬ ‫تكون مجردة عن البدن‬ Ḥikma 102 (folio 2, ligne 2) ; Majlis (folio 69, ligne 4) et Tashkent (folio 105, ligne 2) ont : ‫وان كانت النفس مع ذالك لا تموت ولكن معناه ان النفس لم يوجب لها ان‬ ‫لا تكون مجردة عن‬

52

MERYEM SEBTI

Cet exemple semble indiquer que la tradition manuscrite du Tafsīr n’a pas pour origine la copie d’al-Ṣighnākhī. Cette double tradition remonterait donc avant 1150 et était déjà existante au moment où le disciple d’Avicenne réalise sa propre copie des deux versions du commentaire d’Avicenne. Cette hypothèse nécessite elle aussi d’être corroborée par la collation de la suite du texte. Un certain nombre d’exemples montrent que Ḥikma 102 a souvent des leçons propres contre tous les autres ms. Cependant, à 12 reprises (+ le cas cité d’accord avec la Théologie) sur 34 cas relevés, donc dans 41% des cas on a un accord entre Majlis, Tashkent et Ḥikma 102. On peut donc penser qu’ils ont une source commune. Par ailleurs, Majlis et Tashkent ont quasiment toujours les mêmes leçons dans 90% des cas. Il ne semble pas que l’un soit une copie de l’autre mais plutôt qu’ils sont l’un et l’autre la copie directe d’un même manuscrit. Les exemples que je cite un peu plus loin montrent pourquoi l’un ne peut être une copie directe de l’autre. Cependant, d’autres exemples relevés lors de la collation montrent de nombreux cas de contamination. (1) Par exemple sur ce passage Ḥikma 6 a (folio 146 r, lignes 10-12): ‫وإن كانت النفس مع ذلك لا تموت ولكن معناه أن النفس لم وجب لها أن‬ ‫لا تكون مجردة عن البدن تجرد الأمور العقلية المذكورة فيما بعد الطبيعة بل‬ .‫وجد له في الطبع علاقة و ميل مع البدن في أول الأمر‬ Marsh a (folio 69, lignes 6-8): ‫وإن كانت النفس مع ذلك لا تموت ولكن معناه أن النفس لم تجب لها أن‬ ‫لا يكون مجردة عن البدن بتجرد الأمور العقلية المذكورة فيما بعد الطبيعة بل‬ .‫وجب له في الطبع علاقة و ميل مع البدن في أول الأمر‬ Ḥikma 102 a (folio 2, lignes 4-6): ‫وإن كانت النفس مع ذلك لا تموت ولكن معناه أن النفس لم يوجب لها أن‬ ‫الصور العقلية المذكورة فيما بعد الطبيعة بل‬ ّ ‫لا تكون مجردة عن البدن بمجرد‬ .‫وجب له في الطبع علاقة و ميل مع البدن في أول الأمر‬

LE COMMENTAIRE D’AVICENNE À LA THÉOLOGIE D’ARISTOTE

53

Et Majlis (folio 29, lignes 1-4) a : ‫وإن كانت النفس مع ذلك لا تموت ولكن معناه أن النفس لم يوجب لها أن‬ ‫لا تكون مجردة عن البدن بمجرد الأمورالعقلية المذكورة فيما بعد الطبيعة بل‬ .‫وجب له في الطبع علاقة و ميل مع البدن في أول الأمر‬ Tashkent, 105 v lignes 2-4 ‫وإن كانت النفس مع ذلك لا يموت ولكن معناه أن النفس لم يوجب لها أن‬ ‫لا تكون مجردة عن البدن بمجرد الأمور العقلية‬ ‫المذكورة فيما بعد الطبيعة بل وجد له في الطبع علاقة و ميل مع البدن في‬ .‫أول الأمر‬ Nous constatons ici un accord entre Ḥikma 6, Marsh et Tashkent (bien que ce dernier ait une leçon différente sur ‫وجد‬/‫ وجب‬qui pourrait être une faute d’inattention du copiste) alors que Majlis est plus ou moins proche de Ḥikma 102 (il en reprend un mot sur deux). (2) Dans un autre cas, on a un accord de tous les manuscrits contre Ḥikma 102. Ainsi, on lit dans Ḥikma 6 (folio 148, ligne 13) : ‫وانما صارت في هذا العالم فرارا من سخط الله أي فرارا من أن تكون ناقصة‬ .‫الجوهر فتبقي بعيدة من عناية الله‬ C’est la même leçon qui est dans tous les manuscrits (Majlis, folio 35, ligne 11 ; Marsh folio 73v, ligne 7 ; Tashkent, folio 107v, ligne 10). Seul Ḥikma 102 a la leçon suivante : ‫وانما صارت في هذا العالم فرارا من سخط الله أي فرارا من أن تكون ناقصة‬ .‫الذات فتبقي بعيدة من عناية الله‬ (3) Par ailleurs on a un cas d’homéotéleute qui ne se trouve que dans Ḥikma 102 (folio 7, ligne 7 qui correspond à Ḥikma 6, folio 148r, ligne 16). Cela indique donc que Tashkent et Majlis ne sont pas des copies de Ḥikma 102. Tout cela montre que la tradition est contaminée.

54

MERYEM SEBTI

Si on examine maintenant la tradition dont est issu le manuscrit Ḥikma 6, on constate que le Sharḥ, dans un cas de leçon divergente, se rapproche davantage de la formulation de Ḥikma 6 et Marsh que de celle des autres mss. Ainsi, dans le Sharḥ on lit dans Ḥikma 6 (folio 142 r, ligne 21) : .‫وهو الحياة العقلية التي بين امرها في الكتب‬ Dans le Tafsīr dans Ḥikma 6 (folio 146r, ligne 21) on a : .‫وهو الحياة العقلية التي بين امرها في كتبه‬ Cette dernière leçon est également dans Marsh (folio 70r, lignes 3-4). Dans Majlis (folio 29, ligne 10), Ḥikma 102 (folio 2, ligne 16) et Tashkent (folio 105 v, ligne 13) on a : .‫وهو الحياة العقلية التي بين امرها في ف ّنه‬ Voici un autre cas de leçon totalement divergente : Dans Ḥikma 6 (folio 146 v, lignes 1-2) ; Marsh (folio 70r, lignes 7-9) – il s’agit ici du Tafsīr : ‫وقال ثابت فيه دائما لا يزول عنه أي ليس هذا التجرد والتفرد يكون وقتا دون‬ ‫وقت بل دائما لأن سببه وهو الإستغناء بالكمال الأول في الوجود‬ Dans Majlis (folio 30, ligne 1) ; Tashkent (folio 105v ligne 16) et Ḥikma 102 (Folio 2, ligne 19) : ‫وقال ثابت فيه دائما لا يزول عنه أي ليس هذا التجرد والتفرد يكون وقتا دون‬ ‫وقت بل دائما لأن سببه وهو الإستكمال بالكمال الأول في الوجود‬ Sur les 34 relevés de variantes effectués, on constate 20 cas d’accord entre Ḥikma 6 et Marsh. Ce qui fait un ratio d’un peu plus 58% des cas. Voici à présent une série de cas qui montrent une contamination entre les différents manuscrits. Nous avons ainsi le cas d’une leçon

LE COMMENTAIRE D’AVICENNE À LA THÉOLOGIE D’ARISTOTE

55

commune à Majlis (folio 30, ligne 1), Tashkent (folio 105r, ligne 21) Ḥikma 102 (Folio 3, ligne 2) et Ḥikma 6 (folio 146 v, ligne 7) : ‫وهو أن كل حالة تكون بعد ما لم تكن فإنها تتعلق بالحركات الجسمانية‬ ‫وتنسب إلى الحركة المستديرة‬ Alors que Marsh (folio 65, ligne 18) a : ‫وهو أن كل حالة تكون بعد ما لم تكن فإنها تتعلق بالحركات الجسمانية‬ ‫وتنسب إلى الحركا ت المستديرة‬ Autre cas : Majlis (folio 30, ligne 15), Ḥikma 6 (folio 106r, ligne 7) et Ḥikma 102 (folio 2, ligne 10) ont : ‫لأن رؤية الأشياء هو قبول صورته ولكن مغزاه إلى رؤية الأشياء التي في العقل‬ Alors que l’on trouve dans Tashkent 6 (folio 66, ligne 10) et Marsh (folio 66, ligne 10) : ‫لأن رؤية الأشياء هو قبول صورته ولكن معناه إلى رؤية الأشياء التي في العقل‬ Encore un autre cas : Ḥikma 6 (folio 147r, ligne 10), Majlis (folio 32, ligne 2), Ḥikma 102 (folio 4, ligne 10) ‫حتى صار لكل شيء منها نصيب بحسبه‬ Alors que Tashkent (folio 106 v, ligne 5) et Marsh (Folio 71v, ligne 3) ont : ‫حتى صار لكل شيء منها نصيبه بحسبه‬ Autre cas : Ḥikma 6 (folio 147 r, ligne 14), Majlis (folio 32, ligne 9) et Tashkent (folio 106 v, ligne 9): ‫لها امکان اتصال بالجواهر العالیة التي لها اللذات الحقيقية والجمال الحقيقي‬ Alors que Marsh (folio 71v, ligne 11) et Ḥikma 102 (folio 4, ligne 17) ont : ‫لها امکان اتصال بالجواهر العالیة التي لها اللذات الحقيقية والكمال الحقيقي‬

56

MERYEM SEBTI

Nous avons là des cas de contamination manifeste. La même différence de leçon, al-jamal / al-kamal se retrouve plus loin dans les mss, avec encore une fois un exemple de contamination évident : Ḥikma 6 (folio 147r, ligne 21) a : ‫متصلة بالعالم الأعلى الأشبه بالجمال الأبهى‬ Ḥikma 102 (folio 5, ligne 4), Tashkent (folio 106 v, ligne 16), Majlis (folio 32, ligne 19) ‫متصلة بالعالم الأعلى الأشبه بالجمال الالهي‬ Marsh (folio 62r, ligne 4) ‫متصلة بالعالم الأعلى الأشبه بالكمال الأبهى‬ Marsh (folio 14, ligne 3) a une leçon commune avec Majlis (folio 14, ligne 1): ‫نعم ما حكم بان جعل مبدا الخلق المعقول‬ Ḥikma 102 (folio 7, ligne 13), Ḥikma 6 (folio 148r, ligne 20), Majlis (folio 41, ligne 2) ont : ‫نعم ما حكم بان جعل مبدا الحق المعقول‬ Autre exemple. On trouve dans Ḥikma 6 (folio 146v, lignes 9-10), Marsh 70v, lignes 3-4, Tashkent (folio 106r, ligne 3) et Majlis (folio 30, ligne 12): ‫فاذا استفاد العقل شوقا كان طلبه لما يشتاق اليه في قبيل غير قبيل العقل‬ Ḥikma 102 (3, ligne 5) ‫فاذا استفاد العقل شوقا كان طلبه لما يشتاق اليه في قبل غير قبل العقل‬ Comme on peut le constater à travers ces exemples, les différents cas de contaminations entre les manuscrits rendent difficile, voire impossible à ce stade de la collation, la constitution d’un stemma.

LE COMMENTAIRE D’AVICENNE À LA THÉOLOGIE D’ARISTOTE

57

La nature du texte La première constatation que l’on peut faire, c’est que, s’il est vrai que le texte du commentaire d’Avicenne se présente sous forme de lemmes, comme le commentaire à la Métaphysique Lambda15, il n’en reste pas moins qu’il existe une différence notable entre les deux. Alors que les lemmes qui constituent le commentaire à Lambda se présentent comme un énoncé suivi et construit, qui peut être lu quasiment indépendamment du texte commenté, les notes qui constituent notre commentaire sont incompréhensibles si l’on n’a pas le texte de la Théologie sous les yeux. Les lemmes, distingués dans les manuscrits par des lettres de l’alphabet arabe (pas toujours selon l’abjad), sont émaillés de ‫ أي‬qui renvoient à des passages de la Théologie. On peut compter 25 ‫ أي‬dans le seul commentaire du Mīmar 1 (à titre de comparaison, il y en a 15 dans tout le commentaire à Métaphysique Lambda où l’on a un seul passage où l’on trouve 4 ‫ أي‬en 4 lignes16 : ‫ أي الحقيقة الأولى للحق الأول فليس‬،‫ وأما ما هو بالأنية الأولى‬: ‫ثم يقول‬ ‫ فإذا المحرك‬،‫أي لأنه أنية قائمة بالفعل لا تخالط القوة‬ ّ ،‫له عنصر لأنه تمام‬ ‫الأول واحد بالكلمة والعدد أي كلمته والقول الشارح لاسمه واحد وهو أيض ًا‬ .‫ أي إنما له ذات واحدة فقط لا تشترك فيها ذاتان‬،‫واحد بالعدد‬ Cela est en revanche tout à fait courant dans le commentaire à la Théologie. Par exemple17: ‫قال وكل جوهر عقلي أي من حيث هو جوهر مستغن عن ان يقوم بالمادة‬ ‫يناله شوق اي يحتاج ان يحصل له شيء غير حاصل له فذلك الجوهر بعد‬ ‫الجوهر الذي هو عقل فقط أي من حيث ليس واحدا للكمال في اول ما به‬ ‫يتجوهر فهو في درجة ثانية أي لا يكون عقليا صرفا في قوام ذاته وفي كمال‬ 15. Pour l’édition du Commentaire d’Avicenne à Métaphysique Lambda, voir : M. GEOF– J. JANSSENS – M. SEBTI, Avicenne. Le commentaire à Métaphysique Lambda d’Aristote (édition critique, traduction et annotation), Paris 2014. 16. GEOFFROY – JANSSENS – SEBTI, Avicenne. Le commentaire à Métaphysique Lambda d’Aristote, pp. 65, 211-214. 17. Sharḥ Kitāb Uthūlūjiyā al-mansūb ilā Arisṭāṭālīs, dans BADAWĪ, Ariṣṭū ‘inda-l-‘Arab, p. 38, lignes 11-16. FROY

58

MERYEM SEBTI

‫ذاته الذي يتلوا قوامه بل وان كان من حيث ذاته مجردا عن المادة فانه من‬ .‫حيث كمال ذاته الذي يكون له بعد ما لم يكن يحتاج الى مادة‬ Ou encore ce paragraphe18: ‫ تعاني عذابا‬/147v/ ‫أي اذا فارقت البدن لم تتصل الا بعد تعب شديد أي‬ ‫شديدا كثيرا حتى تنمحي عنها كل دنس ووسخ ٮعلق بها من البدن أي لانها‬ ‫انما تستبقي بالافعال الرديئة فاذا تعطلت جاز ان ٮبطل بل وجب فان قال قائل‬ ‫كما ان الهيئات والكمالات التي تستحدث للنفس لا تتم الا بالبدن على ما‬ .‫ادعيتم وٮٮٮٮم فكذلك بطلان الهيئات لا يكون الا بالبدن‬ Ou encore celui-ci ou le ‫ أي‬donne l’impression qu’il s’agit de notes non continues19 : ‫قال هي مفارقة عند انتقاضه وتحلله أي النفس الانسانية التي هي الاصل ولها‬ ‫هذه فان الصحيح ان للانسان ولكل حيوان نفسا واحدة ولها قوى عدة وانها‬ .‫اصل القوى لانبعاث القوى واما أي القوى ٮـبقى معها فهو بحث اخر‬ Plus qu’un commentaire suivi, ce texte donne plutôt l’impression d’être une série de gloses qui accompagnent le texte de la Théologie. On pense à des notes marginales sur les marges d’un manuscrit ou à des notes prises par un tiers lors d’une lecture donnée par Avicenne. On trouve par exemple un cas où un lemme a été déplacé. Le ‫أي‬ se réfère à l’avant dernier paragraphe du texte suivi et non au texte qui précède immédiatement. Il s’agit du passage suivant – la traduction suit ici notre édition provisoire et non l’édition Badawī : Lorsque l’usage que l’âme fait [du corps]20 n’advient pas conformément à ce qui est requis, mais qu’il provient d’un état (hay’a) qui l’affecte par accident du fait de son orientation (iqbāl) [vers l’inférieur] – qu’elle devient donc corporelle, que s’emparent d’elle des états (hay’āt) de soumission aux choses corporelles tels le désir, la colère, et cetera, voire que ces états deviennent en elle des habitus – après [avoir quitté] le corps, l’âme est semblable à l’ensemble qui était dans le corps et elle est empêchée d’accéder au monde supérieur. 18. Sharḥ Kitāb Uthūlūjiyā, p. 42, lignes 7-11. 19. Sharḥ Kitāb Uthūlūjiyā, p. 41, lignes 6-8. 20. On lit isti‘malihi et non isti‘malihā

LE COMMENTAIRE D’AVICENNE À LA THÉOLOGIE D’ARISTOTE

59

[…ni salie par les impuretés, etc.] Il entend par « impuretés » des appendices vils, ignobles, non naturels, inconvenants, qui sont concomitants à une chose, laquelle est pure en comparaison avec eux. Lorsque l’âme se sépare du corps, elle reste jointe au monde supérieur, qui est le plus semblable à la Beauté suprêmement splendide, coupée du monde où elle était. [explicitation de l’empêchement de l’avant dernier §] C’est-à-dire que lorsqu’elle se sépare du corps, elle ne se joint [au monde intelligible] qu’après une intense fatigue, c’est-à-dire qu’elle endure un tourment très (kathīran) intense, jusqu’à ce que s’efface d’elle toute souillure et toute impureté qui lui est attachée du fait du corps, et ce [ay, qui explicite la raison de l’assertion précédente] parce que ces dernières ne persistent qu’à cause des actes vils, et si ces actes ne sont plus opérés [puisqu’il n’y a plus de corps pour les accomplir], il est possible, voire nécessaire, qu’elles soient détruites.

Nous avions eu également un cas d’interversion de paragraphe dans le commentaire à Lambda. Néanmoins, comme je le mentionnais plus haut, il s’agit d’un texte beaucoup plus construit en dépit de cette interversion. La multiplication des ‫ أي‬dans des lemmes parfois très courts dans le commentaire à la Théologie font penser que ce texte n’a pas été rédigé comme un commentaire continu. Ce qui distingue le Sharḥ du Tafsīr Le Sharḥ n’est pas un simple résumé du Tafsīr dans la mesure où il n’en reprend pas tous les thèmes même de façon abrégée. La question de l’âme est au cœur du commentaire d’Avicenne sur la Théologie. Sur 50 lemmes qui constituent ce commentaire, 32 sont consacrés à la question de l’âme. Il s’agit avant tout pour Avicenne d’établir que l’âme ne préexiste pas au corps et que les passages de la Théologie d’Aristote qui semblent aller dans ce sens sont, soit corrompus, soit mal interprétés. Or, lorsqu’on analyse le Sharḥ, on se rend compte que tous les lemmes portant sur la question de l’âme et de l’intelligible ont été supprimés. Il ne subsiste que les développements sur le Principe Premier et son rapport à Sa création. Le lemme 1 du Sharḥ, correspond à Tafsīr (lemme 2 (‫ )ب‬p. 37, ligne 14 - p. 38, ligne 12 de l’édition Badawī) : il s’agit d’un passage sur l’intellection qu’a le Premier Principe de lui-même et par laquelle il intellige ce qui procède de lui. Ce lemme évoque aussi la nature du

60

MERYEM SEBTI

monde intelligible. Le lemme 2 correspond à Tafsīr (lemme ‫ يا‬p. 42, lignes 3-15 de l’édition Badawī), qui correspond à l’avant dernier chapitre du Mīmar 1. Il s’agit d’un passage sur le Premier qui est le Bien et qui est toujours en acte. Entre les deux, le premier lemme du Tafsīr concernant la non-existence de l’âme avant le corps a été supprimé. Le troisième lemme concernant la perfection de l’âme humaine qui est progressive, est supprimé ; le quatrième lemme sur le fait que toute substance intelligible a un désir qui la fait tendre vers un autre, a été supprimé ; le cinquième lemme qui est un commentaire du passage de la Théologie où il est dit que « l’âme avait vu dans l’intelligible » est supprimé ; le sixième lemme concernant la nécessité de l’actualisation de la forme intelligible en puissance par une forme intelligible en acte, est supprimé ; le septième que l’âme a un désir pour le monde sensible, est supprimé ; le huitième, que l’âme est une substance intelligible par soi, est supprimé ; le neuvième, que l’âme se perfectionne au moyen du particulier et de l’universel, est supprimé ; le dixième, que l’âme humaine est une substance une de laquelle procèdent des puissances diverses, est supprimé ; le onzième que l’âme, qui est une substance inaltérable et pure, lorsqu’elle quitte le corps revient à son origine, est supprimé ; le douzième que l’âme doit éprouver de grands tourments afin de se débarrasser des concrétions acquises, est supprimé ; le treizième sur le retour de l’âme à son monde d’origine, est supprimé ; le quatorzième sur le mode de pensée discursive, est supprimé ; le quinzième sur le rapport de l’âme à son amour véritable et au corps, est supprimé ; le seizième sur le perfectionnement de l’âme ici-bas, est supprimé ; le dix-septième lemme sur la même question, est supprimé. Si l’on poursuit l’examen du Sharḥ au-delà du Mīmar 1, on constate la même chose. Le Sharḥ n’a conservé que les passages relatifs au Premier et à la création : lemme 3, sur l’émanation de l’acte à partir du Premier ; lemme 4, sur l’intellection de soi du Premier ; le mode de connaissance du Premier par l’intellect ; de la lumière qui procède du Premier sur toutes les créatures ; que toute créature est possible de soi et nécessaire par Dieu ; de la générosité divine (al-jūd) et comment elle procède de Dieu sur les créatures ; de l’unité du Premier et de comment la multiplicité est créée ; de la miséricorde divine ; que le monde intelligible contient toutes les perfections ; que le Premier connaît tout ce qui procède de Lui ; de la création (al-ibdāʻ), du possible et du nécessaire ; sur l’omnipotence (qudra) de Dieu.

LE COMMENTAIRE D’AVICENNE À LA THÉOLOGIE D’ARISTOTE

61

La tradition manuscrite nous apprend que ce compendium centré sur les passages proprement théologiques du Commentaire d’Avicenne a été constitué très tôt. Par qui, à quel usage et dans quel but ? Ce sont là les questions qui demeurent ouvertes et auxquelles nous devrons tenter de trouver une réponse. Un long travail reste à effectuer afin de parvenir à une édition critique digne de ce nom. Cette courte étude vise à montrer la complexité de la tradition manuscrite malgré la présence parmi elle d’un manuscrit copié de la main même d’un disciple de la troisième génération d’Avicenne. Elle nous permet aussi de voir que, avant 1150, un compendium avait été constitué à partir du commentaire d’Avicenne à la Théologie, autour d’une thématique exclusivement métaphysique. Ces informations nous renseignent sur les modalités de transmission des manuscrits d’Avicenne et leur probable présence dans des milieux de transmission de la connaissance où de tels compendiums étaient utiles. Meryem SEBTI CNRS Centre Jean Pépin-UMR 8320-ENS-PSL 7 rue Guy Môquet F-94800 Villejuif [email protected]

L’IMAM, PHILOSOPHE ET PROPHÈTE DE LA CITÉ VERTUEUSE (AL-MADĪNA AL-FĀDILA) D’AL-FĀRĀBĪ, EST-IL UN IMAM SHI‘ITE ? Daniel DE SMET Abstract Al-Fārābī’s Mabādi’ ārā’ ahl al-madīna al-fāḍila was extensively used by the Ismā‘īlī author Ḥamīd al-Dīn al-Kirmānī, who adopted parts of its cosmology and theory of knowledge, but also the idea of a perfect state ruled by an Imam who is also a philosopher. The smooth integration of al-Fārābī’s political philosophy into Ismā‘īlī doctrine again raises the thorny question of possible Ismā‘īlī, or general Shī‘ī, influences on al-Fārābī. In this paper, I argue that he probably read Abū Ḥātim al-Rāzī’s A‘lām al-nubuwwa and the “political” epistles of the Ikhwān al-Ṣafā’, although this does not necessarily imply that he was an Ismā‘īlī himself, as these texts circulated widely outside the inner circle of the community. In the last part, I formulate the hypothesis that the Mabādi’, written shortly after the beginning of the Great Occultation of the twelfth Imam, was a kind of Fürstenspiegel for the Shī‘ī amīr Sayf al-Dawla, who was al-Fārābī’s protector in Syria.

1. Introduction Le philosophe ismaélien de tradition fatimide Ḥamīd al-Dīn al-Kirmānī (m. vers 411/1020) était un lecteur enthousiaste des Mabādi’ ārā’ ahl al-madīna al-fāḍila d’al-Fārābī (m. 339/950). Sans jamais citer sa source, il lui emprunte jusque dans les détails la description du premier Intellect auto-intelligeant, avec toutefois cette différence fondamentale que pour l’ismaélien cet Intellect n’est pas l’Instaurateur (mubdi‘), mais le premier être instauré (al-mubda‘ al-awwal). Puis, il adopte de façon créative l’émanation successive des dix Intellects avec les sphères et les corps célestes qui leur correspondent, ainsi que le rôle d’un Intellect transcendant toujours en acte dans le processus d’actualisation de l’intellect humain1. J’avais écrit 1. D. DE SMET, « Al-Fārābī’s Influence on Ḥamīd al-Dīn al-Kirmānī’s Theory of Intellect and Soul », dans : P. ADAMSON (éd.), In the Age of al-Fārābī : Arabic Philosophy

64

DANIEL DE SMET

un peu trop hâtivement à l’époque que les emprunts cessent avec le chapitre 15 des Mabādi’ (d’après la division de l’ouvrage dans l’édition de Richard Walzer)2, là où commence la « philosophie politique », et j’avais laissé en suspens la question épineuse de savoir si al-Fārābī avait lui-même subi une influence shi‘ite, voire ismaélienne3. Dans la présente contribution, que je dédie chaleureusement à Jules Janssens4, j’essaierai de combler cette double lacune en explorant d’abord l’influence d’al-Fārābī sur la conception qu’al-Kirmānī se faisait de l’imam ismaélien ; puis, en remontant dans le temps, je reprendrai la question déjà ancienne, mais toujours irrésolue, d’une possible influence shi‘ite sur la manière dont al-Fārābī se représentait l’imam de sa madīna al-fāḍila, bien conscient que je me risque sur un terrain glissant, pour ne pas dire choquant aux yeux d’un certain mainstream dans les études farabiennes. Au préalable, résumons brièvement le portrait de ce souverain idéal, tel qu’il ressort des Mabādi’. Nous y relevons en particulier les points suivants5 : (1) Le chef (ra’īs) occupe le premier rang dans la structure hiérarchique de la cité vertueuse, tout comme le cœur est l’organe dominant dans le corps ; ils gouvernent l’ensemble de leurs subordonnés, dont ils représentent le principe d’harmonie (XV, § 5). (2) Ce roi (malik) se rapporte à ses sujets comme la Cause première se rapporte aux autres êtres ; dès lors, il correspond à l’Intellect divin ; in the Fourth/Tenth Century, Londres / Turin 2008, pp. 131-150. À cela s’ajoute une série de termes techniques qui ne figurent pas dans la littérature ismaélienne antérieure à al-Kirmānī et proviennent, eux aussi, d’al-Fārābī : la première et la seconde intention (al-qaṣd al-awwal, al-qaṣd al-thānī), la première et la seconde perfection (al-kamāl al-awwal, al-kamāl al-thānī), ‘ināya pour désigner la providence, etc. 2. AL-FĀRĀBĪ, Mabādi’ ārā’ ahl al-madīna al-fāḍila, éd. et trad. R. WALZER, Al-Farabi On the Perfect State, Oxford 1985. 3. DE SMET, « Al-Fārābī’s Influence », p. 133. 4. Le bel article de J. JANSSENS, « Al-Farabi : la religion comme imitation de la philosophie », dans : M. DELGADO – C. MÉLA – F. MÖRI (éds.), Orient-Occident, racines spirituelles de l’Europe, Paris 2014, pp. 497-512, a nourri les réflexions que je présenterai ici. 5. Je suis le découpage en chapitres et paragraphes de l’édition de Walzer, qui diffère de celui adopté par F. DIETERICI, Alfārābī’s Abhandlung der Musterstaat, Leiden 1895 (le chapitre XV de Walzer y correspond aux chapitres 26 à 29). L’édition de Dieterici a été reprise et traduite en français par Y. KARAM – J. CHLALA – A. JAUSSEN, Al-Fārābī. Idées des habitants de la cité vertueuse, Beyrouth / Le Caire 1980. Voir en outre la traduction d’A. CHERNI, Abû Nasr al-Fârâbî. Opinions des habitants de la cité vertueuse, Paris 2011. Toutes les traductions qui figurent dans cet article sont les miennes.

L’IMAM, PHILOSOPHE ET PROPHÈTE DE LA CITÉ VERTUEUSE

65

les deux fonctionnent comme cause finale pour les êtres qui leur sont inférieurs (XV, § 6). (3) Le chef de la cité vertueuse n’est pas n’importe quel homme, car il est prédisposé à gouverner par une disposition innée et par nature (bi l-fiṭra wa-l-ṭab‘) ; cela génère en lui une attitude (hay’a) et une aptitude volontaire (malaka irādiyya) à diriger la cité (XV, § 7). (4) Personne ne peut régner sur lui, puisqu’il a atteint la perfection (istakmala), de sorte qu’il est intellect et intelligé en acte (‘aql wa ma‘qūl bi l-fi‘l). Sa faculté imaginative (quwwa mutakhayyila) a elle aussi, par nature (bi l-ṭab‘), atteint le plus haut niveau de perfection (ghāyat al-kamāl) ; elle est disposée par nature (bi l-ṭab‘) à recevoir, pendant le sommeil ou en état de veille, de l’Intellect agent les particuliers (al-juz’iyyāt) tels qu’ils sont ou sous la forme d’images sensibles qui les imitent, ainsi que les intelligibles (ma‘qūlāt) sous la forme d’images les imitant. Son intellect passif (‘aql munfa‘il) a atteint la perfection de sorte qu’il intellige l’ensemble des intelligibles : il est parvenu ainsi au stade de l’intellect acquis (al-‘aql al-mustafād) (XV, § 8). (5) Il est l’homme en qui s’infuse l’Intellect agent (ḥalla fīhi al-‘aql al-fa‘‘āl) (XV, § 9). (6) Cette infusion affecte les parties théorétique (naẓariyya) et pratique (‘amaliyya) de sa faculté rationnelle, ainsi que sa faculté imaginative ; par cette voie, il reçoit la révélation (waḥī) divine par l’intermédiaire de l’Intellect agent (XV, § 10). Ce qui émane à partir de Dieu Très-Haut sur l’Intellect agent, l’Intellect agent le fait émaner sur son intellect passif par l’intermédiaire de l’intellect acquis, puis sur sa faculté imaginative. Il est par ce qui en émane sur son intellect passif un sage philosophe (ḥakīm faylasūf), qui intellige de façon parfaite « avec un intellect dans lequel réside le divin » (bi-‘aql fīhi al-ilāhī) [?], et par ce qui en émane vers la faculté imaginative, il est un prophète (nabī), un avertisseur de ce qui sera et un rapporteur des particuliers existant actuellement « avec un intellect dans lequel réside le divin » (bi-‘aql fīhi al-ilāhī) [?]6 ». 6. Ce passage est visiblement corrompu. Walzer (p. 244) met dans les deux cas bi ‘aql fīhi al-ilāhī entre crochets, alors que l’apparat critique nous apprend qu’il s’agit (du moins dans le premier cas) d’une conjecture de Baneth adoptée par Walzer : (1) wa muta‘aqqilan ‘ala al-tammām bi ‘aql fīhi al-ilāhī ; (« an accomplished thinker who employs an intellect of divine quality ») ; (2) mukhbiran bi-mā huwa al-ān mina l-juz’iyyāt mawjūd bi ‘aql fīhi al-ilāhī (« who tells of particular things which exist at present » ; ici, il omet donc de traduire bi ‘aql fīhi al-ilāhī). Le texte arabe de Dieterici (pp. 58-59) omet la première occurrence (wa muta‘aqqilan ‘ala al-tammām) et lit pour la seconde mukhbiran bi-mā

66

DANIEL DE SMET

(7) Cet homme occupe le plus haut rang de l’humanité et a atteint le plus haut stade de la félicité (sa‘āda). C’est comme si son âme est unie (muttaḥida) à l’Intellect agent. Ainsi, il connaît toutes les actions qui mènent à la félicité, ce qui est la première condition requise pour diriger la cité. En outre, il doit être capable d’exercer son autorité verbalement, d’agir sur la faculté imaginative de ses auditeurs, afin de les guider (irshād) vers la félicité ; il doit avoir une bonne condition physique pour mener la guerre (ḥarb) : « Ceci est le chef sur qui aucun autre homme n’a de pouvoir ; c’est lui l’imam et le premier chef de la cité vertueuse » (XV, § 11). (8) Il réunit en lui par nature (bi l-ṭab‘) douze qualités (khaṣla) qui lui sont innées (XV, § 12). (9) Les hommes doués d’une telle nature innée (fiṭra) étant très rares, ils n’apparaissent que l’un après l’autre (al-wāḥid ba‘da-l-wāḥid). Si un tel homme n’est pas disponible, mais qu’on dispose de quelqu’un qui réunit ces qualités sans avoir le pouvoir « d’énoncer un message à partir de la faculté imaginative » (al-indhār min jihat al-quwwa al-mutakhayyila), c’est-à-dire sans être un prophète, il sera le chef. Mais si ce type de souverain vient également à manquer à un certain moment, alors qu’auparavant ils se sont succédés (tawāllū) dans la cité, on adoptera alors leurs lois et leurs coutumes (sharā’i wa sunan). Le second chef (al-ra’īs al-thānī), qui succèdera aux premiers, devra réunir en lui six qualités : être philosophe (ḥakīm) ; connaître et préserver les lois et coutumes des premiers souverains ; pouvoir déduire (istinbāṭ) de nouvelles lois à partir des préceptes établis par les premiers imams ; exceller dans la rawiyya (délibération) et l’istinbāṭ pour pouvoir répondre à des questions qui ne se posaient pas à l’époque des premiers souverains, et cela en vue du bien-être (ṣalāḥ) de la cité ; guider (irshād) par la parole vers les lois des premiers souverains et vers celles qui ont été déduites après eux ; avoir la condition physique requise pour mener la guerre (XV, § 13). (10) Ces qualités peuvent être présentes en différentes personnes, qui s’associent pour diriger la cité. Mais si aucun philosophe ne participe au gouvernement, la cité se corrompra et périra inévitablement (XV, § 14). huwa al-ān al-juz’iyyāt bi-wujūd ya‘qil fīhi al-ilāhī, ce que Karam e.a. traduisent par « narrateur des (événements) particuliers présents, et ce grâce à un être dans lequel il intellige le divin » (p. 92). Cherni (pp. 240-241), dans le texte qu’il a établi à partir des éditions existantes, omet simplement le bi ‘aql fīhi al-ilāhī dans les deux cas, parce que, selon lui, cela « n’ajoute rien au sens, mais le déforme même » (p. 240, n. 5).

L’IMAM, PHILOSOPHE ET PROPHÈTE DE LA CITÉ VERTUEUSE

67

Voyons tout d’abord ce qu’al-Kirmānī a fait de ces préceptes, qui restent après tout assez énigmatiques. 2. L’imam selon al-Kirmānī : le principe de la rationalité dans la cité L’ouvrage principal d’al-Kirmānī, le Kitāb Rāḥat al-‘aql, est conçu comme une cité fortifiée avec ses sept murailles (aswār) et ses cinquante-six chemins (mashāri‘) qui renferment la connaissance du cycle cosmique : le Créateur ineffable, la création de l’Intellect, l’émanation des dix Intellects séparés de la matière, la génération des corps et des sphères célestes, la formation du monde sublunaire avec ses trois règnes, végétal, minéral et animal, l’apparition de l’homme, doté d’une âme aux multiples facultés et, enfin, les voies menant au salut de cette âme (remontée vers son origine céleste) ou à sa damnation éternelle7. Une simple comparaison avec le résumé succinct (Iḥṣā’ abwāb hādha l-kitāb) qui figure en tête des Mabādi’ ārā’ ahl al-madīna al-fāḍila, fait apparaître une étroite similitude quant à la thématique et à sa présentation8. Une différence toutefois, de prime abord fondamentale, saute aux yeux : pour le faylasūf al-Fārābī, le sauveur qui indique la voie vers la béatitude de l’âme (sa‘āda) est l’imam de la cité vertueuse, à la fois philosophe et prophète, tandis que le dā‘ī ismaélien al-Kirmānī assigne ce rôle aux prophètes et aux imams (dans le sens shi‘ite du terme) dont les cycles se sont succédés au cours de notre histoire terrestre, pour finalement aboutir à l’avènement du Résurrecteur (qā’im), qui clôturera le cycle cosmique. La « cité idéale » d’al-Kirmānī est en effet basée sur un célèbre hadith shi‘ite qui fait dire au Prophète Muhammad : « Je suis la cité de la connaissance et ‘Alī en est la porte. Quiconque veut la connaissance, qu’il entre par cette porte » (anā madīnat al-‘ilm wa ‘Alī bābuhā fa-man arāda al-‘ilm fal-ya’ti l-bāb)9. Le salut de l’âme réside dans l’acquisition de la connaissance. Les prophètes et les imams, représentés 7. D. DE SMET, La Quiétude de l’Intellect. Néoplatonisme et gnose ismaélienne dans l’œuvre de Ḥamīd ad-Dīn al-Kirmānī, Louvain 1995, pp. 16-17. 8. AL-FĀRĀBĪ, Mabādi’, pp. 38-49, à comparer avec la traduction de la table des matières du Rāḥat al-‘aql dans P. WALKER, Ḥamīd al-Dīn al-Kirmānī. Ismaili Thought in the Age of al-Ḥākim, Londres / New York 1999, pp. 131-141. 9. Ḥamīd al-Dīn AL-KIRMĀNĪ, Kitāb Rāḥat al-‘aql, éd. M. GHĀLIB, Beyrouth 1983, p. 575 ; ID., al-Risāla al-lāzima fī ṣawm shahr Ramaḍān, dans : Majmū‘at rasā’il al-Kirmānī, éd. M. GHĀLIB, Beyrouth 1987, p. 76.

68

DANIEL DE SMET

par leurs archétypes Muhammad et ‘Alī, en sont les enseignants (mu‘allimūn) ; leur enseignement (ta‘līm) est la voie menant au salut et à la libération (najāt, khalāṣ) de l’âme10. À l’instar d’al-Fārābī, pour qui les connaissances requises des habitants de la cité vertueuse (et par conséquent enseignées par son imam) afin d’atteindre la félicité (sa‘āda) sont celles qui sont exposées dans les Mabādi’11, la connaissance salvatrice enseignée par l’imam d’al-Kirmānī recoupe le contenu de son Rāḥat al-‘aql12. Cet ensemble de « sciences nobles » représente la « seconde connaissance » (‘ilm thānī), liée à la seconde perfection de l’âme, et lui procure la félicité éternelle (al-sa‘āda al-abadiyya) en la transformant en un intellect en acte, capable de subsister sans le substrat matériel du corps13. J’ai montré ailleurs qu’al-Kirmānī emprunte pour une large part sa noétique à al-Fārābī, y compris les différentes phases dans l’actualisation de l’intellect humain14. Or, pour l’ismaélien, cette actualisation s’effectue par l’enseignement des imams : Grâce à leur position (makān) et à leur enseignement (ta‘līm), l’existence de l’homme en tant qu’homme devient possible. Ainsi, la perfection de l’âme humaine, suite à l’influence (ta’thīr) qu’ils exercent sur elle par instruction (ta‘līm) et direction (hidāya), atteint le degré de l’accomplissement et des Intellects, devenant analogue aux Anges qui veillent sur le monde15.

En assimilant « les sciences nobles » enseignées par les imams, l’âme humaine passe graduellement de la puissance à l’acte, pour finalement devenir un intellect qui s’intellige soi-même. Cela implique qu’elle s’est détachée de ses liens avec le corps, qui forment obstacle à l’intellection pure. Une fois transformée en un intellect qui intellige 10. Ḥamīd al-Dīn AL-KIRMĀNĪ, Kitāb al-Riyāḍ, éd. ‘Ā. ṬĀMIR, Beyrouth 1960, p. 229 ; ID., Rāḥat, pp. 164-165 ; WALKER, Ḥamīd al-Dīn al-Kirmānī, pp. 76-79 ; DE SMET, Quiétude, pp. 356-360. 11. AL-FĀRĀBĪ, Mabādi’, XVII, § 1, pp. 276-277. 12. AL-KIRMĀNĪ, Rāḥat al-‘aql, pp. 383, 462, 470. 13. AL-KIRMĀNĪ, Rāḥat al-‘aql, pp. 288, 469-471, 493, 497. 14. DE SMET, « Al-Fārābī’s Influence », pp. 146-149 ; voir également ID., « La noétique ismaélienne : Abū Ya‘qūb al-Sijistānī et Ḥamīd al-Dīn al-Kirmānī », dans : M. SEBTI – D. DE SMET (éds.), Noétique et théorie de la connaissance dans la philosophie arabe du IXe au XIIe siècle. Des traductions gréco-arabes aux disciples d’Avicenne, Paris 2019, pp. 133-135. 15. AL-KIRMĀNĪ, Rāḥat al-‘aql, p. 164 ; cf. ibid., p. 363.

L’IMAM, PHILOSOPHE ET PROPHÈTE DE LA CITÉ VERTUEUSE

69

sa propre essence, l’âme devient pareille aux Intellects cosmiques, qu’elle pourra intelliger à son tour. Par cette intellection, elle entre en conjonction (ittiṣāl) avec l’Intellect, allant même jusqu’à former une union (ittiḥād) avec lui, sans toutefois se fondre complètement en lui : même dans son actualisation maximale, l’intellect de l’homme reste toujours inférieur à l’Intellect cosmique16. Tout cela, bien sûr, rappelle étroitement al-Fārābī17, à cette différence près qu’al-Kirmānī n’utilise qu’une seule fois le terme « intellect agent » (al-‘aql al-fa‘‘āl), manifestement emprunté à al-Fārābī18. Pour le philosophe ismaélien, la conjonction se fait plutôt avec le premier Intellect, le premier être instauré. Désormais libérée des passions corporelles, l’âme humaine unie à l’Intellect partage son bonheur éternel et sa quiétude : « la quiétude de l’Intellect » (rāḥat al-‘aql), qui n’est autre que l’état des bienheureux au paradis dont parlent les livres révélés19. Le paradis lui-même est le monde des Intellects séparés, qu’al-Kirmānī décrit comme une « cité idéale » avec sept portes, chaque porte donnant accès à une forteresse de lumière dont la cour intérieure renferme une source ruisselante, pourvue de sièges, de gobelets et de tapis, ainsi qu’un jardin planté de palmiers, de grenadiers et de vignes. Chaque forteresse possède une multitude de chambres dans lesquelles sont installées les âmes ; chaque chambre contient douze trônes qui portent chacun un nombre infini d’anges. La cité entière renferme ainsi en ses murs la totalité des âmes ayant atteint la félicité20. En revanche, les âmes qui n’ont pas accédé à leur seconde perfection parce qu’elles ont refusé l’enseignement des imams, se retrouvent après la mort du corps en un état de privation qui les tourmente. Al-Kirmānī rapporte à ce sujet une tradition qu’il attribue à l’imam-calife fatimide al-Ḥākim, selon laquelle les âmes ayant raté l’occasion d’acquérir la vertu et la connaissance, restent frustrées après la mort du corps, étant conscientes des bienfaits qu’elles ont manqués ainsi. Cette frustration les accable et les brûle, comme le feu torture et dévore les corps physiques : c’est 16. AL-KIRMĀNĪ, Rāḥat al-‘aql, pp. 458, 466-470, 480-481, 551-552 ; cf. DE SMET, Quiétude, pp. 368, 375. 17. AL-FĀRĀBĪ, Mabādi’, XIII, § 5, pp. 204-207 ; XVI, § 4, pp. 264-267. 18. AL-KIRMĀNĪ, Rāḥat al-‘aql, p. 245 ; cf. DE SMET, « Noétique ismaélienne », p. 153. 19. AL-KIRMĀNĪ, Rāḥat al-‘aql, pp. 506, 519, 528. 20. AL-KIRMĀNĪ, Rāḥat al-‘aql, pp. 116-117 ; cf. DE SMET, Quiétude, p. 17, n. 69.

70

DANIEL DE SMET

cela l’enfer21. Al-Fārābī conçoit lui-aussi le paradis des habitants de la cité vertueuse comme un prolongement de la cité modèle qui était la leur durant leur vie terrestre, tandis que les citoyens des cités corrompues subissent un sort qui rappelle l’enfer tel qu’al-Ḥākim l’aurait décrit à al-Kirmānī22. Les prophètes et les imams font passer, par leur enseignement, l’intellect des croyants de la puissance à l’acte, de la première à la seconde perfection. Ils ont ce pouvoir pour autant qu’ils possèdent un intellect qui est toujours en acte et qui n’a pas besoin d’être actualisé puisqu’il réunit en son essence les deux perfections, un intellect donc qui, à l’instar des Intellects cosmiques, s’intellige soi-même et intellige en même temps le premier Intellect. Tout ce qui est en puissance est imparfait et ne passe à l’acte, qui est le rang de la perfection, qu’en dépendant de ce qui est déjà en acte, parfait en son essence et en son acte. Or, les âmes humaines dans le monde de la nature sont en puissance et donc imparfaites. Dès lors, elles ne peuvent passer à l’acte que par ce qui est déjà en acte, en son essence et en son action. Parmi les âmes humaines, certaines sont déjà passées à l’acte, comme celles des prophètes, légataires et imams, ainsi que leurs successeurs, pour autant qu’elles aient acquis les deux perfections et réalisé les deux béatitudes, et qu’elles aient réuni en elles tous les bienfaits, étant exemptes de vices et parfaites. Par conséquent, c’est par celui qui est en acte, parfait en son essence et en son action, qu’ils ont acquis leur perfection et se sont élevés vers le rang de ce qui est en acte ; c’est en s’appuyant sur lui qu’ils ont acquis une existence parfaite. Sans lui, ils n’auraient pu passer à l’acte, car c’est par ce qui est déjà en acte, parfait en son essence et en son action, que l’être en puissance peut accéder à l’acte23.

L’intellect des prophètes et des imams joue ainsi le rôle de l’Intellect agent d’al-Fārābī, tout en ressemblant à celui de l’imam de la cité vertueuse qui est parvenu au stade de l’intellect acquis (al-‘aql al-mustafād) : il est un intellect (‘aql) s’intelligeant (‘āqil) soi-même et intelligé (ma‘qūl) par soi-même, tout en intelligeant l’Intellect agent24. 21. AL-KIRMĀNĪ, Rāḥat al-‘aql, p. 522. Pour un exposé plus complet de l’eschatologie complexe d’al-Kirmānī, voir D. DE SMET, « Ismā‘īlī Shī‘ī Visions of Hell. From the “Spiritual” Torment of the Fāṭimids to the Ṭayyibī Rock of Sijjīn », dans : C. LANGE (éd.), Locating Hell in Islamic Traditions, Leiden / Boston 2016, pp. 255-260. 22. AL-FĀRĀBĪ, Mabādi’, XVI, § 3-5, pp. 262-269 ; § 8, pp. 272-275. 23. AL-KIRMĀNĪ, Rāḥat, p. 160-161 ; cf. DE SMET, « Noétique ismaélienne », pp. 135, 152. 24. AL-FĀRĀBĪ, Mabādi’, XV, § 8, pp. 240-243. Rappelons que le terme al‘aql al-mustafād apparaît une seule fois chez al-Kirmānī, toujours sous l’influence d’al-Fārābī : Rāḥat, p. 552 ; cf. DE SMET, « Noétique ismaélienne », pp. 134, 152.

L’IMAM, PHILOSOPHE ET PROPHÈTE DE LA CITÉ VERTUEUSE

71

Selon al-Kirmānī, l’actualisation intégrale de l’intellect des prophètes et des imams, ainsi que la connaissance qui en découle pour eux, ne sont pas le résultat d’un enseignement (ta‘līm) qu’ils auraient reçu, mais plutôt le fruit d’un influx (fayḍ) procédant directement du monde de l’Intellect : Leur science provient des degrés (ḥudūd) supérieurs [les Intellects cosmiques] par le biais d’une inspiration (ta’yīd) continue et sans fin, comme le mouvement circulaire, alors que la science des dignitaires inférieurs est acquise par une instruction (ta‘allum) soumise à une limite25.

En d’autres termes, l’intellect des prophètes et des imams est continuellement en acte grâce à une inspiration céleste et divine (ta’yīd samawī ilāhī) qui provient des Intellects séparés, tandis que l’intellect de tous les autres hommes dépend de l’enseignement des prophètes et des imams pour passer de la puissance à l’acte. L’inspiration ou le soutien (ta’yīd, ce terme peut prendre les deux sens) dont ils bénéficient est un influx pénétrant dans leurs âmes, raison pour laquelle al-Kirmānī les appelle al-mu’ayyadūn min al-samā’, « ceux qui sont inspirés ou soutenus à partir du ciel ». Grâce à cet influx, ils sont en mesure d’intelliger de façon continue et parfaite les Intellects du monde intelligible (le ‘ālam al-ibdā‘ ou « monde de l’instauration ») auquel ils sont unis par une conjonction parfaite (ittiṣāluhu bi-‘ālam al-ibdā‘ ittiṣālan kulliyyan)26. Malgré le contexte et la terminologie ismaélienne (ta’yīd, mu’ayyad, ḥudūd), cette description de l’intellect des prophètes et des imams s’avère proche de la manière dont al-Fārābī présente l’intellect acquis du souverain de la cité vertueuse : il ne peut s’agir d’un homme comme les autres (XV, § 7), son intellect est actualisé de façon maximale ; il est intellect, intelligeant et intelligé, tout en recevant de l’Intellect agent l’ensemble des intelligibles (XV, § 8) ; « il est l’homme en qui s’infuse l’Intellect agent (huwa l-insān alladhī ḥalla fīhi al-‘aql al-fa‘‘āl) (XV, § 9)27. Certes, en tant qu’ismaélien modéré, al-Kirmānī n’aurait jamais osé employer le verbe ḥalla en ce contexte, car il évoque le ḥulūl, « l’infusion ou l’inhérence » d’un principe divin dans les prophètes et 25. AL-KIRMĀNĪ, Rāḥat, p. 346. 26. AL-KIRMĀNĪ, Rāḥat, pp. 299-300 ; cf. DE SMET, Quiétude, p. 366. 27. AL-FĀRĀBĪ, Mabādi’, XV, § 7-9, pp. 238-245.

72

DANIEL DE SMET

les imams, une thèse bien connue des shi‘ites radicaux (les ghulāt ou « exagérateurs ») dont il voulait se démarquer à tout prix28. Cela n’empêche que pour lui aussi, l’imam représente l’Intellect sur terre ; il en est en quelque sorte l’incorporation : le ‘aql mujassad ou « intellect incorporé ». D’ailleurs, un collègue d’al-Kirmānī, Ibrāhīm al-Nisābūrī (m. après 386/996), le dit explicitement : L’imam est la lueur (ḍiyā’) des âmes. De lui procède la lumière intellectuelle (al-nūr al-‘aqlī), ainsi que le mouvement et l’élévation [...]. L’imam occupe dans le monde qui est le sien la place de l’Intellect universel (al-‘aql al-kullī) [...]. La plus noble des facultés et des âmes dans l’homme est l’intellect. L’imam est l’Intellect universel dans le monde. Les gens doués de science (ahl al-‘ilm) dépendent de lui, tendent vers lui et s’unissent à lui29.

La révélation (waḥī) dont le prophète est le bénéficiaire n’est autre que l’inspiration (ta’yīd) qui rayonne sur son âme à partir du monde de l’Intellect. En se référant à Cor. 42:51 : « Il n’a pas été donné à un mortel que Dieu lui parle si ce n’est par inspiration ou derrière un voile ou bien encore en lui envoyant un messager à qui est révélé, avec sa permission, ce qu’Il veut », al-Kirmānī distingue trois modes de révélation prophétique, qu’il désigne par les termes énigmatiques al-jadd, al-fatḥ et al-khayāl. Le premier mode, celui d’al-jadd, désigne la révélation « directe » : l’inspiration coule directement de l’Esprit saint (c’est-à-dire du premier Intellect) dans l’âme du prophète, de sorte qu’il connaît les essences (ma‘ānī) des choses telles qu’elles sont. Le deuxième mode, al-fatḥ, est la révélation qui s’effectue « derrière un voile », à l’aide d’images et de symboles tirés des réalités sensibles, qui masquent les essences réelles, intelligibles. Enfin, dans le troisième cas, celui d’al-khayāl, l’âme inspirée du prophète perçoit par sa faculté imaginative (khayāl) des formes et des sons que les autres hommes ne peuvent percevoir, comme la figure angélique de Gabriel qui apparaît à Muhammad et lui dicte le Coran30. 28. D. DE SMET, « Kufr et takfīr dans l’ismaélisme fatimide : le Kitāb Tanbīh al-hādī de Ḥamīd al-Dīn al-Kirmānī », dans : C. ADANG e.a (éds.), Accusations of Unbelief in Islam. A Diachronic Perspective on Takfīr, Leiden / Boston 2016, pp. 92-97. 29. AL-NĪSĀBŪRĪ, Kitāb Ithbāt al-nubuwwa, éd. A. LALANI, Degrees of Excellence. A Fatimid Treatise on Leadership in Islam, Londres / New York 2010, pp. 16-17 et 50 (de la partie arabe). 30. AL-KIRMĀNĪ, Rāḥat, pp. 559-560, 562.

L’IMAM, PHILOSOPHE ET PROPHÈTE DE LA CITÉ VERTUEUSE

73

Dans ce passage remarquable, al-Kirmānī transforme de fond en comble l’ancienne doctrine ismaélienne de la triade al-Jadd, al-Fatḥ et al-Khayyāl, trois hypostases qui servent d’intermédiaire entre l’Âme universelle et le prophète31, en les situant à l’intérieur de l’âme prophétique, suivant en cela de près la théorie de la prophétie et des visions prophétiques d’al-Fārābī. En effet, nous reconnaissons dans le premier mode de révélation l’action de l’Intellect agent sur la faculté rationnelle du souverain de la cité vertueuse, faisant de lui un philosophe accompli. Le deuxième mode correspond à l’action de l’Intellect agent sur sa faculté représentative : en tant que prophète, il produit alors des images sensibles qui imitent parfaitement les concepts de la philosophie. Enfin, le troisième mode est celui du prophète visionnaire, dont l’imagination projette les images sensibles vers l’extérieur, de sorte qu’il voit à l’état de veille des apparitions que personne d’autre ne peut percevoir32. La sixième « lampe » (miṣbāḥ) de la première partie du Kitāb al-Maṣābīḥ établit une distinction bien nette entre la fonction du prophète et celle de l’imam. Il incombe au prophète d’exprimer des réalités qui ne sont pas accessibles aux sens, comme Dieu, les Intellects, la félicité de l’âme humaine ou sa damnation. Pour ce faire, il a recours à des images sensibles (amthila maḥsūsa), comme lorsqu’il décrit le paradis et l’enfer. Bien qu’al-Kirmānī ne mette pas en doute que Muhammad ait été le meilleur des prophètes et que ses images contiennent une sagesse profonde (ḥikma bāligha), il affirme néanmoins : « l’aspect extérieur (ẓāhir) de ce que Muhammad a apporté du Coran et de la loi (sharī‘a) est contraire aux principes de la raison ». Il en cite plusieurs exemples : des versets coraniques et des hadith dont le sens littéral est absurde, des rites comme ceux du pèlerinage à La Mecque qui semblent insensés, des prescriptions légales qui sont contraires à la justice et au bon sens. Pour cette raison, le message prophétique doit être soumis à une exégèse (ta’wīl) qui extrait le bāṭin, le sens 31. D. DE SMET, « La fonction noétique de la triade al-Jadd, al-Fatḥ et al-Khayyāl. Les fondements de la connaissance prophétique dans l’ismaélisme », dans : H. BIESTERFELDT – V. KLEMM (éds.), Differenz und Dynamik im Islam. Festschrift für Heinz Halm zum 70. Geburtstag, Würzburg 2012, pp. 319-336 ; A. STRAFACE, « The Representation of al-Jadd, al-Fatḥ and al-Khayāl in the Ismaili Literature », dans : M. A. AMIR-MOEZZI e.a. (éds.), L’ésotérisme shi‘ite, ses racines et ses prolongements, Turnhout 2016, pp. 423-440. 32. AL-FĀRĀBĪ, Mabādi’, XIV, § 8, pp. 220-223 ; XV, § 10, pp. 244-245.

74

DANIEL DE SMET

interne, du ẓāhir. Cette exégèse a pour critère absolu la raison (‘aql) : « Puisque le prophète est un sage (ḥakīm, terme qui peut aussi désigner un philosophe) qui n’est pas affecté par l’ignorance, il est impératif que tout ce qu’il a apporté ne soit pas dénué de significations (ma‘ānī) qui sont conformes aux intellects et qui sont acceptées par eux ». Le sens interne, ésotérique, et l’exégèse qui y mène relèvent donc de la raison33. Or, bāṭin et ta’wīl ont été confiés par le Prophète à son légataire ‘Alī, qui les a transmis aux imams de sa descendance. Par conséquent, dans la « cité de la connaissance », Muhammad représente le prophète et ‘Alī le philosophe. Contrairement aux ismaéliens plus radicaux, al-Kirmānī n’en conclut pas pour autant que l’imam est supérieur au prophète. Bien au contraire, dans l’exégèse du hadith sur la dite cité rapportée dans sa Risāla al-lāzima, il compare Muhammad au soleil et ‘Alī à la lune qui tient sa lumière du soleil34. Pour autant qu’il recevait directement de l’Intellect les ma‘ānī des choses, Muhammad était lui-aussi un philosophe, dans le sens fārābien du terme. Toutefois, depuis sa disparition, la présence continue d’un imam est requise, afin de prémunir la religion et le système juridique qui en découle, de toute corruption. L’imam est ainsi le garant de la rationalité, à l’aune de laquelle les pratiques religieuses et l’interprétation du Coran doivent être évaluées. Le sens interne, ésotérique, du texte révélé et de la charia, qu’il dévoile par son exégèse (ta’wīl), se traduit dans les concepts intelligibles de la philosophie. Ainsi, par exemple, si la langue imagée du Prophète nous parle du Calame, de la Tablette et des Anges rapprochés, le ta’wīl de l’imam transpose ces images ou symboles en concepts intelligibles, respectivement l’Intellect universel, l’Âme universelle et les Intellects cosmiques35. Tout cela recoupe une fois de plus ce qu’al-Fārābī nous apprend au sujet de la direction de la cité vertueuse : si elle peut éventuellement 33. Ḥamīd al-Dīn AL-KIRMĀNĪ, Kitāb al-Maṣābīḥ fī ithbāt al-imāma, éd. et trad. P. WALKER, Master of the Age. An Islamic Treatise on the Necessity of the Imamate, Londres / New York 2007, pp. 28-32 (partie arabe), 63-67 (trad.). 34. AL-KIRMĀNĪ, Risālat al-lāzima, pp. 76-77 ; cf. supra, p. 67. 35. Le Rāḥat al-‘aql contient de nombreux exemples d’exégèses de versets coraniques qui se rapportent à des thèses aristotéliciennes ou néoplatoniciennes ; voir D. DE SMET, « Philosophie grecque et religion musulmane : Aristote comme exégète du Coran selon la tradition shi‘ite ismaélienne », dans : Ishrāq. Islamic Philosophy Yearbook 2 (2011), pp. 344363.

L’IMAM, PHILOSOPHE ET PROPHÈTE DE LA CITÉ VERTUEUSE

75

se passer de prophète, elle ne pourra pas survivre sans philosophe, car il faut à tout moment un herméneute pour retraduire les images des prophètes en concepts philosophiques, dont elles ne sont que des imitations36. 3. L’imam de la cité vertueuse d’al-Fārābī Il ressort de tout ce qui précède qu’al-Kirmānī a trouvé dans les Mabādi’ ārā’ ahl al-madīna al-fāḍila une source d’inspiration pour adapter à sa façon la notion ismaélienne de la prophétie et de l’imamat au cadre conceptuel de la falsafa. Toutefois, la question se pose de savoir si cette adaptation a été facilitée par le fait que le souverain de la cité vertueuse d’al-Fārābī se présentait déjà sous les traits d’un imam shi‘ite, voire ismaélien. Il s’agit là d’une question épineuse, qui reste encore de nos jours très débattue et controversée. Les rares données biographiques dont nous disposons au sujet d’alFārābī ne permettent pas de trancher s’il était de confession sunnite ou shi‘ite, hormis le fait, peu concluant en-soi, qu’il a passé les dernières années de sa vie en Syrie, dans l’entourage de l’émir ḥamdānide Sayf al-Dawla, qui était un shi‘ite duodécimain37. Cela n’a pas empêché certains chercheurs de se prononcer pour ou contre une affiliation shi‘ite du philosophe, généralement en usant d’arguments vagues et peu probants, ce qui fait que la question reste indécise38. Dès lors, il me semble opportun de rependre la question à nouveaux frais. 36. AL-FĀRĀBĪ, Mabādi’, XV, § 14, pp. 252-253. 37. Les données biographiques sur le philosophe et la bibliographie qui s’y rapporte ont été réunies par U. RUDOLPH, « Abū Naṣr al-Fārābī », dans : U. RUDOLPH – R. HANSBERGER – P. ADAMSON (éds.), Philosophy in the Islamic World. Volume 1: 8th - 10th Centuries, Leiden / Boston 2017, pp. 536-541, 636-639 ; D. GUTAS, « Fārābī I. Biography », dans : Encyclopaedia Iranica, éd. E. YARSHATER, IX/2, New York 1999, pp. 208-213. Nous reviendrons sur Sayf al-Dawla à la fin de cet article. 38. Sans prétendre à l’exhaustivité, nous trouvons dans le camp pro-shi‘ite : H. CORBIN, Histoire de la philosophie islamique, Paris 1964, p. 223 ; WALZER, On the Perfect State, pp. 441-449 ; F. NAJJAR, « Fārābī’s Political Philosophy and Shī‘ism », dans : Studia Islamica 14 (1961), pp. 57-72 ; D. STEIGERWALD, « La pensée d’al-Fārābī. Son rapport avec la philosophie ismaélienne », dans : Laval Théologique et Philosophique 55 (1999), pp. 455476. Pour la thèse contraire, voir, entre autres, P. CRONE, « What was al-Fārābī’s “Imamic” Constitution? », dans : H. SIURUA (éd.), P. Crone. The Iranian Reception of Islam. The Non-Traditionalist Strands, Leiden / Boston 2016, p. 270 (article paru originellement dans Arabica 50 [2003], pp. 306-321) ; RUDOLPH, « Abū Naṣr », pp. 541, 630-631.

76

DANIEL DE SMET

Il est indéniable que la « cité vertueuse » d’al-Fārābī s’inspire de la République et des Lois de Platon – qu’il connaissait par les résumés de Galien ou par quelque autre voie – auxquelles s’ajoutent des éléments aristotéliciens39. Dès lors, nous reconnaissons dans le roi (malik), le premier chef (al-ra’īs al-awwal), l’imam, le philosophe (faylasūf ) et le nomothète (wāḍi‘ al-nawāmīs)40 d’al-Fārābī le Roi philosophe et législateur qui dirige la cité platonicienne. Bien sûr, ce « premier chef » prend les traits d’un prophète musulman : il reçoit une révélation (waḥī) de la part de Dieu par l’intermédiaire de l’Esprit fidèle (al-rūḥ al-amīn) ou l’Esprit saint (rūḥ al-quds)41 ; il a pour tâche d’instruire (ta‘līm) et de guider (irshād) les hommes qui sont incapables de trouver par eux mêmes la voie vers la félicité (sa‘āda)42. À cette fin, il édicte des lois religieuses (sharā’i‘), quitte à abroger les lois de ses prédécesseurs43, et il fonde une communauté religieuse (milla)44. Muhsin Mahdi a longuement insisté sur le souci d’al-Fārābī d’intégrer dans la philosophie politique de l’Antiquité l’apport de la religion musulmane. Mais dès qu’il s’agit d’indiquer des sources islamiques concrètes, le lecteur reste sur sa faim45. Rappelons que l’idée centrale de la « philosophie politique » d’alFārābī concerne l’origine commune de la philosophie et de la religion, à savoir l’Intellect agent, qui agit sur la faculté rationnelle du philosophe en produisant des concepts abstraits et sur la faculté imaginative du prophète en produisant des images qui imitent les concepts de la philosophie. Ces deux qualités, celle du philosophe et celle du 39. Il existe une vaste littérature sur ce sujet ; voir, entre autres, M. MAHDI, La fondation de la philosophie politique en islam. La cité vertueuse d’Alfarabi. Traduit par François Zabbal, Paris 2000, pp. 172-231 ; H. DAIBER, The Ruler as Philosopher. A New Interpretation of al-Fārābī’s View (Mededelingen der Koninklijke Nederlandse Akademie van Wetenschappen. Afd. Letterkunde, Nieuwe Reeks, 49/4), Amsterdam / Oxford / New York 1986. 40. AL-FĀRĀBĪ, Taḥṣīl al-sa‘āda, éd. ‘A. BŪ MALḤUM, Beyrouth 1995, pp. 92-94, où il est dit explicitement que ces cinq qualificatifs ont le même sens (ma‘nā wāḥid) et désignent une même personne. 41. AL-FĀRĀBĪ, Kitāb al-Siyāsa al-madaniyya, éd. F. NAJJAR, Beyrouth 1993, p. 32 (spécifiant qu’il s’agit de deux dénominations de l’Intellect agent) ; ID., Kitāb al-Milla. Texte, traduction et commentaire par A. CHERNI, Paris 2012, pp. 38-39, 108-113. 42. AL-FĀRĀBĪ, al-Siyāsa, pp. 75-76, 78. 43. AL-FĀRĀBĪ, al-Siyāsa, pp. 80-81. 44. AL-FĀRĀBĪ, Milla, pp. 30-31. 45. Voir le recueil d’articles de Muhsin Mahdi cité n. 39.

L’IMAM, PHILOSOPHE ET PROPHÈTE DE LA CITÉ VERTUEUSE

77

prophète, sont réunies dans la personne du souverain de la cité vertueuse. Pour autant que je puisse voir, cette manière spécifique de concevoir l’origine commune de la philosophie et de la religion n’apparaît nulle part dans la pensée musulmane antérieure à al-Fārābī, si ce n’est chez des auteurs ismaéliens ou inspirés par ce courant shi‘ite. Cela a très bien été vu par Hans Daiber. a. Abū Ḥātim al-Rāzī Daiber a en effet découvert qu’un certain nombre d’idées développées par al-Fārābī sur l’imam-philosophe se retrouvent dans le Kitāb A‘lām al-nubuwwa du propagandiste ismaélien Abū Ḥātim al-Rāzī (m. 322/934). Il en conclut, un peu trop rapidement sans doute, qu’alFārābī avait été, du moins pendant une période de sa vie, gagné à la cause ismaélienne46. Voyons cela de plus près. Le Kitāb A‘lām al-nubuwwa rapporte un débat public (majlis) qui aurait eu lieu entre le philosophe-médecin « libre penseur » Abū Bakr al-Rāzī (m. 313/925 ou 323/935) et son concitoyen ismaélien, Abū Ḥātim al-Rāzī47. Abū Bakr y soutient que l’homme, grâce à son intellect inné, est capable de percer les secrets de l’univers et qu’il n’a nul besoin de révélations ou de prophètes pour arriver à cette fin. Le fait que ces révélations et les livres qui les contiennent (Thora, Évangile, Coran) se contredisent, prouverait que les prophètes qui les ont rédigés sont des imposteurs. Abū Ḥātim répond à cette attaque frontale contre les religions et leurs prophètes en invoquant l’argument de la disparité qui règne au sein de l’espèce humaine. Si les animaux appartenant à une même 46. H. DAIBER, « The Ismaili Background of Fārābī’s Political Philosophy. Abū Ḥātim ar-Rāzī as a Forerunner of Fārābī », dans : U. TWORUSCHKA (éd.), Gottes ist der Orient. Gottes ist der Okzident. Festschrift für Abdoldjavad Falaturi, Cologne 1991, pp. 143-150 ; ID., « Abū Ḥātim ar-Rāzī (10th century A. D.) on the Unity and Diversity of Religions », dans : J. GORT (éd.), Dialogue and Syncretism. An Interdisciplinary Approach, Grand Rapids / Amsterdam 1989, pp. 87-104 ; sur Abū Ḥātim al-Rāzī, voir D. DE SMET, « The Religious Application of Philosophical Ideas », dans : RUDOLPH e.a. (éds.), Philosophy, pp. 743-744. 47. Il n’est pas clair si ce débat a réellement eu lieu ou s’il s’agit d’une fiction littéraire ; voir D. DE SMET, « From Khalaf to Ḥasan al-Ṣabbāḥ : Ismailism in Rayy before and under the Seljūqs », dans : Der Islam 93 (2016), p. 444 ; sur Abū Bakr al-Rāzī, voir, entre autres, S. STROUMSA, Freethinkers of Medieval Islam. Ibn al-Rāwandī, Abū Bakr al-Rāzī, and Their Impact on Islamic Thought, Leiden / Boston / Cologne 1999, pp. 87-120.

78

DANIEL DE SMET

espèce sont tous plus ou moins pareils, il en est tout autre des hommes, dont certains sont faibles, soumis et idiots, tandis que d’autres sont forts, dominants et intelligents. L’enseignant (‘ālim) s’y distingue de l’enseigné (muta‘allim), le dirigeant (imām) du dirigé (ma’mūm). Par conséquent, l’humanité a besoin de guides qui ont reçu une révélation divine48. Al-Fārābī tient un raisonnement similaire dans sa Siyāsa al-madaniyya : chaque homme n’ayant pas la disposition innée (fiṭra) lui permettant de connaître par soi-même les voies menant à la félicité, il faut un enseignant (mu‘allim) et un guide (murshid) ; l’homme qui, par nature, est disposé à guider et qui en reçoit l’inspiration de l’Intellect agent, est un chef (ra’īs), un imam49. D’après Abū Ḥātim, cet imam peut être un philosophe, car « le sage Platon était un imam pour Aristote » (Aflāṭūn al-ḥakīm kāna imāman li-Arisṭāṭālīs)50. Les contradictions relevées par Abū Bakr entre, d’une part, les différentes révélations prophétiques et, de l’autre, la science des philosophes, s’expliquent par le fait que les prophètes s’expriment à l’aide d’images (muthul) et de symboles (rumūz). Si le contenu (le « sens », ma‘nā) du message qu’ils ont reçu par révélation est toujours le même et reste inchangé, les prophètes successifs l’ont traduit dans un langage imagé qui convient au milieu historique du peuple auquel ils étaient envoyés51. Nous reconnaissons dans cette idée centrale du livre d’Abū Ḥātim une thèse caractéristique de la prophétologie d’al-Fārābī : grâce à l’action de l’Intellect agent sur leur faculté imaginative, les prophètes produisent des images et des symboles, chacun à sa façon, de sorte qu’il peut y avoir des cités vertueuses avec des religions différentes qui, toutes, conduisent à la félicité52. Ce qui vaut pour les livres révélés par les prophètes, observe Abū Ḥātim, vaut également pour les livres des philosophes. Eux aussi 48. Abū Ḥātim AL-RĀZĪ, Kitāb A‘lām al-nubuwwa, éd. Ṣ. AL-ṢĀWĪ – G. A‘WĀNĪ, Téhéran 1977, pp. 6-8, 185, 301 ; traduction anglaise par T. KHALIDI, The Proofs of Prophecy, Provo 2011, pp. 4-6, 134, 228. La plupart des parallèles avec al-Fārābī qui vont suivre, ont déjà été relevés par Daiber dans les articles cités n. 46, mais je les présente ici dans un ordre différent. 49. AL-FĀRĀBĪ, al-Siyāsa, pp. 75-76, 78. 50. AL-RĀZĪ, A‘lām, p. 7 (éd.), 5 (trad.). 51. AL-RĀZĪ, A‘lām, pp. 71, 104-105, 108, 155-156, 173 (éd.), 52, 77-78, 80, 113115, 127 (trad.). 52. AL-FĀRĀBĪ, al-Siyāsa, pp. 85-86 ; ID. Taḥṣīl, p. 85, 88-90 ; ID., Mabādi’, XV, § 8, 10-11, XVII, § 2, pp. 240-247, 278-281.

L’IMAM, PHILOSOPHE ET PROPHÈTE DE LA CITÉ VERTUEUSE

79

contiennent des passages obscurs qui demandent à être expliqués par un maître qualifié. Les philosophes ont eu recours à des images et des mythes, comme en témoignent les œuvres de Platon. Puisque les vrais philosophes (al-ḥukamā’ al-muḥiqqūn) sont des êtres humains avec une intelligence supérieure au même titre que les prophètes, il est virtuellement impossible qu’ils aient tenu des propos contradictoires ou irrationnels. Par conséquent, lire Platon ou Aristote sans guide est tout aussi absurde que d’essayer d’interpréter le Coran au gré de sa propre fantaisie53. Pour contrecarrer l’assertion d’Abū Bakr selon laquelle les philosophes ont découvert leur science de par eux-mêmes, par déduction (istinbāṭ), Abū Ḥātim prône au contraire que les prophètes sont à la base des enseignements des philosophes. Ou plus précisément, que les sciences philosophiques, géométriques, astronomiques ou médicales ont été instituées par les véritables imams de leur époque, les « preuves » (ḥujaj) de Dieu sur terre qui ont reçu leur science par révélation. L’exemple d’Hermès, que la tradition islamique identifie avec le prophète Idrīs mentionné dans le Coran, doit prouver cette symbiose : il s’agit du même personnage, appelé Hermès en sa qualité de philosophe et Idrīs en sa qualité de prophète54. La science ayant été transmise par une lignée continue d’imams philosophes, Abū Ḥātim fait commencer la chaîne avec le « premier sage » (al-ḥakīm al-awwal) : Ce sage a connu ces sciences subtiles par une inspiration (ta’yīd) de Dieu et par une révélation (waḥī) de sa part. Il fait partie de l’ensemble des prophètes, car personne n’aurait pu atteindre par une autre voie une connaissance d’une telle ampleur55.

Ce ḥakīm al-awwal rappelle étroitement le ra’īs al-awwal d’al-Fārābī, le « premier chef » ou premier imam dont dépendent les souverains des cités vertueuses qui lui ont succédé et qui, lui aussi, était à la fois prophète et philosophe56. Les ressemblances sont si frappantes que l’on peut conclure avec une certaine probabilité qu’al-Fārābī avait lu le Kitāb A‘lām al-nubuwwa. 53. AL-RĀZĪ, A‘lām, pp. 72-73, 107 (éd.), 52-53, 79 (trad.). 54. AL-RĀZĪ, A‘lām, pp. 273, 277-278 (éd.), 205-206, 209-210 (trad.). 55. AL-RĀZĪ, A‘lām, p. 318 (éd.), 242 (trad.). 56. AL-FĀRĀBĪ, Siyāsa, pp. 78-80 ; ID., Milla, pp. 38-39, 104-105 ; ID., Taḥṣīl, pp. 78, 88, 94.

80

DANIEL DE SMET

Cela ne prouve pas pour autant qu’il était ismaélien ou qu’il avait des sympathies pour l’ismaélisme. En effet, la tradition ismaélienne considère ce livre comme ẓāhir (« exotérique »), pouvant circuler librement en dehors des circuits internes, ce qu’il a visiblement fait, puisqu’il a connu une large diffusion dans le monde musulman. Aussi, ne contient-il aucun exposé explicite de doctrines ismaéliennes « secrètes », ce qui ne veut pas dire pour autant qu’elles en soient absentes. Il est sûr, par exemple, que l’auteur emploie le terme imām dans un sens ismaélien, tout comme ta’yīd ou ta‘līm. Mais un lecteur profane n’est pas censé le remarquer57. Sans doute al-Fārābī a-t-il été séduit par l’idée de l’origine prophétique de la philosophie, une thèse qui remonte à la Doxographie du Pseudo-Ammonius rédigée dans le cercle d’al-Kindī et dont le Kitāb A‘lām al-nubuwwa est notre plus ancien témoin58. b. Les Frères de la pureté (Ikhwān al-Ṣafā’) Dès 1962, Yves Marquet avait détecté un nombre de similitudes frappantes entre la cité vertueuse d’al-Fārābī et la « philosophie politique » des Ikhwān al-Ṣafā’. Plus récemment, Carmela Baffioni a confirmé ces ressemblances59. Dans les dernières épîtres de leur « Encyclopédie » (en particulier les Ép. 42 à 48), les Ikhwān décrivent leur cité idéale, qu’ils qualifient de madīna al-fāḍila60. Mais contrairement à al-Fārābī, ils présentent 57. Il est significatif à cet égard que Tarif Khalidi, dans sa traduction anglaise, semble ignorer (délibérément ou non ?) tout apport ismaélien, et plus généralement shi‘ite. Sa manière de rendre les termes techniques est dès lors souvent sujette à caution. 58. Voir à ce sujet D. DE SMET, « La Doxographie du Pseudo-Ammonius dans ses rapports avec le néoplatonisme ismaélien », dans : E. CODA – C. MARTINI BONADEO (éds.), De l’Antiquité tardive au Moyen Âge. Études de logique aristotélicienne et de philosophie grecque, syriaque, arabe et latine offertes à Henry Hugonnard-Roche, Paris 2014, pp. 491-518. 59. Y. MARQUET, « Imamat, résurrection et hiérarchie selon les Ikhwan as-Safa », dans : Revue des Études Islamiques 30 (1962), pp. 49-142 (les parallèles avec al-Fārābī ont été relégués dans les notes de bas de page) ; C. BAFFIONI, « Al-Madīnah al-fāḍila in al-Fārābī and in the Ikhwān al-Ṣafā’: a Comparison », dans : S. LEDER (éd.), Studies in Arabic and Islam. Proceedings of the 19th Congress, Union Européenne des Arabisants et Islamisants, Halle 1998, Louvain 2002, pp. 3-12. Notons que les rapprochements suggérés par ces auteurs ne sont pas tous aussi convaincants. Je ne reprendrai que ceux qui me semblent les plus significatifs. 60. L’Épître 44 raconte l’histoire d’une madīna al-fāḍila bâtie sur le sommet d’une île au milieu des océans et dont les habitants vivaient dans une harmonie parfaite, loin de

L’IMAM, PHILOSOPHE ET PROPHÈTE DE LA CITÉ VERTUEUSE

81

cette « cité vertueuse » comme une cité spirituelle (madīna fāḍila rūḥāniyya), qui ne doit pas être construite sur terre, afin de la prémunir contre l’influence néfaste des cités injustes (al-mudun al-jā’ira), ni près du rivage pour la protéger des vagues dévastatrices, mais haut dans le ciel, hors de la portée des miasmes des cités iniques61. Les frères qui y habitent sont si unis qu’ils ne forment « qu’un seul homme, un seul corps, une seule âme », expression qui revient comme un refrain sous la plume des Ikhwān, mais que l’on trouve également chez al-Fārābī62. L’accès à la cité vertueuse, qui est protégée par quatre remparts, est réservé aux frères qui ont acquis les sciences nécessaires à la félicité (sa‘āda), à savoir l’ensemble des sciences décrites dans les Rasā’il Ikhwān al-Ṣafā’ 63. Cette cité a été construite « dans le royaume du maître suprême de la loi » (fī mamlakat ṣāḥib al-nāmūs al-akbar)64. Il s’agit du législateur (wāḍi‘ al-sharī‘a ou wāḍi‘ al-nāmūs, au pluriel al-nawāmīs)65, fondateur de la cité vertueuse. Selon les Ikhwān, les prophètes (al-anbiyā’) occupent le rang le plus élevé parmi les hommes, proches des anges ; ils sont immédiatement suivis par les sages philosophes (al-falāsifa al-ḥukamā’). Les deux groupes s’accordent sur les mêmes doctrines, comme l’unicité du Créateur, la nature générée de tous les existants ou le retour (ma‘ād) de l’âme sauvée vers son origine céleste. La prophétie suppose toute corruption et discorde ; voir S. TRABOULSI – T. MAYER – I. NETTON, Epistles of the Brethren of Purity, On Companionship and Belief. An Arabic Critical Edition and English Translation of Epistles 43-45, Oxford 2016, pp. 86-87 (éd.), 108-109 (trad.). 61. A. HAMDANI – A. SOUFAN, Epistles of the Brethren of Purity, The Call to God. An Arabic Critical Translation of Epistle 48, Oxford 2019, pp. 59-62 (éd.), 82 (trad.) ; cf. Ibid., p. 48 (éd.), 79 (trad.). 62. Ainsi, p. ex., TRABOULSI e.a., Companionship, p. 113 (éd.), 124 (trad.) (Ép. 44) ; HAMDANI e.a., Call, p. 55 (éd.), 81 (trad.) ; IKHWĀN AL-ṢAFĀ’, Rasā’il, Beyrouth 1958, IV, p. 127 (présenté ici comme un hadith du Prophète : « les croyants sont comme un seul homme et une seule âme »), 134 (Ép. 47) ; cf. AL-FĀRĀBĪ, Mabādi’, XVI, § 1, pp. 258-260. 63. HAMDANI e.a., Call, pp. 66-67 (éd.), 83-84 (trad.). Outre la ressemblance avec les murailles de la cité de la connaissance décrite par al-Kirmānī (voir supra), l’idée que l’obtention de la félicité est liée à la connaissance exposée dans le livre (en ce cas les Mabādi’), se retrouve dans AL-FĀRĀBĪ, Mabādi’, XVII, § 1, pp. 276-279. L’Ép. 45 contient un faṣl sur la sa‘āda (TRABOULSI e.a., Companionship, pp. 115-117 [éd.], 126-127 [trad.]), mais outre le terme il n’y a que peu de ressemblances avec al-Fārābī. 64. HAMDANI e.a., Call, p. 59 (éd.), 82 (trad.). La traduction « in the kingdom of the Great Divinity » est trompeuse et erronée. 65. Al-Fārābī emploie couramment ces mêmes termes ; voir, p. ex., Taḥṣīl, pp. 91-94.

82

DANIEL DE SMET

quarante-six qualités (khaṣla), dont la première est la vision véridique (al-ru’ya al-ṣādiqa), d’après un hadith attribué au Prophète : « La vision véridique est la première des quarante-six parties de la prophétie ». Si, à un certain moment, ces quarante-six qualités sont réunies en un seul homme, il sera l’envoyé (al-mab‘ūth), le maître du temps (ṣāḥib al-zamān) et l’imam des hommes. Après sa mort, son successeur (khalīfa) réunira la plupart de ces qualités. Si une telle personne n’est pas disponible, mais les qualités sont dispersées sur un groupe d’individus, alors ce groupe gouvernera collectivement la cité et coopèrera pour préserver la religion et la loi66. En des termes très analogues, al-Fārābī admet qu’après la disparition du premier chef, la cité vertueuse peut être dirigée par un « second chef » ou, à défaut, par un collectif de gouvernants qui portent chacun une partie des qualités requises67. Le législateur (wāḍi‘ al-sharī‘a) doit être doté de douze qualités innées, qui sont décrites les unes après les autres68. Une liste similaire (mais avec quelques modifications) figure dans les Mabādi’ d’alFārābī69, alors que le Taḥṣīl al-sa‘āda, qui reprend la liste, l’attribue explicitement à la République de Platon70. Outre ces qualités, le législateur doit maîtriser dix bases (qā’ida) de connaissances, concernant Dieu, la création, les intellects, les âmes, la nature de l’âme humaine 66. IKHWĀN, Rasā’il, IV, pp. 124-125 ; HAMDANI e.a., Call, pp. 81-82 (éd.), 88-89 (trad.). 67. AL-FĀRĀBĪ, Siyāsa, pp. 78-81 ; ID., Milla, pp. 60-65, 84-87 ; ID., Mabādi’, XV, § 13-14, pp. 248-253. 68. IKHWĀN, Rasā’il, IV, pp. 129-130. Le rapport avec les quarante-six qualités du prophète mentionnées précédemment, mais non spécifiées, ne ressort pas clairement du texte. Conformément à leur habitude, les Ikhwān semblent avoir juxtaposé deux traditions différentes, sans prendre le soin de les harmoniser. Une allusion à ces douze qualités apparaît également dans l’Ép. 48 ; voir HAMDANI e.a., Call, pp. 97-98 (éd.), 93-94 (trad.). 69. AL-FĀRĀBĪ, Mabādi’, XV, § 12, pp. 246-249 ; le texte précise que ces qualités sont présentes dans le chef de la cité vertueuse « par nature » (bi l-ṭab‘). Pour une comparaison détaillée entre les deux listes, voir MARQUET, « Imamat », pp. 49-50 et n. 2. 70. AL-FĀRĀBĪ, Taḥṣīl, pp. 95-96. WALZER, On the Perfect State, pp. 445-446, montre que cette liste est compilée à partir des premiers chapitres du livre VI de la République, par le biais d’une paraphrase remontant à Galien et qui est citée dans le Commentaire de la République d’Ibn Rushd, dont seul une traduction en hébreu nous est parvenue ; voir R. LERNER, Averroes on Plato’s Republic, Ithaca / Londres 1974, pp. 73-75. Les trois listes (Ikhwān, al-Fārābī et Ibn Rushd) ont été comparées par A. HAMDANI, « The Ikhwān al-Ṣafā’: Between al-Kindī and al-Fārābī », dans : O. ALI-DE-UNZAGA (éd.), Fortresses of the Intellect. Ismaili and Other Islamic Studies in Honour of Farhad Daftary, Londres / New York 2011, p. 200.

L’IMAM, PHILOSOPHE ET PROPHÈTE DE LA CITÉ VERTUEUSE

83

et son retour. Il tient cette connaissance par révélation (waḥī), étant en contact avec l’Intellect (‘aql), par l’intermédiaire de l’Âme universelle et des anges71. Il expose sa science directement, mais en secret, à l’élite (visiblement les philosophes), sans faire usage de symboles ou sans faire appel à la dissimulation (ghayr marmūz wa lā maktūm). En revanche, quand il s’adresse au vulgaire, il s’exprime de façon symbolique, en n’usant que d’énoncés (alfāẓ) et de notions (ma‘ānī) qu’ils sont susceptibles de comprendre. Dès lors, si la science que les législateurs dissimulent de cette façon est toujours la même, les images et les symboles diffèrent d’une époque à l’autre et d’une culture à l’autre, une thèse que nous avons déjà relevée chez al-Fārābī et Abū Ḥātim al-Rāzī72. Le prophète opère ainsi comme le médecin de la cité, dont la loi est le médicament qui guérit les âmes malades73. À défaut de prophète, il faut que des « intellectuels » (‘uqalā’) gouvernent la cité, pour préserver les lois contre toute tentative visant à les altérer 74. Or, ces « intellectuels » ne sont autres que les philosophes, dont nous avons vu qu’ils occupent un rang immédiatement inférieur aux prophètes. Manifestement, les prophètes étaient eux aussi philosophes et certains philosophes prophètes, comme en témoigne ce hadith dans lequel Muhammad aurait déclaré : « Je suis l’Aristote de cette communauté » (anā Arisṭāṭālīs hādhihi l-ummati)75. Une indéniable parenté se dessine ainsi entre al-Fārābī et les Ikhwān al-Ṣafā’. Cependant, déterminer lequel des deux dépend de l’autre, est une question délicate, liée au problème de la datation du corpus ikhwānien. La position dominante encore aujourd’hui accorde foi au témoignage d’Abū Ḥayyān al-Tawḥīdī selon lequel les Rasā’il auraient été composées par un groupe de clercs à Baṣra entre 360/970 et 370/98076, donc après la mort d’al-Fārābī, ce qui signifie que les 71. IKHWĀN, Rasā’il, IV, pp. 127, 129-132, 136. 72. IKHWĀN, Rasā’il, IV, p. 132 ; HAMDANI e.a., Call, p. 97 (éd.), 93 (trad.). 73. HAMDANI e.a., Call, p. 237 (éd.), 133 (trad.). Al-Fārābī lui-aussi affectionne cette comparaison avec le médecin ; voir Milla, pp. 86-87 ; Taḥṣīl, p. 98 ; Siyāsa, p. 83. 74. IKHWĀN, Rasā’il, IV, p. 137. De même pour AL-FĀRĀBĪ, Mabādi’, XV, § 14, pp. 252253, la cité vertueuse ne peut survivre sans la présence d’un philosophe. 75. IKHWĀN, Rasā’il, IV, p. 263. 76. Pour un état de la question de la datation des Rasā’il, voir DE SMET, « Religious Application », pp. 752-754 ; G. DE CALLATAŸ, Ikhwan al-Safa’. A Brotherhood of Idealists on the Fringe of Orthodox Islam, Oxford 2005, pp. 3-4.

84

DANIEL DE SMET

Ikhwān seraient tributaires du philosophe77. Yves Marquet, en revanche, croyait que la majorité des Épîtres ont été écrites entre 350/961 et 370/980, bien que certaines (dont celles relatives à la cité vertueuse) remonteraient à une période plus ancienne, même antérieure à l’avènement des Fatimides en Tunisie au début du 10e siècle. Par conséquent, al-Fārābī dépendrait des Ikhwān ou aurait employé les mêmes sources qu’eux78. Wilferd Madelung a récemment émis une hypothèse complexe, qui situe la rédaction des Rasā’il dans la première moitié du 10e siècle79 : les Ikhwān et al-Fārābī seraient alors contemporains. Enfin, une datation pré-fatimide est défendue par Abbas Hamdani. Certes, les nombreuses publications de ce chercheur ismaélien poursuivent un « agenda caché », qui consiste à prouver la thèse ismaélienne tardive selon laquelle les Épîtres auraient été écrites par l’imam Aḥmad, petit-fils du 7e imam Muḥammad b. Ismā‘īl et grand-père de ‘Abd Allāh al-Mahdī (m. 322/934), le premier calife fatimide, selon la généalogie officielle80. Si les arguments invoqués par Hamdani font feu de tout bois et s’avèrent parfois hautement improbables, il a raison d’observer que le vocabulaire philosophique des Ikhwān est plus proche d’al-Kindī et 77. Déjà S. PINÈS, « Philosophy », dans : The Cambridge History of Islam, Cambridge 1970, II, p. 804, soutenait que les Ikhwān ont plagié al-Fārābī, notamment en copiant leur liste avec les douze qualités de l’imam. Il a été suivi, e.a., par WALZER, On the Perfect State, p. 446 ; H. ENAYET, « An Outline of the Political Philosophy of the Rasā’il of the Ikhwān al-Ṣafā’ », dans : S. H. NASR (éd.), Ismā‘īlī Contributions to Islamic Culture, Téhéran 1977, pp. 42-44 ; RUDOLPH, « Abū Naṣr al-Fārābī », p. 542. La liste des douze qualités de l’imam est souvent invoquée comme « preuve » de la dépendance des Ikhwān envers al-Fārābī. Mais, en réalité, elle ne prouve rien : les deux listes ne sont pas identiques et les citations platoniciennes – authentiques et fictives – abondent dans les Rasā’il. Ainsi, dans l’Ép. 47, quelques pages après la fameuse liste, les Ikhwān citent Timée 90a7-b2, le célèbre passage sur l’homme comme un arbre renversé qui prend racine dans le ciel, sans aucun rapport avec al-Fārābī (IKHWĀN, Rasā’il, IV, p. 136). Les Ikhwān auraient donc pu trouver la liste dans leurs sources platoniciennes. Pour un inventaire des citations, qui pourraient provenir d’une anthologie, voir C. BAFFIONI, Frammenti e testimonianze di autori antichi nelle Rasā’il degli Iḫwān al-Ṣafā’, Rome 1994. 78. Y. MARQUET, « La procession des âmes chez Fārābī ou le dédoublement de l’itinéraire », dans : Journal asiatique 275 (1987), p. 65 ; ID., « Imamat », p. 50 n. 2. 79. W. MADELUNG, « Maslama al-Qurṭubī’s Contribution to the Shaping of the Encyclopaedia of the Ikhwān al-Ṣafā’ », dans : A. STRAFACE – C. DE ANGELO – A. MANZO (éds.), Labor Limae. Atti in onore di Carmela Baffioni, 2 vols., Naples 2014, I, pp. 403-417. 80. Sur cette thèse, voir D. DE SMET, « L’auteur des Rasā’il Ikhwān al-Ṣafā’ selon les sources ismaéliennes ṭayyibites », dans : Shii Studies Review 1 (2017), pp. 151-166 ; pour la généalogie officielle des Fatimides, voir F. DAFTARY, The Ismā‘īlīs. Their History and Doctrines. Second Edition, Cambridge 2007, p. 507.

L’IMAM, PHILOSOPHE ET PROPHÈTE DE LA CITÉ VERTUEUSE

85

des écrits néoplatoniciens élaborés dans son cercle (comme la PseudoThéologie d’Aristote) que d’al-Fārābī81. Cela vaut en particulier dans le domaine de la « philosophie politique » : ce qui fait l’originalité des exposés d’al-Fārābī sur la cité vertueuse et son imam, est le cadre noétique aristotélicien, avec des termes comme Intellect agent, intellect acquis, conjonction, etc. Or ces termes, qui n’apparaissent qu’occasionnellement dans l’ensemble des Épîtres82, font totalement défaut dans les Ép. 47 et 48 qui sont centrales pour notre sujet. Ici, la terminologie est néoplatonicienne, proche du néoplatonisme ismaélien antérieur à al-Kirmānī. Si les Ikhwān dépendaient d’al-Fārābī, on peut se demander pourquoi ils n’aient rien emprunté à son vocabulaire technique. Certainement pas par souci de cohérence, car les Ikhwān, dans leur brassage éclectique, sont tout sauf cohérents. Peut-on en conclure qu’ils ne connaissaient pas les écrits politiques d’al-Fārābī, qui sont généralement considérés comme ayant été rédigés vers la fin de sa vie ? Le rapport entre al-Fārābī et les Ikhwān al-Ṣafā’ reste donc indécis, bien que je penche plutôt en faveur d’une dépendance d’al-Fārābī envers les Ikhwān, sans pouvoir apporter de preuve décisive. Ce qui, en revanche, apparaît clairement de notre étude, est le fait qu’alFārābī avait une certaine affinité intellectuelle avec ces milieux shi‘ites imbibés de philosophie antique, quitte à leur emprunter l’idée d’un imam, à la fois philosophe et prophète, qui dirige la cité de la connaissance préparant à la félicité éternelle. Cette affinité intellectuelle explique en même temps pourquoi ses ouvrages politiques ont influencé des ismaéliens comme al-Kirmānī. Toutefois, elle n’implique pas nécessairement qu’al-Fārābī ait été lui-même ismaélien. Comme le Kitāb A‘lām al-nubuwwa d’Abū Ḥātim al-Rāzī, les Rasā’il Ikhwān al-Ṣafā’ ne faisaient pas partie de la littérature interne de la da‘wa ismaélienne à son époque et avaient une diffusion plus large que le cercle restreint des initiés83. 81. HAMDANI, « The Ikhwān al-Ṣafā’ », pp. 189-212. 82. En des passages qui pourraient être des interpolations, du genre : « L’Intellect, à savoir l’Intellect agent ... ». Mais, il n’y a pas de théorie de l’Intellect agent dans les Rasā’il, leur noétique étant néoplatonicienne, avec des termes comme Intellect universel, Âme universelle, l’âme humaine comme partie de l’Âme universelle, etc. ; voir D. DE SMET, « Les Ikhwān al-Ṣafā’ (Frères de la Pureté) », dans SEBTI – DE SMET, Noétique, pp. 159-189. 83. De surcroît, le rapport des Épîtres avec l’ismaélisme est un sujet tout aussi controversé que leur datation. Disons pour simplifier qu’ils sont issus d’un même milieu intellectuel.

86

DANIEL DE SMET

4. Les Mabādi’ ārā’ ahl al-madīna al-fāḍila : un « Fürstenspiegel » pour période d’occultation (ghayba) ? Je ne peux résister à la tentation de clore cet article avec une hypothèse qui viendra s’ajouter aux innombrables hypothèses qui ont déjà été émises au sujet de la « philosophie politique » d’al-Fārābī. Notre philosophe serait né vers 256/870 et il est mort en 339/95084. Par conséquent, la plus grande partie de sa vie coïncide avec la « petite occultation » (al-ghayba al-ṣughrā) du douzième imam shi‘ite Muḥammad al-Mahdī, qui commence en 260/874 avec la mort de son père al-Ḥasan al-‘Askarī, et se termine avec la proclamation de l’« occultation majeure » (al-ghayba al-kubrā) en 329/941 : désormais, l’imam n’a plus de représentant (safīr) sur terre, tout lien avec lui est coupé jusqu’à sa réapparition comme sauveur messianique85. Cette interruption de la succession des imams jeta la communauté shi‘ite dans le désarroi : beaucoup doutaient, certains se ralliaient à la branche rivale des ismaéliens, tandis que d’autres devinrent sunnites. La grande ghayba posait ouvertement la question du pouvoir religieux. L’imam étant la seule autorité en matière de droit et d’interprétation du Coran et de la Tradition, que faire en son absence86 ? Al-Fārābī a commencé la rédaction des Mabādi’ à Bagdad en 330/942 et il l’acheva à Damas en 331/94387. En Syrie, plus précisément à Alep, il vécut sous la protection (fī kanaf) de l’émir Sayf al-Dawla, qui appartenait à la famille shi‘ite duodécimaine des Ḥamdānides88. Le livre a 84. RUDOLPH, « Abū Naṣr al-Fārābī », pp. 537-538. 85. H. HALM, Die Schia, Darmstadt 1988, pp. 41-47. 86. Cette question continue à diviser les shi‘ites duodécimains jusqu’à aujourd’hui. Certains théologiens contestent en effet la légitimité du régime islamique en Iran, en prônant que le clergé shi‘ite a usurpé des pouvoirs qui ne reviennent qu’à l’imam. Selon eux, tant que dure la ghayba, il ne peut y avoir qu’un pouvoir exercé par un état laïc ; voir M. A. AMIR-MOEZZI – C. JAMBET, Qu’est-ce que le shî’isme ?, Paris 2004, pp. 207-220. 87. GUTAS, « Fārābī i. Bibliography », p. 210, basé sur IBN ABĪ UṢAYBI‘A, ‘Uyūn al-anbā’ fī ṭabaqāt al-aṭibbā’, éd. N. RIḌĀ, Beyrouth, s.d., p. 608, ainsi que sur les notes marginales de certains manuscrits (WALZER, On the Perfect State, p. 20). Selon ces mêmes notes, toujours confirmés par Ibn Abī Uṣaybi‘a, al-Fārābī aurait achevé la rédaction définitive de son livre lors d’un voyage en Égypte en 337 ou 338. 88. SĀ‘ID AL-ANDALUSĪ, Ṭabaqāt al-umam, éd. Ḥ. MU’NIS, Le Caire 1998, p. 74 ; AL-QIFṬĪ, Ta’rīkh al-ḥukamā’, éd. J. LIPPERT, Leipzig 1903, p. 279 ; IBN ABĪ UṢAYBI‘A, ‘Uyūn al-anbā’, p. 603. Les adversaires de la « thèse shi‘ite » invoquent volontiers l’argument qu’al-Fārābī ne peut avoir été invité par Sayf al-Dawla à cause de ses sympathies shi‘ites, pour la simple raison que le philosophe quitta Bagdad en 331/942-3, alors que

L’IMAM, PHILOSOPHE ET PROPHÈTE DE LA CITÉ VERTUEUSE

87

donc été écrit peu après le début de la grande ghayba. En supposant qu’al-Fārābī était de confession shi‘ite duodécimaine et qu’il ne vivait pas dans une tour d’ivoire, déconnecté de l’actualité de son époque, il ne pouvait pas ne pas être au courant des débats soulevés par l’occultation majeure du douzième imam89. Or, à mon avis, cette problématique a laissé des traces dans les Mabādi’. En effet, l’auteur déclare que l’imam de la cité vertueuse n’est pas un homme comme les autres, puisqu’il doit réunir de naissance et par nature les douze qualités que nous avons déjà évoquées (XV, § 11-12). De tels hommes étant très rares, ils n’apparaissent « que l’un après l’autre » (al-wāḥid ba‘da l-wāḥid) (XV, § 13). Visiblement, il s’agit des prophètes législateurs, dont nous avons vu qu’ils étaient, tout comme pour Abū Ḥātim al-Rāzī et les Ikhwān al-Ṣafā’, en même temps des philosophes. Si un tel homme n’est pas disponible, il est remplacé par des souverains doués des mêmes qualités, hormis le pouvoir « d’énoncer un message à partir de la faculté imaginative ». Ces chefs, qui n’ont donc pas le don de la prophétie, se succèdent (tawāllū) à la tête de la cité (XV, § 13) : il s’agit manifestement de la lignée ininterrompue des imams, qui selon l’imamologie shi‘ite relie un prophète à l’autre. Mais si cette succession s’arrête, donc s’il n’y a plus d’imams, la cité sera gouvernée par un « second chef » (al-ra’īs al-thānī), Sayf al-Dawla ne prit possession d’Alep qu’en 333/944 (RUDOLPH, « Abū Naṣr al-Fārābī » p. 541, se référant à Ph. VALLAT, Farabi et l’École d’Alexandrie, Paris, 2004, p. 16). Outre le fait qu’aucune source ne prétend qu’al-Fārābī a été « invité » par Sayf al-Dawla, cet argument ne nous dit strictement rien sur les convictions religieuses du philosophe. Plus problématique, toutefois, est le fait que nos sources situent unanimement le décès du philosophe en 339 à Damas, alors que cette ville, après le partage de la Syrie en 336/947, semble avoir été sous le contrôle des Ikhshidides sunnites d’Égypte (M. Canard, « Ḥamdānides », dans : Encyclopédie de l’islam. Deuxième édition). Mais ici encore, cela n’exclut pas pour autant qu’al-Fārābī ait été shi‘ite et qu’il ait appartenu au cercle des érudits réunis par Sayf al-Dawla à Alep. 89. Cette supposition va à contre-courant de ce qui semble être le mainstream des études fārābiennes, dont RUDOLPH, « Abū Naṣr al-Fārābī », p. 631, se fait le porte-parole éloquent : « Al-Fārābī was not interested in day-to-day politics nor in the religious legitimation of a specific dynasty ; he was concerned with philosophical reflection about the community based on fundamental principles. He thus followed a tradition which had already been avidly cultivated in antiquity ... ». Il est évident que nos sources bibliographiques lacunaires ne permettent pas de tirer une telle conclusion. Mais, pour Rudolph, c’est un moyen commode d’évacuer les thèses dérangeantes de Fawzi Najjar, Henry Corbin, Richard Walzer et, par anticipation, de l’auteur du présent article, et de passer aux choses sérieuses : la question de l’influence du platonisme sur la pensée d’al-Fārābī.

88

DANIEL DE SMET

qui devra réunir les qualités suivantes : être philosophe (ḥakīm), connaître et préserver les lois et coutumes des premiers souverains, déduire (istinbāṭ) de nouvelles lois à base des préceptes établis par les premiers imams ; exceller dans la ru’ya (la délibération) et l’istinbāṭ pour pouvoir répondre à des questions qui ne se posaient pas à l’époque des premiers souverains, et cela en vue du bien-être (ṣalāḥ) de la cité ; guider (irshād) par la parole vers les lois des premiers souverains et vers celles qui ont été déduites après eux ; et enfin avoir la condition physique requise pour mener la guerre (XV, § 13). Ces qualités peuvent être présentes en différentes personnes, qui s’associent pour diriger la cité. Mais si aucun philosophe ne participe au gouvernement, la cité se corrompra et périra (XV, § 14)90. Nous reconnaissons dans les caractéristiques de ce « second chef » le gouverneur idéal, tel que le conçoit al-Fārābī en période d’occultation, alors que l’imam n’est plus disponible. Ce gouvernement de substitution n’édicte plus de nouvelles lois – c’est le privilège des prophètes et des imams – mais a recours à des méthodes de déduction (istinbāṭ) afin d’adapter le droit à de nouvelles situations qui n’étaient pas prévues par les législateurs. Il me semble très probable qu’al-Fārābī nous décrive ici la situation juridique qui apparut dans le shi‘isme duodécimain après la ghayba. En effet, les différentes méthodes de logique juridique (ra’y, qiyās, ijtihād, istiḥsān) employées par les juristes sunnites pour déduire des lois concrètes à partir des deux bases de la loi musulmane que sont le Coran et la Tradition91, ou pour adapter des lois existantes à de nouvelles conditions, sont, en théorie du moins, rigoureusement interdites par les shi‘ites, l’imam étant pour eux l’unique autorité en matière de droit. Si cela reste toujours le cas dans l’ismaélisme nizarite, où la succession des imams continue jusqu’à aujourd’hui, le shi‘isme duodécimain a été contraint, suite à l’occultation de l’imam, à adopter certaines de ces méthodes tant décriées92. Il en résulte un basculement 90. Le règne du « second chef » est également évoqué dans AL-FĀRĀBĪ, Siyāsa, pp. 80-81 ; ID., Milla, pp. 60-67. 91. Sur ces méthodes, voir, entre autres, C. CHEHATA, « Logique juridique en droit musulman », dans : Studia Islamica 23 (1965), pp. 5-25 ; É. TYAN, « Méthodologie et sources du droit en Islam (Istiḥsān, Istiṣlāḥ, Siyāsa šar‘iyya) », dans : Studia Islamica 10 (1959), pp. 79-109. 92. I. POONAWALA, « Al-Qāḍī al-Nu‘mān and Isma‘ili Jurisprudence », dans : F. DAFTARY (éd.), Medieval Isma‘ili History and Thought, Cambridge 1996, pp. 117-145 ;

L’IMAM, PHILOSOPHE ET PROPHÈTE DE LA CITÉ VERTUEUSE

89

du shi‘isme duodécimain vers le sunnisme, une évolution qui provoqua de vives réactions auprès de ceux qui trouvaient que les juristes s’arrogeaient des droits qui ne leur revenaient pas93. Dans un tel contexte, la démarche d’al-Fārābī prend sens : pour éviter une dérive totale et prémunir la religion et la loi contre des interprétations arbitraires, il faut la présence d’un philosophe dans le gouvernement de la cité. Grâce à son intellect acquis, inspiré par l’Intellect agent, il veille à ce que la cité ne s’écarte pas des voies tracées par les prophètes et les imams. Si cette hypothèse tient la route, les Mabādi’ pourraient être lus comme un « Fürstenspiegel »94 écrit pour Sayf al-Dawla, afin de l’aider à organiser un état shi‘ite désormais dépourvu d’imam. Daniel DE SMET CNRS Laboratoire d’Études sur les Monothéismes-UMR 8584 Campus Condorcet, 14 cours des Humanités F-93322 Aubervilliers [email protected]

F. DAFTARY, « Al-Qāḍī al-Nu‘mān, Ismā‘īlī Law and Imāmī Shī‘ism », dans : M. A. AMIRMOEZZI – M. BAR-ASHER – S. HOPKINS (éds.), Le shī‘isme Imāmite quarante ans après. Hommage à Etan Kohlberg, Turnhout 2009, pp. 179-186. 93. Sur ce processus de sunnitisation, qui s’accentua sous le règne des Buyides shi‘ites à Bagdad (334/945-447/1055), voir AMIR-MOEZZI – JAMBET, Qu’est-ce que le shi’isme ?, pp. 183-194. 94. Bien sûr, l’ouvrage d’al-Fārābī n’a rien en commun avec le genre des « Fürstenspiegel » tel qu’il fut développé dans la littérature arabe. Le terme est ici employé dans un sens métaphorique, pour dire qu’al-Fārābī a inséré dans sa philosophie politique sa vision sur la direction de la cité en l’absence d’un prophète ou d’un imam.

TASHKĪK AL-WUJŪD AND THE LAWĀZIM IN AVICENNA’S METAPHYSICS* Damien JANOS Abstract In this article, I have two main aims. First, I intend to examine how Avicenna’s theory of tashkīk al-wujūd fits in his general description of the subject matter of metaphysics in the Ilāhiyyāt of the Shifā᾿. Second, I analyze the complicated relation between tashkīk and the māhiyya-lawāzim distinction, which is relevant to many facets of his philosophy. The analysis underscores deep conceptual correspondences between Avicenna’s discussions of wujūd and mawjūd in Ilāhiyyāt, I, 1‒5, of māhiyya and the lawāzim of māhiyya in Ilāhiyyāt, V, 1, and of tashkīk in the various passages that broach that notion. It emerges that while tashkīk applies for the most part to the concomitants and accidents of essence, it can also apply to notions that designate substance and essence themselves. More specifically, Avicenna exceptionally allows for the essential predication of tashkīk in a theological and ontological context. This feature of his philosophical doctrine, and the disagreements it prompted, can help to explain three major interpretive trends that developed during the post-classical period.

1. Introduction Recently there has been a flurry of studies focusing on the notion of tashkīk al-wujūd in Arabic philosophy. Most of these studies have dealt with Avicenna’s (or Ibn Sīnā’s) theory of tashkīk al-wujūd and its reception in the post-classical tradition.1 Thanks to the ongoing discussion * I am grateful to Daniel De Smet and Meryem Sebti for their kind invitation to contribute to this volume. Many thanks also go to the anonymous reviewer, whose comments helped me to improve the draft. 1. S. RIZVI, Mullā Ṣadrā and Metaphysics: Modulation of Being, London 2009; T. KOUTZAROVA, Das Transzendentale bei Ibn Sīnā: Zur Metaphysik als Wissenschaft erster Begriffsund Urteilsprinzipien, Leiden / Boston 2009, pp. 211–230; A. TREIGER, “Avicenna’s Notion of Transcendental Modulation of Existence (taškīk al-wuğūd, analogia entis) and its Greek and Arabic Sources,” in: F. OPWIS – D. REISMAN (eds.) Islamic Philosophy, Science, Culture, and Religion. Studies in Honor of Dimitri Gutas, Leiden / Boston 2012, pp. 327–

92

DAMIEN JANOS

on this topic, we are beginning to grasp more clearly the important role this notion played in the intellectual history of Islam, starting with Fārābī and Avicenna in the tenth and eleventh centuries and extending all the way to Mullā Ṣadrā and the School of Isfahan in the seventeenth century. Although some disagreement still subsists as to how to define and translate the notion of tashkīk, as well as what its metaphysical implications are, the idea that Avicenna played a major role in its theorization seems well established in the scholarship. In the present paper, I wish to tackle a residual conceptual problem that arises from the investigation of tashkīk al-wujūd in his thought. The challenge taken up in this piece is to elucidate how tashkīk al-wujūd relates to Avicenna’s māhiyya-lawāzim paradigm, that is, how it relates to quiddity (māhiyya), on the one hand, and to the external concomitants of quiddity (lawāzim al-māhiyya), on the other.2 As we will see, this issue bears directly on the way we understand Avicenna’s ontology and theology. With regard to the former, it raises the question of how Avicenna conceived of the place and role of tashkīk al-wujūd in his metaphysical project and how it intersects with his description of this science as the study of “the existent inasmuch as it is existent” (al-mawjūd bi-mā huwa mawjūd).3 With regard to the latter, it will help us to shed light on the issues of whether, and if so how, tashkīk al-wujūd applies to God. These two topics directly impact our understanding of 363 (this is a revised version of Treiger’s article published in Documenti e studi sulla tradizione filosofica medievale 21 [2010], pp. 165–198); D. DE HAAN, “The Doctrine of the Analogy of Being in Avicenna’s Metaphysics of the Healing,” The Review of Metaphysics 69.2 (2015), pp. 261–286; ID., Necessary Existence and the Doctrine of Being in Avicenna’s Metaphysics of the Healing, Leiden 2020, pp. 294–336; R. WISNOVSKY, “On the Emergence of Maragha Avicennism,” Oriens 46 (2018), pp. 263‒331; D. JANOS, Avicenna on the Ontology of Pure Quiddity, Berlin 2020, pp. 58–60, 434–477; ID., “Avicenna on Equivocity and Modulation: A Reconsideration of the asmā᾿ mushakkika (and tashkīk al-wujūd),” Oriens, pp. 1–62 (advance article published online on September 1, 2021); R. P. ANSARI, “The Ambiguity of ‘Being’ in Arabic and Islamic Philosophy,” Ph.D. dissertation, Georgetown University 2020; F. BENEVICH, “The Necessary Existent (wāǧib al-wuǧūd): From Avicenna to Faḫr al-Dīn al-Rāzī,” in: A. SHIHADEH – J. THIELE (eds.) Philosophical Theology in Islam: Later Ash῾arism East and West, Leiden 2020, pp. 123–55; and F. O. ZAMBONI, “Is Existence One or Manifold? Avicenna and his Early Interpreters on the Modulation of Existence (taškīk al-wuǧūd),” Documenti e studi sulla tradizione filosofica medievale 31 (2020), pp. 121–149. 2. These notions are explained in section 3 of this paper. 3. This question was already raised by DE HAAN, “The Doctrine,” who perceived an “apparent inconsistency between his [Avicenna’s] doctrine of the analogy of being with his presentation of the subject of metaphysics as absolute being” (p. 286).

TASHKĪK AL-WUJŪD AND THE LAWĀZIM IN METAPHYSICS

93

the reception of Avicenna’s thought in the post-classical period. It should be noted that the question of the theological relevance of tashkīk alluded to above was itself connected with two other, more specific issues that were intensely debated in the post-classical period: the univocity vs. equivocity of wujūd, and whether essence and existence are identical or distinct in God. The structure of the present paper reflects these two broad ontological and theological problematics, which it discusses separately. In the process of addressing these topics, particular emphasis will be placed on the centrality of the notion of the concomitants (lawāzim) in Avicenna’s metaphysics. The overarching aim of the study is to establish tashkīk and tashkīk al-wujūd as core and fundamental notions in Avicenna’s philosophy.4 2. Key considerations regarding tashkīk al-wujūd Before turning to these issues, however, it is desirable to say a few words about recent developments in the study of tashkīk al-wujūd. In what follows, I will limit myself to briefly summarizing the various findings that have recently been brought to light and refer the readers to the relevant literature on the topic. The first noteworthy point is that tashkīk applies to many different philosophical notions and terms in Avicenna’s works, and not just to existence, even though the latter notion, together with oneness (waḥda), is arguably the most important modulated term the shaykh identifies in his logical and metaphysical works. Indeed, wujūd and waḥda may be considered the paradigmatic cases of tashkīk.5 Taken together with the other notions Avicenna identifies as modulated, these constitute a distinct category of terms called asmā᾿ or alfāẓ mushakkika. Even though they are primarily grounded in logic and discussed in Avicenna’s logical works, the modulated terms play a key role in this thinker’s metaphysics. Since Treiger’s 4. On this point, I take my cue from Rizvi and Treiger, whose studies established the importance of tashkīk al-wujūd in the Arabic metaphysical tradition. RIZVI, Mullā Ṣadrā, p. 1, claims that “modulation is central to Sadrian ontology.” Likewise, I want to show that it lies at the very core of Avicenna’s ontology. Compared to Ṣadrā, however, the textual evidence in Avicenna dealing directly with tashkīk al-wujūd is limited and has to be painstakingly collected and analyzed. 5. For a list of the various terms Avicenna describes as modulated, see JANOS, “Avicenna on Equivocity and Modulation.” The notions of wujūd and waḥda appear in virtually all the passages that expressly discuss tashkīk in Avicenna’s corpus. For the various references to the primary sources, see JANOS, Avicenna, pp. 434–477.

94

DAMIEN JANOS

seminal article, it has become apparent that tashkīk assumes a full-blown metaphysical relevance in Avicenna’s thought, well beyond the domain of logic.6 Moreover, we now know that the metaphysical and theological significance of this theory was amplified during the post-classical period, with the debate focusing on the question of its application to God’s existence. Among other things, it had an impact on Fakhr al-Dīn al-Rāzī’s evaluation of Avicenna’s ontology and on Naṣīr al-Dīn al-Ṭūsī’s response to al-Rāzī.7 On the other hand, and returning to Avicenna, the questions of how to translate the term tashkīk, of whether it lies closer to univocity or equivocity, and to what objects it applies, have proven quite controversial in the secondary literature. As recent studies have shown in detail, the theory of tashkīk al-wujūd situates existence between the two extremes of absolute univocity (tawāṭu᾿ ) and absolute equivocity (ishtirāk). Although Treiger proposed to regard it as a kind of “modulated univocity of existence,” other scholars have argued that tashkīk al-wujūd can also be viewed as expressing a kind of “modulated equivocity of existence,” which bears a direct link to Aristotelian pros hen homonymy and the various iterations that theory underwent in the late-antique philosophical tradition.8 Both approaches appear to be 6. DE HAAN, “The Doctrine,” also emphasizes the metaphysical relevance of tashkīk in Avicenna’s philosophy. The present study will lend additional weight to that hypothesis. It is worth pointing out in this connection that Avicenna himself rarely if ever uses the expression tashkīk al-wujūd; he does, however, explicitly describe wujūd and mawjūd as mushakkikān (or mushakkakān, both vocalizations having been retained by scholars), thereby legitimating to some extent the former locution. In spite of this, one still has to account for the transition from a purely logical analysis of tashkīk to its metaphysical implications. 7. For the reception of Avicenna’s theory of tashkīk al-wujūd in the works of Bahmanyār, al-Rāzī, al-Ṭūsī, and other post-classical thinkers, see WISNOVSKY, “On the Emergence,” BENEVICH, “The Necessary Existent,” and ZAMBONI, “Is Existence?” 8. For the view that tashkīk al-wujūd expresses a kind of univocity of existence, see TREIGER, “Avicenna’s Notion”; T.-A. DRUART, “Ibn Sīnā and the Ambiguity of Being’s Univocity,” in: M. ARFA MENSIA (ed.) Views on the Philosophy of Ibn Sīnā & Mulla Ṣadrā Shīrāzī, Carthage 2014, pp. 15–24; and BENEVICH, “The Necessary Existent.” Others, in contrast, have argued that it expresses a kind of equivocity; see Rizvi, Mullā Ṣadrā, especially pp. 1, 38–39, 42, 46–47, and 52; A. BERTOLACCI, “The ‘Ontologization’ of Logic. Metaphysical Themes in Avicenna’s Reworking of the Organon,” in: M. CAMERON – J. MARENBON (eds.), Methods and Methodologies: Aristotelian Logic East and West, 500– 1500, Leiden 2010, pp. 25–51, especially pp. 41–49; T. M. JOHANNESEN, “The Avicennian Moment: Separation and the Science of Being (qua Being) in Aristotle and Avicenna,” Ph.D. dissertation, Asia Institute, University of Melbourne 2019, especially pp. 114–122; and JANOS, “Avicenna on Equivocity and Modulation.”

TASHKĪK AL-WUJŪD AND THE LAWĀZIM IN METAPHYSICS

95

warranted and find traction in Avicenna’s works. On the one hand, Avicenna stresses the unity and cohesiveness of the notion of wujūd, endowing it with an overarching or focal meaning, and making it the primary subject matter of the metaphysical enterprise, whose aim is to study “being qua being” or “the existent inasmuch as it is existent.” On the other hand, he also stresses that tashkīk “differentiates” (ikhtalafa, ikhtilāf ) between nuances in the notion of existence, even going so far as to say that wujūd has many senses (ma῾ānī kathīra).9 The fact that these elements appear side by side in Avicenna’s works accounts largely for the divergence in the modern interpretations of that notion, as well as for the post-classical debates regarding wujūd: some post-classical thinkers, such as Fakhr al-Dīn al-Rāzī (d. 1210 C.E.), opted to stress the first aspect, leading them to regard wujūd as a completely univocal notion possessing a single nature (ṭabī῾a), while others, such as Naṣīr al-Dīn al-Ṭūsī (d. 1274 C.E.), made the second aspect the cornerstone of their ontology and theology, stressing modulation or a mild form of equivocity as the key notion at play. Recent studies have identified three key avenues of research pertaining to tashkīk al-wujūd in Avicenna and beyond, which are outlined below, and of which the third is the most important for our purposes. The first is to examine the question of whether tashkīk amounts to a kind of transcendental or categorial predication, that is, whether it applies only to the contingent entities of Avicenna’s ontology, or to God as well, Who, technically, lies beyond the categories of substance and accident. In their studies, Treiger and De Haan very perceptively argued that tashkīk al-wujūd extends to, and encompasses, the First Cause or God. This led Treiger to formulate the main argument of his paper, which is that tashkīk al-wujūd in Avicenna amounts to a kind of transcendental univocal modulation. As I will explain later, I believe Treiger was right to regard tashkīk as transcendental, although the philosophical reason he adduces to justify this point is in my eyes problematic.10 9. See AVICENNA, The Metaphysics of The Healing, A parallel English-Arabic text translated, introduced, and annotated by M. E. MARMURA, Provo, Utah 2005, I, 5, p. 24. 11–12. In order to be consistent in my use of the Arabic titles, I shall henceforth refer to Marmura’s edition and translation as the Ilāhiyyāt. 10. I also take issue with the description of tashkīk as a kind of univocity, but will not engage with this question here, as it is not the main focus of the present study. In addition

96

DAMIEN JANOS

The second topic concerns the constitutive “aspects,” “modes,” or “criteria” of tashkīk al-wujūd, in other words, the notions that are subsumed under it and are responsible for “performing” modulation. Although the terms “aspects,” “modes,” and “criteria” are mine, Treiger shed valuable light on this topic by analyzing key texts from the Avicennian corpus. He identified the following three sets of notions as being relevant to Avicenna’s theory of tashkīk al-wujūd: “priority and posteriority” (al-taqaddum wa-l-ta᾿akhkhur); “more deserved and more appropriate” or “degree of deservingness” (al-awlā wa-l-aḥrā); and “strength and weakness” (al-shidda wa-l-ḍu῾f ).11 Nevertheless, the list of aspects subsumed under tashkīk al-wujūd that Treiger provides, which is drawn from the well-known passage in Maqūlāt, I, 2, is incomplete. There are other aspects that need to be taken into account. For example, Avicenna’s distinction between possibility and necessity is a crucial aspect of tashkīk al-wujūd that should be added to the list. This move is justified inter alia by a passage of Ilāhiyyāt, VI, 3, which mentions possibility and necessity as aspects that differentiate wujūd, even though the term tashkīk is not used expressly in that excerpt. I quote it here in full because of its intrinsic importance, and also because it will play a role in the later analysis of tashkīk al-wujūd: Existence inasmuch as it is existence is not differentiated according to strength and weakness (lā yakhtalifu fī l-shidda wa-l-ḍu῾f ), nor does it accept degrees in deficiency. Rather, it is differentiated (yakhtalifu) according to

to Treiger’s paper, some recent articles have discussed the transcendental quality of tashkīk al-wujūd in the post-Avicennian period; see BENEVICH, “The Necessary Existent,” and ZAMBONI, “Is Existence?” As interesting as they are, these subsequent developments do not concern Avicenna directly and will be tackled only briefly towards the end of this paper. 11. See TREIGER, “Avicenna’s Notion,” pp. 353–358; DE HAAN, “The Doctrine,” pp. 272–278 also limits his analysis to these three aspects; cf. JANOS, Avicenna, pp. 441– 449. As has been emphasized on various occasions in the scholarship, for Avicenna, the most salient and important aspects of tashkīk al-wujūd are priority and posteriority, which are always connected with it in his accounts. For example, it is the only aspect mentioned in connection with tashkīk al-wujūd in Kitāb al-Hidāya, ed. M. ABDUH, Second edition, Cairo 1974, p. 232. 9–12; for other references to the primary sources, see JANOS, Avicenna, pp. 445–446, 460–477. Moreover, it should be stressed that the aspect of “strength and weakness” or “degrees of intensity” that is associated with tashkīk al-wujūd is highly problematic, inasmuch as Avicenna is inconsistent in upholding its validity, as the excerpt quoted below explicitly shows.

TASHKĪK AL-WUJŪD AND THE LAWĀZIM IN METAPHYSICS

97

a number of aspects [or modes, ῾iddat aḥkām], which are: priority and posteriority, autonomy and need, and necessity and possibility.12

It is quite evident that this passage directly addresses modulation, even though Avicenna omitted to include the term tashkīk itself. Two factors establish this unequivocally. First, the idea that the meaning of wujūd is differentiated (ikhtalafa) appears in virtually all the passages that deal expressly with tashkīk al-wujūd. Second, the inclusion of priority and posteriority is recurrent in Avicenna’s expositions of that theory. Hence, on the basis of this passage, the modalities (jihāt), as well as the notions of autonomy and need, can also be said to participate in the modulation of existence. This, it should be noted, is entirely consistent with the place and role of the modalities in Avicenna’s metaphysics, which distinguishes between necessary and possible/contingent beings, and which regards possibility and necessity as attributes or accidents of “the existent” (mawjūd).13 In addition to the aspects outlined above, Avicenna sometimes includes “the essential” or “what is by essence” (bi-l-dhāt) and “the accidental” or “what is by accident” (bi-l-῾araḍ) — or, alternatively, what is “through one’s self” (min or bi-dhātihi) and what is “through another” (min or bi-ghayrihi) — as aspects of tashkīk al-wujūd. These are mentioned, for example, in Maqūlāt, I, 2, Jadal, II, 2, and in the Ilāhiyyāt of the Dānesh-nāmeh.14 Whereas in Maqūlāt, I, 2, Avicenna employs these notions merely to explain al-awlā wa-l-aḥrā — what has existence min dhātihi is “worthier” and “more deserving” than what has existence min ghayrihi — in the Jadal passage, the distinction 12. AVICENNA, Ilāhiyyāt, VI, 3, p. 213. 14–17, my translation; cf. AVICENNA, Ilāhiyyāt, V, 8, p. 186. 5–7, which explains that “the existent” and “the one” are predicated according to priority and posteriority, but which omits the term tashkīk. See also JANOS, Avicenna, pp. 443–444. Treiger quotes this passage in note 92 of his article, but apparently did not take it to refer to tashkīk directly, perhaps because the term itself is absent. Here it is important to clarify the following terminological point: the term jihāt refers in Avicenna to the modalities (possibility and necessity), whereas the terms anḥā᾿ and aḥkām refer to the aspects of tashkīk al-wujūd. Hence, the jihāt are one naḥw or ḥukm of tashkīk al-wujūd alongside many others. 13. See AVICENNA, Ilāhiyyāt, I, 1, section 15. 14. AVICENNA, Maqūlāt, al-Manṭiq, al-Shifā᾿, ed. J. SH. QANAWATĪ – M. AL-KHUḌAYRĪ – A. F. AL-AHWĀNĪ – S. ZĀYID, Cairo 1959, I, 2, pp. 10–11, especially p. 10. 13–16; ID., Jadal, al-Manṭiq, al-Shifā᾿, ed. A. F. AL-AHWĀNĪ, Cairo 1965, p. 117. 7–9, and more generally pp. 117–120; ID., Ilāhiyyāt: Dānesh-nāmeh-i ῾Alā᾿ī, ed. M. MU῾ ĪN, Tehran 1952, section 11.

98

DAMIEN JANOS

between bi-l-dhāt and bi-l-῾araḍ is presented as a core notion of tashkīk. That these various interrelated notions would also be subsumed under tashkīk al-wujūd is again not surprising, given that Avicenna states at Ilāhiyyāt, I, 4 that metaphysics investigates “the state of what is in essence and what is accidental” (ḥāl alladhī bi-l-dhāt wa-alladhī bi-l῾araḍ).15 Another important point to dwell on concerns the criterion of alshidda wa-l-ḍu῾f, which has been treated rather hastily in the secondary literature on Avicenna. It can be translated roughly as “intensity in existence,” “strength and weakness,” or “grades of existence.” Scholars have often pointed out that although Avicenna dismisses this aspect in his implementation of tashkīk al-wujūd, it became preponderant in the later metaphysical tradition in Islam, especially in the works of Mullā Ṣadrā.16 It is true that the shaykh in many instances shuns this aspect when discussing tashkīk. At times, as in the text quoted above, he even cautions that it has no role to play in the metaphysical investigation. However, on other occasions, Avicenna seems to approve the use of al-shidda wa-l-ḍu῾f. This is the case of the locus classicus in Maqūlāt, I, 2, as well as of passages in the Maqūlāt of al-Mukhtaṣar al-awsaṭ fī l-manṭiq and the Dānesh-nāmeh.17 In the first text, he gives a restricted application of this aspect that seems limited to certain accidents and concomitants, such as whiteness, which can be of different intensity in snow and in ivory. He also somewhat oddly gives philosophy as an example. But in the Maqūlāt of al-Mukhtaṣar al-awsaṭ fī l-manṭiq, the phrasing and context seem to suggest that Avicenna applies these notions to wujūd itself. And in section 11 of the Ilāhiyyāt of the Dānesh-nāmeh, he states that “existence applies to these things [the categories] according to prior and posterior and to more or less (kamā bīshī), though it applies to one meaning. This [kind of predicate] is called modulated (moshakkek).”18 This suggests that degrees 15. AVICENNA, Ilāhiyyāt, I, 4, p. 19. 13–14. 16. RIZVI, Mullā Ṣadrā, pp. 20 and 44; TREIGER, “Avicenna’s Notion,” p. 357, note 92. 17. AVICENNA, Maqūlāt, I, 2, pp. 10–11; A. KALBARCZYK, “The Kitāb al-Maqūlāt of the Muḫtaṣar al-awsaṭ fī l-manṭiq: A Hitherto Unknown Source for Studying Ibn Sīnā’s Reception of Aristotle’s Categories,” Oriens 40 (2012), pp. 305–354, p. 326 of the Arabic text. 18. See AVICENNA, Ilāhiyyāt: Dānesh-nāmeh-i ῾Alā᾿ī, section 11; the English translation is taken from TREIGER, “Avicenna’s Notion,” p. 359; cf. P. MOREWEDGE, The Metaphysica of Avicenna (ibn Sīnā). A critical translation-commentary and analysis of the fundamental

TASHKĪK AL-WUJŪD AND THE LAWĀZIM IN METAPHYSICS

99

in strength and weakness in existence apply to substances and accidents. In any case, whenever Avicenna employs the formula al-shidda wa-l-ḍu῾f – which is a consecrated expression in his works – it is typically in connection with concomitants and accidents. For instance, the expression is mentioned twice in al-Nafs to describe the activities (af῾āl) of the soul, which are liable to increase and decrease in intensity.19 Hence, even though one should acknowledge that the evidence pertaining to this aspect of tashkīk al-wujūd is ambiguous and problematic, it is nevertheless the case that Avicenna significantly downplays its importance in his works and that he usually connects it with minor accidents and concomitants. On the other hand, one might ask, if it truly has nothing to do with the modulation of existence, then why is it included at all in the expositions of tashkīk al-wujūd that Avicenna provides in his works? Moreover, Naṣīr al-Dīn al-Ṭūsī, one of the greatest exponents of Avicennian philosophy in the thirteenth century, enthusiastically embraced this notion and made it a cornerstone of his ontological system. Unlike al-Ṭūsī, who applies it squarely to wujūd, the shaykh seems to allow a certain restricted application of al-shidda wa-l-ḍu῾f to the concomitants and accidents. But then again, is wujūd not described as a concomitant of essence in his system? More research on this point is called for. To recap, and for the sake of the forthcoming analysis, I provide below a table of the various aspects or modes of tashkīk al-wujūd as they appear in the Avicennian corpus:20 Table 1: The aspects of tashkīk al-wujūd mentioned in Avicenna’s works “Priority and posteriority” (al-taqaddum wa-l-ta᾿akhkhur, awwalan wa-thāniyyan)

Maqūlāt of al-Mukhtaṣar al-awsaṭ fī l-manṭiq; Maqūlāt, I, 2; Jadal, II, 2; Dānesh-nāmeh, section 11; al-Samā῾ al-ṭabī῾ī, I, 3 and II, 2; Ilāhiyyāt, VI, 3, V, 8; K. al-Hidāya

arguments in Avicenna’s Metaphysica in the Dānish Nāma-i ῾alā᾿ī (The Book of Scientific Knowledge), London 1973, p. 32. 19. AVICENNA, al-Nafs, al-Shifā᾿, Avicenna’s De anima (Arabic text), being the Psychological Part of Kitāb al-Shifā᾿, ed. F. RAHMAN, London 1959, p. 33. 7–8 and 18. 20. The list is limited to passages where tashkīk is explicitly mentioned, the exceptions being Ilāhiyyāt, VI, 3 and V, 8.

100

DAMIEN JANOS

“Worthier and more deserving” (al-awlā wa-l-aḥrā)

Maqūlāt of al-Mukhtaṣar al-awsaṭ fī l-manṭiq; Maqūlāt, I, 2

“Possibility and necessity” (al-imkān wa-l-wujūb)

Ilāhiyyāt, VI, 3

“Autonomy and need” (al-istighnā᾿ wa-l-ḥājja)

Ilāhiyyāt, VI, 3

“Intensity and weakness” (al-shidda wa-l-ḍu῾f )

Maqūlāt of al-Mukhtaṣar al-awsaṭ fī l-manṭiq; Maqūlāt, I, 2; Dānesh-nāmeh, section 11

“Essentially and accidentally” (bi-l-dhāt wa-bi-l-῾araḍ) or “in itself and through another” (min dhātihi wa-min ghayrihi)

Maqūlāt, I, 2; Jadal, II, 2; Dānesh-nāmeh, section 11

The third issue broached by scholars explores how tashkīk al-wujūd relates to the state of being an external concomitant or attribute.21 Basing himself mostly on a passage from Mubāḥathāt, Treiger explains that Avicenna frames wujūd as a “non-constitutive concomitant” (lāzim ghayr muqawwim) of essence. As Treiger put it: According to this solution, related to Avicenna’s famous quiddity/existence (māhīya/wuǧūd) distinction, existence is neither a genus nor a quasigenus, but a non-constitutive concomitant (lāzim ġayr muqawwim) that is, an inseparable accident of every quiddity … This theory [taškīk al-wuǧūd] explains how, in Avicenna’s view, existence is differentiated into the Necessarily Existent and the contingent existents, and further into the ten categories. It is differentiated not as a genus (or a quasigenus) by differentiae, but by the very quiddities of the things of which it is predicated; this becomes possible precisely because it does not form a part, and is not a constituent, of these quiddities.22

On the one hand, this presumably explains why, on Treiger’s reconstruction, tashkīk al-wujūd expresses a kind of univocity of being, since it is always predicated as an external attribute of essence and remains distinct from quiddity. On the other hand, it plays a role in 21. TREIGER, “Avicenna’s Notion,” pp. 360–363, although it should be noted that this issue is not clearly spelled out as such in his article. Recently, DE HAAN, Necessary Existence, pp. 314–335, BENEVICH, “The Necessary Existent,” and ZAMBONI, “Is Existence?” especially pp. 126, 132, have broached this issue in a more explicit way. 22. TREIGER, “Avicenna’s Notion,” pp. 361–363; the emphasis is Treiger’s.

TASHKĪK AL-WUJŪD AND THE LAWĀZIM IN METAPHYSICS

101

explaining why tashkīk al-wujūd is transcendental in nature, as well as why the necessary existence of the First does not lead to an internal duality of necessity and existence, as was argued by Muḥammad b. ῾Abd al-Karīm al-Shahrastānī (d. 1153 C.E.) and others in an attempt to undermine Avicenna’s metaphysics. The unstated assumption of Treiger’s argumentation is that, in God as well, wujūd may be regarded as a lāzim of His essence.23 In two recent and insightful studies, Fedor Benevich and Francesco Zamboni also emphasized the connection between the status of existence as a concomitant and its modulation. According to Benevich, Avicenna rejected the idea of the equivocity of existence and laid the ground for al-Rāzī’s strong interpretation of univocity by emphasizing that wujūd is always a lāzim of the essence.24 In addition, he elaborates on Treiger’s hint that God’s existence could be viewed as an external concomitant of His essence.25 Benevich shows convincingly that al-Rāzī establishes a conceptual connection between the status of wujūd as a concomitant, the univocity of being, and the real and conceptual essence-existence distinction in all beings. According to al-Rāzī, wujūd is univocal because it is always predicated as something external and posterior to the essence, and not as one of its constituents. If that were not the case, and if existence were somehow constitutive of essence, then existence and essence would either overlap significantly or be identical, with the result that each being would have a different type of essence-existence predicated of it, leading to equivocity.26 Note that al-Rāzī’s contention that existence is a univocal attribute of essence applies both to contingent beings and to God. One upshot of this view is that what al-Rāzī criticizes in Avicenna is not so much his theory of tashkīk al-wujūd (which al-Rāzī largely ignored and construed as being consistent with a univocal conception of wujūd), but rather his lack of consistency in making God an exception to the rule that existence must be predicated as something distinct from, as well as external and posterior to, the essence. As for 23. TREIGER, “Avicenna’s Notion,” p. 362, note 106. I return to these two points (the univocal and transcendental nature of tashkīk al-wujūd) at the end of this paper. 24. BENEVICH, “The Necessary Existent,” especially pp. 130, 134–135, 145, and 149. 25. BENEVICH, “The Necessary Existent,” p. 150. 26. As BENEVICH, “The Necessary Existent,” pp. 125–126, notes, this is the position that is often ascribed to Ash῾arites in the later doxographical literature.

102

DAMIEN JANOS

Zamboni, he also bases the idea of modulated univocity, or what he calls Avicenna’s theory of “the essential unity of existence,” on the external and concomitant status of existence vis-à-vis essence. In his reconstruction as well, tashkīk al-wujūd presupposes a link between the concomitant status of wujūd and its unitary meaning.27 The emphasis placed on wujūd as a concomitant in these scholars’ analyses of tashkīk is fully vindicated in Avicenna’s metaphysics. In fact, the correlation between tashkīk and the lawāzim extends well beyond the paradigmatic case of wujūd. For instance, it also applies to the modulation of oneness (waḥda), since oneness, for Avicenna, is a concomitant and attribute of essence. The notion of “concomitant” (lāzim, lāḥiq) is crucial not only to understand the relation between essence and existence as Avicenna conceives it, but also to conceptualize the various additional aspects that are related to both essence and existence, such as unity and multiplicity, possibility and necessity, and particularity and universality, to name only a few. For according to the shaykh, these various notions have in common the fact that they are concomitants. With that being said, I believe that there are two crucial features of these scholars’ accounts that need to be revised, especially as they pertain to Avicenna. The first is the idea that there is a strict correlation between the status of existence as an external concomitant and its univocal predication. The fact that Avicenna often describes wujūd as an external concomitant of essence has been widely taken to justify the univocal or modulated-univocal nature of that notion. It is highly plausible that al-Rāzī was partly responsible for this interpretive trend, since this is the position that he promoted in his works.28 The assumption is that existence is modulated-univocal because it is always 27. ZAMBONI, “Is Existence?” p. 122. Zamboni does not explicitly construe tashkīk al-wujūd as a kind of univocity in his article and maintains the notion of modulation as one distinct from univocity. But his interpreting modulation in light of an essential meaning of existence, as well as the fact that he praises al-Rāzī (who himself explicitly upheld univocity) as a more faithful interpreter of Avicenna than either Bahmanyār or al-Ṭūsī (p. 122), certainly go in that direction. 28. At least in the case of Benevich and Zamboni, who focus mostly on post-classical thinkers in their studies, there are reasons to suspect that their interpretation of Avicenna is influenced by al-Rāzī’s own construal of the master’s ontology and theory of tashkīk al-wujūd. At any rate, these three scholars all link univocity to the status of wujūd as a concomitant.

TASHKĪK AL-WUJŪD AND THE LAWĀZIM IN METAPHYSICS

103

predicated externally of things according to a single core meaning, even though tashkīk also allows for certain differentiations to be introduced in the act of predication. Conversely, if existence were taken to be constitutive of, identical to, or interchangeable with, essence – as many Ash῾arite theologians apparently claimed – then existence would be fully equivocal, because it would be said essentially differently of different beings (horses, giraffes, etc.). However, this reconstruction of the evidence runs into three obstacles. First, it assumes that wujūd is treated as a lāzim throughout the entire spectrum of Avicenna’s ontology, which, as we shall see, is unlikely to be the case. Second, it somehow conflates the notions of univocity and tashkīk, which is also problematic, since it is far from obvious that, semantically and conceptually, tashkīk signifies univocity rather than a mild form of equivocity.29 Finally, although I agree with these scholars regarding the transcendental scope of tashkīk al-wujūd in Avicenna, I think that the philosophical reason they adduce as a ground for it needs to be reevaluated. The view that God’s wujūd is a lāzim external to His quiddity is one that was defended by some later, post-Avicennian thinkers, especially Fakhr al-Dīn al-Rāzī. As such, it should probably be regarded as a later elaboration on the master’s views. When it comes to Avicenna’s corpus, the only traction this interpretation finds lies in a few obscure passages drawn from the Ta῾līqāt and the Mubāḥathāt, whose authorship, transmission, and interpretation remain problematic.30 29. On this important point, see JANOS, “Avicenna on Equivocity and Modulation.” 30. For the references to the primary sources and an analysis of these passages, see TREIGER, “Avicenna’s Notion,” p. 363, note 106; BENEVICH, “The Necessary Existent,” pp. 150–151; cf. D. DE SMET – M. SEBTI, “Avicenna’s Philosophical Approach to the Qur᾿an in the Light of his Tafsīr Sūrat al-Ikhlāṣ,” Journal of Qur᾿anic Studies 11.2 (2009), pp. 134‒148, especially pp. 140–141. For another reading of this aspect of Avicenna’s theology, see JANOS, Avicenna, pp. 545–550. The point is that it is likely that the Ta῾līqāt and the Mubāḥathāt mark certain important departures from Avicenna’s original doctrines and reflect the views of some of his disciples who were involved in the task of compiling these texts; see D. C. REISMAN, The Making of the Avicennan Tradition: The Transmission, Content, and Structure of Ibn Sīnā’s al-Mubāḥathāt (The Discussions), Leiden 2002, for the Mubāḥathāt; and J. JANSSENS, “Ibn Sīnā’s Ta῾līqāt: The Presence of Paraphrases of and Super-commentaries on the Ilāhīyāt of the Šhifā᾿,” in: F. OPWIS – D. REISMAN (eds.), Islamic Philosophy, Science, Culture, and Religion: Studies in Honor of Dimitri Gutas, Leiden / Boston 2012, pp. 201–222, for the Ta῾līqāt. This does not mean, of course, that these works should not be consulted and brought within the fold of the analysis, but rather that their contents need to be systematically collated with that of Avicenna’s other works.

104

DAMIEN JANOS

In contrast, in all of his core works, from the early al-Mabda᾿ wa-lma῾ād, to the middle-period works al-Najāt and al-Shifā᾿, to the late al-Ishārāt wa-l-tanbīhāt, Avicenna is adamant about the fact that wujūd and māhiyya are identical in God and that wujūd is not something added to or distinct from God’s essence, but something essential to Him. One clear upshot of Avicenna’s numerous provisos to that effect would be that the connection between existence (wujūd) and the concomitants (lawāzim) applies only to the ontological structure of contingent beings, that is, to all things other than God. So, if tashkīk al-wujūd applies to God as well, it must be in a different way and with regard to God’s special, essential existence.31 In brief, modulation would relate differently to Avicenna’s māhiyya-lawāzim distinction depending on whether one investigates existence in a theological context or in his general ontology.32 The foregoing has identified a tension running through Avicenna’s metaphysics; or rather, in the modern interpretations of it that have been articulated in the path inaugurated by al-Rāzī. If tashkīk al-wujūd 31. This, in fact, is the line of thought followed by Bahmanyār, al-Ṭūsī, and other thinkers who — and this is of course not coincidental — also reject a strictly univocal construal of wujūd and tashkīk al-wujūd. 32. Avicenna’s māhiyya-lawāzim paradigm is a sophisticated conceptual framework that the shaykh uses to address core logical, ontological, and theological issues. In logic, it is used to underpin the linguistic and predicamental structure that exists between subjects and predicates and by extension between substances and accidents. For example, in his logical works, Avicenna makes ample use of the terms māhiyya and lawāzim to explain how some predicates, such as universality or multiplicity, are conceptually external to and distinct from quiddity and added to it in the mind. In ontology, which in Avicenna is closely bound up with logic, it is used to express the causal connection that obtains between a cause and its effect, or between the essence or quiddity of a thing, existent, or entity – conceived of as existing in the mind or in reality – and the external attributes and accidents that are actualized when that thing exists, such as oneness and multiplicity, priority and posteriority, particularity and universality, etc. Finally, in Avicenna’s theology, the māhiyyalawāzim paradigm is used to express either the relation between God’s essence and His external attributes (ṣifāt, lawāzim) – a discourse Avicenna rejects – or between God and the effects He causes to exist. In the latter case, the caused, contingent existents relate to God’s māhiyya or self qua lawāzim. Thus, in Avicenna’s theology, this paradigm is relied upon to contrast God’s special essence to all other things defined as concomitants of His essence, or lawāzim entailed by God’s māhiyya. It should be noted that these three levels of application of the māhiyya-lawāzim paradigm are substantiated terminologically and conceptually in Avicenna’s works. That is, Avicenna himself describes 1/ God’s various effects; 2/ the attributes and accidents of existing entities, as well as existence itself relative to the essence of a thing; and 3/ the predicates of logic, as lawāzim, thereby pointing to the use of an overarching conceptual framework to address these issues.

TASHKĪK AL-WUJŪD AND THE LAWĀZIM IN METAPHYSICS

105

is solely restricted to dealing with wujūd as a lāzim of a māhiyya, then how can God’s existence be said to be essential or identical to His very māhiyya? One way to approach this conundrum would be to argue, like al-Rāzī, that Avicenna’s essence-existence identity thesis with regard to God is simply mistaken, and that even in the case of God, wujūd must be designated as a concomitant of essence. Note that this interpretive option maintains both the transcendental nature and the univocal nature of tashkīk al-wujūd, but contests a core tenet of Avicenna’s theology, the identity thesis. Another attempt is to claim that Avicenna did not extend tashkīk al-wujūd to God.33 The problem is that both the essence-existence identity thesis in God and the transcendental nature of tashkīk al-wujūd seem to be supported by considerable evidence in Avicenna’s works.34 My contention in this piece is that these two theses – i.e., the identity of essence and existence in God, and the transcendental nature of tashkīk al-wujūd – are equally true. However, the complex relation between them needs to be more clearly spelled out. In particular, the claim that tashkīk al-wujūd applies to wujūd as a lāzim of a māhiyya needs to be problematized and investigated in more detail. It will appear that, while generally true, or true in most cases, this rule nevertheless applies to Avicenna’s general ontology, and more precisely to the contingent beings. Yet, his framework also allows for certain crucial exceptions, namely, God and the pure quiddities, with regard to which certain modulated notions, such as existence, oneness, and priority, for example, are predicated essentially rather than accidentally and as concomitants.35 The impact of these exceptions is considerable: they shift the focus of tashkīk away from univocity and more in the direction of a moderate form of equivocity, while at the 33. WISNOVSKY, “On the Emergence,” p. 293, argues against the transcendental scope of tashkīk al-wujūd, which in his eyes is limited to the categories. 34. Some of this evidence has already been discussed by TREIGER, “Avicenna’s Notion,” pp. 360–363; cf. JANOS, Avicenna, pp. 449–460. I reiterate some of the important points at the end of this paper. 35. Extending the argument to encompass the pure quiddities is admittedly more controversial, but the various parts of the argument developed in this paper need not be taken together, i.e., one could limit the argumentation to God and ignore the pure quiddities, even though I would claim that there is a direct connection between these two topics. For a detailed discussion of the pure quiddities in Avicenna’s philosophy, see JANOS, Avicenna.

106

DAMIEN JANOS

same time maintaining the idea of a core meaning; and they suggest that tashkīk al-wujūd – and by extension tashkīk as it applies to other modulated notions, such as oneness, priority, etc. – is predicated essentially of certain entities and encompasses certain pure essences within its scope. As we will see, this harmonizing reading of the essence-existence identity thesis and the transcendental scope of tashkīk al-wujūd seems to cohere with Avicenna’s identification of the various criteria or aspects that constitute the very core of tashkīk and tashkīk al-wujūd – these include, in addition to existence itself, unity, necessity, universality, and priority – all of which apply to God and/or the pure quiddities in an essential way. In light of the foregoing, the question of how tashkīk (and especially tashkīk al-wujūd) relate to quiddity and its concomitants definitely calls for a more detailed investigation in Avicenna’s philosophy. As a corollary, it will also lead to a better understanding of the reception of Avicenna’s theories in the post-classical tradition, from Bahmanyār, al-Shahrastānī, al-Suhrawardī, and Fakhr al-Dīn al-Rāzī, all the way to Mullā Ṣadrā. In fact, the post-Avicennian debate about tashkīk al-wujūd, whether it is acknowledged explicitly or not in the primary sources, seems to have focused largely on the question of whether one should regard existence solely as an external attribute of essence, or whether it can refer to a kind of essential existence (wujūd dhātī) as well. Posited in a more general way, the issue was whether tashkīk focuses solely on the external concomitants of essence or can modulate essence itself. Are waḥda and wujūd external concomitants of God’s essence or are they identical to It? Are these notions predicated accidentally or essentially of things? Is there such a thing as essential oneness and essential existence? These are some of the key points of contention between Bahmanyār, al-Shahrastānī, al-Rāzī, and al-Ṭūsī, which implicitly involve the nature, function, and scope of tashkīk. 3. Modulation (tashkīk) and the state of the concomitant (lāzim) Let us now return to the question raised at the beginning of this paper, namely, how tashkīk generally and tashkīk al-wujūd specifically relate to Avicenna’s māhiyya-lawāzim distinction. This implies shedding additional light on the state of the concomitant and on what it

TASHKĪK AL-WUJŪD AND THE LAWĀZIM IN METAPHYSICS

107

is to be a concomitant. In Avicenna’s metaphysics, concomitants are the things that are not necessary either for the constitution and subsistence (qiwām) of quiddity as such or for its conceptualization (taṣawwur, ta῾aqqul). They are opposed to the internal constituents (muqawwimāt) of quiddity, which enter into the definition of a thing and participate directly in its conceptualization. Concomitants are therefore external or extrinsic to, as well as derived from, the essence they qualify. Thus, while the differentia “rational” and the genus “animal” are internal constituents of the quiddity “humanness” (insāniyya), oneness, actuality, possibility and, according to Avicenna, existence (wujūd), in contrast, are considered external concomitants of it. None of these notions is necessary or a condition to conceptualize what a human being is in itself. This separation between the internal constituents and the external concomitants of quiddity provides the foundation for Avicenna’s theory of the pure quiddity (al-māhiyya min ḥaythu hiya hiya). The latter can be apprehended strictly in itself and without any reference to the accidents and external concomitants that attach to it when essence is considered as a concrete and particular entity (῾ayn), existent (mawjūd), or even thing (shay᾿ ). In this connection, the distinction between the pure quiddity and the lawāzim appears prominently in Avicenna’s ontology, both with regard to mental and concrete existence, where it is combined with a mereological approach. The mental universal, for example, consists fundamentally of a pure quiddity (e.g., farasiyya or “horseness”) combined with a variety of concomitants: “universality” (kulliyya); “oneness” (waḥda); “speciesness” (naw῾iyya); “actuality” (fi῾liyya); “mental existence” (wujūd dhihnī or fī-l-῾aql ), etc. Likewise, in the case of concrete existents, the concomitants proper to this kind of existence – “particularity” (juz᾿iyya) or “individuality” (shakhṣiyya), “oneness” (waḥda), etc. – are related to the pure quiddity in the same way (at least conceptually) that the concomitants of the mental existent are related to the pure quiddity. In brief, a concomitant (lāḥiq, lāzim, ῾āriḍ) according to Avicenna is something that lies outside the quiddity or essence of a thing. It is non-constitutive, external or extrinsic, and can also be conceptually added to (zā᾿id ῾alā) the pure quiddity. According to Avicenna, existence (wujūd) is such an external concomitant, because it is not identical to the essence or quiddity of a thing

108

DAMIEN JANOS

or to its constituents. Moreover, it is acquired from an external cause. As the shaykh puts it in a passage of Mubāḥathāt, existence is “a nonconstitutive concomitant” (lāzim ghayr muqawwim) of essence.36 In Maqūlāt, he states that it is “a thing that is concomitant to the quiddities (amr lāzim lahā).”37 A similar view is expressed in Ilāhiyyāt, V, 2: As for the nature of human inasmuch as it is human, it follows as a concomitant that it should be an existent, even though [the fact] that it is an existent is neither identical to its being human nor included therein.38

In that sense, the concomitant also fulfills one criterion associated with accidentality in Avicenna’s philosophy, which revolves around the idea of being “external to” and “non-constitutive of” a thing’s essence.39 This, it should be noted, is not tantamount to being an accident in the full sense, in the way that “black” or “musical” are accidents of Zayd. Rather, the concomitants are accidental in that they are external to the quiddity, but, unlike true accidents such as “musical,” they necessarily become actualized and realized once that quiddity is instantiated in existence, either in the concrete world or in the mind, or what Avicenna calls al-wujūdān (the “two modes or aspects of existence,” i.e., mental and concrete existence). As the shaykh puts it in the Manṭiq al-mashriqiyyīn: The constituent and the concomitant (al-muqawwim wa-l-lāzim) each have in common that they never separate from the thing [i.e., they always accompany the thing in existence]; and the concomitant and the accident (al-lāzim wa-l-῾āriḍ) each have in common that they are external to the [essential] reality of the thing and follow it (khārij ῾an ḥaqīqat al-shay᾿ lāḥiq ba῾dahā).40

Consequently, there cannot be a thing (shay᾿ ) that is not also an existent characterized by these properties or concomitants. Nevertheless, 36. AVICENNA, Mubāḥathāt, ed. M. BĪDĀRFAR, Qum 1992, p. 218, section 647; cf. TREIGER, “Avicenna’s Notion,” p. 362. 37. AVICENNA, Maqūlāt, II, 1, p. 61. 3–4. 38. AVICENNA, Ilāhiyyāt, V, 2, p. 158. 1–2, translation by Marmura, slightly revised. 39. It should be noted that Avicenna also relies on this language of extrinsicity and accidendality to describe the various properties, attributes, and states of existence or the existent in Ilāhiyyāt, I, 1–5; see A. BERTOLACCI, The Reception of Aristotle’s Metaphysics in Avicenna’s Kitāb al-Šifā᾿. A Milestone of Western Metaphysical Thought, Leiden 2006, pp. 613–616. I return to this important point shortly. 40. AVICENNA, Manṭiq al-mashriqiyyīn, [n. e.], Qum 1985, p. 14. 5–7.

TASHKĪK AL-WUJŪD AND THE LAWĀZIM IN METAPHYSICS

109

oneness, priority, and the other concomitants remain accidental in that they are external to and non-constitutive of the essence. This explains why Avicenna often uses the root ῾-r-ḍ to describe them (῾āriḍ, ῾awāriḍ, ῾araḍī, etc.), often in compound expressions (e.g., al-῾awāriḍ al-lāzima, which appears at the beginning of Ilāhiyyāt, IV, 1). It should be noted that most metaphysical notions discussed by Avicenna fall in this category: this is the case notably of oneness, multiplicity, actuality, potentiality, possibility, necessity, priority, and posteriority, not to mention existence itself. What is more, because they are external and non-constitutive, these notions by the same token apply to a variety of beings – not to say all beings – and, to boot, they apply to these beings in a different way, which explains why Avicenna tends to regard them as modulated. In fact, when it comes to tashkīk al-wujūd, there seems to be a direct correlation between the fact that wujūd is modulated and the fact that it is a concomitant of essence. Existence is modulated because it is predicated as a concomitant of various kinds of beings. Two passages, one from Maqūlāt and one from Mubāḥathāt, can be relied upon to further substantiate this point. In Maqūlāt, II, 1, Avicenna enumerates different ways in which existence is modulated, and immediately afterwards describes it as an external concomitant (amr lāzim) of quiddity.41 Elsewhere, he states that “existence does not apply to the same degree” to the various categories, with the result that “it is not a genus” or constitutive of quiddity.42 The obvious inference one can draw here is that wujūd is modulated because it has the status of a concomitant.43 The second passage appears in section 648 of Mubāḥathāt and was already quoted and analyzed by Treiger in his article. Having clarified in the preceding section (647) that wujūd is a “non-constitutive 41. AVICENNA, Maqūlāt, II, 1, pp. 60. 17–61. 4. 42. AVICENNA, Maqūlāt, II, 1, p. 61. 1–2; cf. ZAMBONI, “Is Existence?” p. 132. 43. To some extent, the reverse is true as well: wujūd is extrinsic and a concomitant, because it is modulated. In his paper, ZAMBONI, “Is Existence?” p. 123 note 5, refers to Avicenna’s “argument from modulation” to show that wujūd is extrinsic to essence; see also p. 132. At any rate, the basic point here is that the two notions of modulation and extrinsicity are interconnected. As we will see later, Fakhr al-Dīn al-Rāzī also stressed this connection between tashkīk and the concomitants. As BENEVICH, “The Necessary Existence,” pp. 134–135, explains, “His [al-Rāzī’s] explanation of tashkīk is very brief: it is priority in respect of ‘the extent of concomitants and effects’ (kathrat al-lawāzim wa-l-āthār).”

110

DAMIEN JANOS

concomitant” (lāzim ghayr muqawwim), Avicenna proceeds to explicitly connect this lāzim with tashkīk al-wujūd: What is more, because existence is predicated of what is below it according to modulation (bi-l-tashkīk), it follows necessarily that each existent must be differentiated from the other through its essence, in the way [for example that] blackness is differentiated from extension. Such two [concepts] do not share a constitutive common [notion], but may share a non-constitutive concomitant [i.e., wujūd].44

Here the connection between the modulated predication of wujūd and its status as an external concomitant of essence is clearly established. Although it is applied to and shared by all beings, existence is a notion that can be modulated from one instance to the other precisely because it is an external concomitant of the essence and is not constitutive of it. The differentiation or modulation of existence is nuanced and relative and spreads across the ontological spectrum, whereas fundamental differences in nature (e.g., humanness, blackness, and extension) will necessarily have to invoke essences and their constitutive elements (muqawwimāt). The modulated notion of wujūd will not be a sufficient criterion to account for the various essences and the fundamentally different natures between beings, because it is an external concomitant of these essences. The strong correlation Avicenna establishes between tashkīk and concomitance (luzūm, luḥūq) seems conceptually justified. After all, tashkīk is intended to “modulate” notions that apply to a variety of contingent and caused things. For example, tashkīk al-wujūd, tashkīk al-waḥda, and tashkīk al-taqaddum aim to “modulate” the way in which existence, oneness, and priority respectively are predicated of a wide variety of things, such as the “celestial bodies,” as opposed, say, to “this horse” or “that human being.” What is more, tashkīk serves to modulate the concomitants and accidents of particular instantiations of the same essence, such as horses A and B, or particular concrete horse A and universal mental horse B, by resorting to the whole gamut of aspects or modes it posits. Given that essence is unchangeable and predicated univocally of things, particular instantiations of a given 44. AVICENNA, Mubāḥathāt, p. 218, section 648. For a discussion of this passage and its relevance for the thesis of the transcendental nature of modulation, see TREIGER, “Avicenna’s Notion,” pp. 362–363, from which the present translation is taken.

TASHKĪK AL-WUJŪD AND THE LAWĀZIM IN METAPHYSICS

111

essence differ and can be individualized only through their concomitants and accidents. Hence, it is understandable that tashkīk would focus on what is different and variable in these beings, i.e., their concomitants and accidents. In turn, this could explain why Avicenna is prone to associate tashkīk with notions expressing difference and diversity (ikhtilāf ) and even equivocity (ishtirāk, ittifāq). Interestingly, this correlation between tashkīk and wujūd taken as a lāzim can be extended to many other notions as well. In Avicenna’s philosophy, modulation seems to apply chiefly to things that logically and ontologically correspond to concomitants, that is, to things that are external to the essence. The two central notions of Avicenna’s metaphysics to which tashkīk is consistently applied, wujūd and waḥda, are the stock examples of a concomitant (lāzim and lāḥiq) of essence, as was shown previously. But upon further investigation, it turns out that the vast majority of notions Avicenna classifies as modulated (mushakkika) pertain to the external attributes of a quiddity, whether these be concomitants (lawāzim) or mere accidents (a῾rāḍ). Among the list of terms Avicenna regards as modulated, one finds – in addition to “existence” (wujūd) and “existent” (mawjūd) – “one” (wāḥid) and “oneness” (waḥda), “possibility” or “contingency” (imkān), “form” (ṣūra), “matter” (mādda), “privation” (῾adam), “perfection” (kamāl), “substance” (jawhar), “number” (῾adad), “quantity” (kamm), “healthy” (ṣiḥḥī), “medical” (ṭibbī), “power” or “potentiality” (quwwa), “priority and posteriority” (al-taqaddum wa-l-ta᾿akhkhur), and “accident” (῾araḍ).45 A good number of these notions clearly relate to things external to essence, and some of them are even accidents in a straightforward sense (“healthy,” “medical”). Others refer to the Aristotelian categories (“quantity,” “motion”), and the generic notion of “accident” (῾araḍ) itself is mentioned as an ism mushakkik. It is true that there are other notions here that do not fit neatly in that picture, namely, “substance,” “form,” and “matter,” which philosophers since Aristotle have opposed to the notion of accidentality, and which instead express substantiality. Indeed, Avicenna also seems to extend tashkīk to substance generally construed. But for all intents and purposes, most of the notions Avicenna describes as mushakkika correspond to things that are external 45. See JANOS, “Avicenna on Equivocity,” which provides specific references to the primary sources.

112

DAMIEN JANOS

to essence. Hence, it appears that there is a narrow connection in his mind between these two considerations. In view of this, to the questions: To what exactly is tashkīk applied in Avicenna’s philosophy? To what kind of objects does it extend?, the first and most general answer will have to be: tashkīk applies to logically non-univocal and non-constitutive notions that can be predicated in different ways of different things. Translated to the sphere of metaphysics, this means that tashkīk will focus on the external concomitants of an essence that are caused to exist when that thing’s essence becomes realized. In that sense, tashkīk intersects directly with Avicenna’s māhiyya-lawāzim paradigm. For this reason, and also because it represents the main framework through which he disambiguates and discusses existence, its importance in his philosophy can hardly be exaggerated. Now that it has been established that some of the core terms and notions that ground Avicenna’s theory of modulation (tashkīk) typically refer to the concomitants of essence (with the crucial exceptions noted above), we may return to the paradigmatic case of wujūd and analyze it in more detail. Returning to the “anatomy” of tashkīk al-wujūd, one notices that modulation also extends to several notions that are part of the very structure or framework of that theory. In other words, tashkīk applies not only to the primary notion of wujūd per se, but also to the secondary notions that are used to “modulate” this primary notion, so that both wujūd and the aspects that modulate it are said to be mushakkika.46 Of these notions, three are explicitly said to be modulated: “priority and posteriority” (al-taqaddum wa-l-ta᾿akhkhur), and “possibility” (or contingency, imkān). Recall that Avicenna identifies priority and posteriority, as well as possibility and necessity, as “aspects” (anḥā᾿, aḥkām) of tashkīk al-wujūd, which participate directly in the task of modulating existence from one object to the other (see Table 1). Now, at Ilāhiyyāt, IV, 1, the shaykh explicitly describes priority and posteriority themselves as being “modulated” or “quasi-modulated” (takādu an tajtami῾a ῾alā sabīl al-tashkīk fī shay᾿ ).47 46. This important point has to my knowledge not been acknowledged in the modern literature. 47. AVICENNA, Ilāhiyyāt, IV, 1, p. 124. 9–10. At Jadal, p. 119. 8–9, Avicenna uses the phrase al-taqaddum wa-l-ta᾿akhkhur fī l-tashkīk.

TASHKĪK AL-WUJŪD AND THE LAWĀZIM IN METAPHYSICS

113

As for possibility or imkān, it is presented as a modulated notion in Qiyās.48 Upon reflection, the idea that the notions participating in tashkīk al-wujūd are themselves modulated is not altogether surprising. In fact, it is only reasonable to assume that the notions that most strongly modulate wujūd (i.e., priority, posteriority, and possibility) be themselves modulated. What is perhaps more arresting, however, is the realization that some of the aspects of tashkīk al-wujūd overlap with what Avicenna typically calls the concomitants (lawāzim, lawāḥiq), accidents (a῾rāḍ, ῾awāriḍ), states (aḥwāl), and properties (khawāṣṣ) of existence (wujūd) or the existent (mawjūd) in Ilāhiyyāt, I, 1–5. Recall that these notions are mentioned in the context of Avicenna’s description of the subject matter and contents of the metaphysical inquiry. Although metaphysics focuses primarily on “being qua being,” or on “the existent inasmuch as it is existent” (al-mawjūd bi-mā huwa mawjūd), it also investigates the attributes and accidents of beings. For the sake of convenience, I provide below a table that captures the notional overlap between these features of Avicenna’s doctrine: Table 2: Notional overlap between existence and its concomitants and tashkīk al-wujūd

The concomitants (lawāzim, lawāḥiq), properties (khawāṣṣ), accidents (a῾rāḍ, Notions included in tashkīk al-wujūd ῾awāriḍ), and states (aḥwāl) of “the existent” (mawjūd) and “existence” (wujūd) in the Ilāhiyyāt “the universal and the particular” (al-kullī wa-l-juz᾿ī) (I, 1, section 15; I, 2, section 13; V, 1 section 1) “potentiality and actuality” (al-quwwa wa-l-fi῾l) (I, 1, section 15; I, 2, section 13; IV, 1, section 20)

48. AVICENNA, Qiyās, al-Manṭiq, al-Shifā᾿, ed. S. ZĀYID, Cairo 1964, pp. 168‒169. In Burhān, al-Manṭiq, al-Shifā᾿, ed. A. ῾AFĪFĪ, Cairo 1956, p. 328, it is quwwa, a notion closely associated with imkān in Avicenna’s metaphysics, which is said to be modulated.

114

DAMIEN JANOS

“possibility and necessity” (al-imkān “necessity and possibility” (al-wujūb wa-l-wujūb) (I, 1, section 15); wa-l-imkān) (Ilāhiyyāt, VI, 3) “the possible and the necessary” (almumkin wa-l-wājib) (I, 2, section 13); the state of the necessary and the state of the possible (ḥāl al-wujūb wa-ḥāl al-imkān) (I, 4, section 1) “the one and the many” (al-wāḥid wa-l-kathīr) (I, 2, section 13); “the one” (al-wāḥid) (I, 4, sections 5–6) “priority and posteriority” (al-taqaddum wa-l-ta᾿akhkhur) (I, 4, sections 5–6; IV, 1, sections 1–2)

“priority and posteriority” (al-taqaddum wa-l-ta᾿akhkhur) Maqūlāt of al-Mukhtaṣar al-awsaṭ fī l-manṭiq; Maqūlāt, I, 2; Jadal, II, 2; Dānesh-nāmeh, section 11; al-Samā῾ al-ṭabī῾ī, I, 3 and II, 2; Ilāhiyyāt, VI, 3; K. al-Hidāya

“the cause and the caused” (al-῾illa wa-l-ma῾lūl) (VI, 1, section 1)

“worthier” (al-awlā) (Mukhtaṣar al-awsaṭ fī manṭiq, p. 326. 14–15); “worthier and more deserving” (al-awlā wa-l-aḥrā) (Maqūlāt, 1, 2); “autonomy and need” (al-istighnā᾿ wa-l-ḥājja) (Ilāhiyyāt, VI, 3, section 27)49

“the state of what is essential and what is accidental” (ḥāl alladhī bi-l-dhāt wa-alladhī bi-l-῾araḍ) (I, 4, section 1)

“Essentially and accidentally” (bi-l-dhāt wa bi-l-῾araḍ or “in itself and through another” (bi-dhātihi wa-bi-ghayrihi) (Maqūlāt, I, 2; Jadal, II, 2; Dānesh-nāmeh, section 11)

What is of interest here is not the variegated terminology Avicenna relies on in order to describe the attributes of wujūd and mawjūd, which was already analyzed by Amos Bertolacci in his masterful 49. It remains unclear whether these notions are to be related primarily to causality. TREIGER, “Avicenna’s Notion,” p. 357, takes it to be so, I think rightly, in light of what Avicenna says in Ilāhiyyāt, VI, 3, p. 213. But they can also be correlated to substance and accident, which often appear side by side with the notion of worthiness and the bi-dhāt and bi-l-῾araḍ distinction. Nevertheless, they do seem to bear a direct relation to causality and more specifically to the cause’s relation to its effect, especially according to a Neoplatonic reading.

TASHKĪK AL-WUJŪD AND THE LAWĀZIM IN METAPHYSICS

115

monograph,50 but the notions themselves that are identified as falling within the scope of the metaphysical investigation. As we can see, wujūd and its direct concomitants are common both to tashkīk al-wujūd and the description of the subject matter of metaphysics Avicenna provides in Ilāhiyyāt, I, 1–5. In fact, there seems to be a direct connection between these sets of notions, and at least some of them overlap neatly: this is the case of priority, posteriority, and possibility, which are crucial aspects of tashkīk al-wujūd, and also sought in metaphysics in connection with the existent or mawjūd.51 At Ilāhiyyāt, IV, 1 the shaykh signals that the discussion from now on will switch from the “species” (i.e., what can be loosely compared to the species) of existence and oneness (substance, quality, and quantity, i.e., the categories in general) to “the properties and accidents that are necessary concomitants” (al-khawāṣṣ wa-l-῾awāriḍ al-lāzima) of existence, of which priority and posteriority are a key component. He then proceeds to define priority and posteriority as modulated notions.52 Hence, priority and posteriority are “aspects” (anḥā᾿, aḥkām) of tashkīk al-wujūd, as well as “accidents” and “concomitants” of existence studied as part of the subject matter of metaphysics. In addition, we observe that the “bi-l-dhāt and bi-l-῾araḍ distinction” is also common to the metaphysical inquiry and to the framework of tashkīk al-wujūd. Finally, there are serious reasons to connect two other aspects of tashkīk al-wujūd, “degrees of deservingness” (al-awlā wa-l-aḥrā) and “autonomy and need” (al-istighnā᾿ wa-l-ḥājja), with another “accident” 50. See BERTOLACCI, The Reception, Appendix F, pp. 613–616. I am not interested here in disambiguating the technical nuances of the terms lawāzim, lawāḥiq, a῾rāḍ, ῾awāriḍ, aḥwāl, and khawāṣṣ that Avicenna uses to describe these notions in connection with wujūd and mawjūd in Ilāhiyyāt, I, 1–5. Naturally, each carries a specific meaning and should be translated differently into English. But the key points here are that these terms stress the external or extrinsic status of the concomitants vis-à-vis essence and that Avicenna uses them interchangeably to refer to the concomitants of existence, sometimes in compound formulae, such as al-῾awāriḍ al-lāzima, etc. 51. AVICENNA, Ilāhiyyāt, I, 4, p. 20. 5–11; I, 5, p. 27. 13–17. For priority and posteriority as key elements of tashkīk al-wujūd, see Maqūlāt, 1, 2: “priority and posteriority” (al-taqaddum wa-l-ta᾿akhkhur); Dānesh-nāmeh (section 11): “priority and posteriority” (pīsh-wa-pas); and Ilāhiyyāt, VI, 3, section 27: “priority and posteriority” (al-taqaddum wa-l-ta᾿akhkhur). See the latter as well for “necessity and possibility” (al-wujūb wa-l-imkān). 52. AVICENNA, Ilāhiyyāt, IV, 1, p. 124. 7. In fact, this section opens with a discussion of priority and posteriority.

116

DAMIEN JANOS

or “attribute of mawjūd, namely, “the cause and the caused” (al-῾illa wa-l-ma῾lūl).53 Here a word should be said about oneness and multiplicity (al-waḥda wa-l-kathra), which represent a special case. “The one” (wāḥid) is explicitly said to be modulated at the beginning of Ilāhiyyāt, III, 2.54 Because “the multiple” is opposed to and derives from “the one,” it is likely that Avicenna regards it as modulated as well. Surprisingly, however, and as is apparent from Table 1, Avicenna nowhere expressly links these two notions to tashkīk al-wujūd, and upon further reflection, it is unclear how they relate to it. On the one hand, there are reasons to think that they participate in that framework as well. After all, oneness and multiplicity are described as concomitants, accidents, and attributes of mawjūd in Ilāhiyyāt, I, 2 and 4. When connected with ontological modulation, waḥda and kathra ask the question whether an existent is “one” or “multiple,” and also, by the same token, whether it is “simple” or “complex.” Hence, when approached from this angle, it is likely that oneness and multiplicity participate directly in tashkīk al-wujūd, as do, presumably, all the other concomitants of mawjūd and wujūd that Avicenna lists in the Ilāhiyyāt. On the other hand, the shaykh sometimes stresses the co-extensionality and co-implicativeness of wujūd and waḥda in his metaphysics, to such an extent that his perception of that science can be said to amount to both an ontology and a henology.55 On this view, the theory of the modulation of oneness would be distinct from the theory of the 53. See AVICENNA, Ilāhiyyāt, VI, 1, p. 194. 6–8, where “the cause and the caused” are described as “among the things that attach to the existent inasmuch as it is existent” (min al-lawāḥiq allatī talḥaqu l-mawjūd bi-mā huwa mawjūd). The ground for connecting these two sets of notions of tashkīk al-wujūd to causality comes in part from Avicenna himself, who at Ilāhiyyāt, VI, 3, p. 213. 17–18, glosses al-istighnā᾿ wa-l-ḥājja in terms of “the cause” and “the effect.” 54. AVICENNA, Ilāhiyyāt, III, 2, p. 74. 4–6. 55. That is, wujūd and waḥda are co-extensional, but remain intensionally distinct. For statements pointing to the importance of waḥda for the subject matter of metaphysics, as well as to its equal or quasi-equal status with wujūd, see, among other passages, AVICENNA, Ilāhiyyāt, I, 2, p. 11. 18: “it [first philosophy or metaphysics] is knowledge of … the first things in generality (awwal al-umūr fī l-῾umūm), namely, existence and oneness (al-wujūd wa-l-waḥda)”; and VII, 1, p. 236. 8–10: “The one and the existent may be equivalent (yatasāwiyān) in being predicates of things, so that everything that is said to be an existent from one consideration is, from a certain consideration, [also] correctly said to be one.” See also BERTOLACCI, The Reception, pp. 162–169.

TASHKĪK AL-WUJŪD AND THE LAWĀZIM IN METAPHYSICS

117

modulation of existence. This interpretation finds traction in the fact that Avicenna usually mentions wujūd and waḥda as two separate examples – albeit side by side – in his explanations of tashkīk, as well as in the aforementioned fact that oneness is not explicitly listed as an aspect (ḥukm, naḥw) of tashkīk al-wujūd. And after all, oneness, like existence, is both transcategorical (i.e., it applies to all substances and accidents; see Ilāhiyyāt, III, 3) and supracategorical or transcendental (it applies also to the immaterial beings and to God, who is neither a substance nor an accident). Hence, it does make sense that waḥda be granted a central position in the metaphysical investigation and its own theory of tashkīk. Furthermore, it appears that most of the notions that Avicenna describes as concomitants or accidents of the existent in Ilāhiyyāt, I, 1–5 – already listed in Table 2 – are also modulated. Some are said explicitly to be modulated (this is the case of oneness, priority and posteriority, as well as potentiality and possibility), while others are implicitly assumed to be such (this is the case of universality and multiplicity).56 Ultimately, this means that tashkīk characterizes most if not all of the notions that are presented as forming part and parcel of the subject matter of metaphysics. To recap: the previous analysis has established an interrelationship between 1/ modulation or tashkīk, 2/ the notions that participate in and are subsumed under tashkīk al-wujūd, and 3/ the notions that are described as concomitants or accidents of wujūd and mawjūd in Avicenna’s Ilāhiyyāt. To put it differently, modulation seems to underpin the very subject matter of metaphysics as exposed in Ilāhiyyāt, I, 1–5. This conceptual nexus had hitherto not been clearly identified and emphasized in the scholarship. Yet, it seems to represent a fundamental aspect of the way Avicenna conceives of the metaphysical investigation. 56. For oneness, see KALBARCZYK, Maqūlāt of al-Mukhtaṣar al-awsaṭ, p. 326; AVIQiyās, pp. 168–169, and ID., Ilāhiyyāt, III, 2, p. 74; for priority and posteriority, see ID., Ilāhiyyāt, IV, 1 p. 124; for possibility, see ID., Qiyās, pp. 168–169; for potentiality, see ID., Burhān, p. 368. The evidence for regarding universality (kulliyya) and the universal (kullī) as modulated notions, or as modulated/equivocal notions, is quite compelling. At the beginning of Ilāhiyyāt, V, 1, p. 148, Avicenna states that “the universal is said in three ways,” and at the beginning of Ilāhiyyāt, V, 2, p. 157, he adds that the universal has several meanings (ma῾ānī). As for multiplicity, Avicenna defines it as a derivative and the contrary of waḥda, which is a core modulated notion, so one assumes that multiplicity is equally modulated due to its narrow relation to oneness.

CENNA,

118

DAMIEN JANOS

The repercussions of the foregoing interpretation are considerable, because they could potentially lead to a new way of conceiving of the place of tashkīk al-wujūd in Avicenna’s metaphysics. Although Avicenna himself does not emphasize the connection between the lawāzim of wujūd and the aspects of tashkīk al-wujūd, the reconstruction provided above seems fully substantiated by the evidence found in his corpus, as well as conceptually consistent with his approach to metaphysics. There is an implicit recognition on his part that many of the accidents or concomitants of being qua being are fundamentally modulated notions and that some of these notions also participate in the framework of tashkīk al-wujūd. Furthermore, it is remarkable that all of these notions are grounded in the concept of “the concomitant” (sing., lāzim, lāḥiq). In fact, many of the notions described in Ilāhiyyāt participate directly in the modulation of existence precisely because they are accidents and concomitants of wujūd. The implication is that the notions and aspects subsumed under tashkīk al-wujūd are to some extent interchangeable with those that constitute the subject matter of metaphysics. In brief, tashkīk al-wujūd lies at the heart of Avicenna’s metaphysics and is another way of referring to the main contents of this science. One additional corroborating remark in this regard is Avicenna’s insistence that wujūd and waḥda are the primary notions investigated by the metaphysical science and constitute its main subject matter, while also representing the paradigmatic cases of modulation or tashkīk.57 Nevertheless, it would seem that the theory of tashkīk al-wujūd goes one step further than the mere description of the contents of metaphysics provided in Book I of the Ilāhiyyāt: while the latter presents these notions as attributes or concomitants of existence or the existent, the theory of tashkīk al-wujūd presupposes that wujūd itself is to be regarded as a concomitant and predicated of things as one of their external attributes. More specifically, the presupposition is that wujūd (and by extension the concomitants of wujūd) are concomitants of the essence or quiddity of a thing. I return to this point in the last section of the paper.

57. For wujūd and waḥda as the main objects studied in metaphysics, see AVICENNA, Ilāhiyyāt, I, 2, p. 11. 17–18; I, 4, p. 21. 2–3; and IV, 1, p. 124. 6–8.

TASHKĪK AL-WUJŪD AND THE LAWĀZIM IN METAPHYSICS

119

4. Tashkīk, substance and accident, and the subject matter of metaphysics Up to now, the analysis has focused on tashkīk as it applies to the accidents and concomitants of existence or the existent, as well as on the anatomy of tashkīk al-wujūd. As we saw, the general framework of tashkīk al-wujūd overlaps significantly with the content and subject matter of metaphysics. But for tashkīk al-wujūd to overlap perfectly with the subject matter of metaphysics, it would need to encompass substances themselves in addition to the accidents and concomitants of existence. If tashkīk occupies a central place in Avicenna’s conception of metaphysics – which appears to be the case – then one would expect substance as well to be subjectable to tashkīk.58 Two points may be stressed in this regard. The first requires that we expand the analysis of tashkīk to other notions apart from existence. As mentioned previously, Avicenna regards a whole array of terms to be modulated, among which are some that designate substance, chiefly jawhar, ṣūra, and mādda.59 Rather than the state of the concomitant and accidentality, these notions express substantiality. According to the shaykh, these notions are also modulated or even slightly equivocal, given that they do not apply in a semantically uniform and identical way to all things. For instance, substance may be said of form, of matter, of the compound of form and matter, as well as in connection with the essence of a thing. It applies to concrete existents as well as to mental ones, to perishable beings as well as to eternal ones, albeit in different ways. The same of course can be said of ṣūra. In Kitāb al-Ḥudūd, Avicenna cautions that ṣūra is “an equivocal term said in many ways” (ism mushtarak yuqālu ῾alā ma῾ānin).60 But in al-Samā῾ al-ṭabī῾ī, he instead denotes it as an ism mushakkik, together with matter and privation.61 In brief, we can ascertain that there are some 58. Admittedly, it would also have to be “transcendental,” as Treiger put it, since the First is one of the beings investigated in metaphysics. 59. See AVICENNA, Maqūlāt, pp. 36 and 74; ID., The Physics of The Healing (al-Šifā᾿, al-Samā῾ al-ṭabī῾ī), Books I & II. A parallel English-Arabic text translated, introduced, and annotated by J. MCGINNIS, Provo, Utah 2009, I, 3, p. 31 (henceforth al-Samā῾ al-ṭabī῾ī). See also JANOS, “Avicenna on Modulation and Equivocity.” For the substantiality of matter in Avicenna, see A. LAMMER, The Elements of Avicenna’s Physics: Greek Sources and Arabic Innovations, Berlin / Boston 2018, pp. 118‒120. 60. AVICENNA, Kitāb al-Ḥudūd, ed. A.-M. GOICHON, Cairo 1963, p. 16. 5. 61. AVICENNA, al-Samā῾ al-ṭabī῾ī, I, 3, p. 31.

120

DAMIEN JANOS

modulated-equivocal terms that designate substance and substantiality, rather than the concomitants of existence. In that sense, and on a purely logical and terminological level, tashkīk is not exclusively limited to the lawāzim of wujūd. The second point is that Avicenna himself explicitly states that tashkīk al-wujūd focuses on the existence of substances in addition to the existence of concomitants and accidents. In fact, this represents a crucial qualification of tashkīk al-wujūd, which usually appears hand in hand with the aspect of priority and posteriority (al-taqaddum wa-l-ta᾿akhkhur). “Existence” and “the existent,” Avicenna tells us, apply first to substances and only secondarily or in a posterior way to accidents; and they apply to certain substances and accidents before others.62 Thus, priority and posteriority in existence, a notion that harks back to Aristotle himself, is applied differently to substances and accidents within the frame of tashkīk al-wujūd. Accordingly, some substances are prior and more worthy than others, and the various accidents and concomitants that characterize mawjūd and also constitute the aspects of tashkīk al-wujūd (priority, actuality, particularity, oneness, etc.) in all cases pertain primarily to substances. Recall that Avicenna compares substance, quality, and quantity to the “quasi-species” of the “quasi-genus” wujūd. If tashkīk applies to wujūd, as it obviously does in the framework of tashkīk al-wujūd, it will necessarily also include substances. So this is basically saying the same thing than that mawjūd is modulated, when mawjūd is taken to mean a substance. Avicenna sums this up neatly in section 11 of the Dānesh-nāmeh when he states that “existence belongs first to substance, and only through the mediation of substance, to quantity, quality, and relation, and through the mediation of these, to the rest [of the categories] … This [kind of predicate] is called modulated (moshakkek).”63 By implication, this also means that tashkīk al-wujūd examines the existence that belongs essentially to a thing or to the thing in itself (bi-dhātihi, bi-l-dhāt). When Avicenna uses these expressions in connection with substance, as in Ilāhiyyāt, II, 1, it is to stress the fact that 62. See, for example, KALBARCZYK, “The Kitāb al-Maqūlāt,” p. 326 of the Arabic text; AVICENNA, Maqūlāt, I, 2, pp. 10–11; ID., Dānesh-nāmeh, section 11; cf. AVICENNA, Ilāhiyyāt, I, 5, p. 27. 13–15: existence “is predicated with respect to priority and posteriority, first, to the quiddity of substance, and subsequently to what comes after it.” This passage, however, does not explicitly mention tashkīk. 63. The translation is by TREIGER, “Avicenna’s Notion,” p. 359.

TASHKĪK AL-WUJŪD AND THE LAWĀZIM IN METAPHYSICS

121

substance does not exist in a subject (lā fī mawḍū῾), that it is selfsubsistent (qā᾿im bi-nafsihi), and that its existence belongs solely to itself. Accordingly, tashkīk al-wujūd applies both to things that exist not in a subject and have existence in themselves (substances) and to things that exist in a subject and have existence accidentally from another (accidents).64 In the final analysis, then, just as substance and its accidents (as well as wujūd and its ῾awāriḍ or lawāḥiq), constitute the main subject matter of metaphysics, so these notions constitute the main objects of tashkīk al-wujūd. The latter subsumes wujūd and its various “quasispecies” (substance, quantity, quality), as well as its various accidents and states (oneness and multiplicity, possibility and necessity, etc.): that is, it subsumes all of the things Avicenna describes as constituting the subject matter of the metaphysical science in sections I, 1–5 of the Ilāhiyyāt. As a result, tashkīk al-wujūd emerges as the single most important methodological and conceptual theory that is articulated by Avicenna in order to study the subject matter of metaphysics. With regard to the objects it studies, it is for all intents and purposes interchangeable with the other ways the shaykh employs to describe the contents and scope of the metaphysical science.65 One may describe tashkīk as the linguistic, logical, and conceptual framework that is articulated in order to study metaphysics. What did not make this point hitherto more pellucid is the fact that the shaykh discusses these things in different 64. See AVICENNA, Ilāhiyyāt, II, 1, p. 45. 6–8. This point, of course, is not incompatible with regarding existence as a concomitant of the essence. 65. I personally do not share De Haan’s concerns about a potential tension regarding Avicenna’s theory of “the analogy of being” (i.e., ontological modulation) and his presentation of the subject matter of metaphysics (see “The Doctrine,” p. 286). As De Haan explains, modulation implies a certain intensional constriction of the general notion of wujūd, with the consequence that the various articulations of ontological modulation might not correspond to being qua being or absolute being (wujūd muṭlaq). But the latter, for Avicenna, is just a concept in the mind, with no external referent as such, and what metaphysics investigates are really the existents, as well as the attributes and accidents of these existents, which necessarily require a modulated interpretation of wujūd. In this connection, Avicenna’s theory of tashkīk al-wujūd is not only transcategorical, but also (as De Haan himself argues at pp. 279–281) “extracategorical” or transcendental, in that it extends to God’s existence in addition to that of the other immaterial beings. In that sense, tashkīk al-wujūd includes all beings and all instances and aspects falling under wujūd that Avicenna recognizes in Ilāhiyyāt, I, 1–7 (including, that is, God as the Necessary Existent in Itself). Moreover, as shown in this paper, the various accidents and attributes of being are also modulated and fall under tashkīk, which makes the overlap between modulation and the subject matter of metaphysics as it is presented in the Ilāhiyyāt even more compelling.

122

DAMIEN JANOS

parts of his works and does not explicitly stress their connections, leaving the onus on the readers to tie them together. 5. How does tashkīk relate to quiddity? Is there such a thing as the modulation of essence in Avicenna? 5.1. Tashkīk and quiddity in Avicenna’s ontology At this juncture, we need to tackle the important issue of how tashkīk applies to the other side of Avicenna’s māhiyya-lawāzim distinction, if at all, that is, to quiddity. On the basis of the previous discussion, one could assume that tashkīk applies only to the things external to quiddity and to the lawāzim, and that for this reason it is predicated only accidentally (bi-l-῾araḍ) of these things, but this proves not to be the case. Yet, exactly why this is not the case represents a particularly thorny feature of Avicenna’s metaphysics. This question will be approached from two angles: ontology and the notion of quiddity in general, and theology and the notion of God’s quiddity in particular. One key point to initiate the first deliberation is that the concomitants, accidents, or properties of mawjūd and wujūd referred to above can equally be framed as concomitants, accidents, or properties of quiddity (māhiyya). Indeed, Avicenna refers to the concomitants sometimes as “concomitants of quiddity” and other times as “concomitants of existence.” For example, throughout Book I of the Ilāhiyyāt, he designates the concomitants of wujūd or mawjūd as things studied by the science of metaphysics, while in Book V, 1–2, when discussing his theory of the pure quiddity, he expounds on the extrinsic nature of the lawāzim vis-à-vis essence, presenting these notions as lawāzim al-māhiyya. As it turns out, many of these concomitants are the same, and there is considerable overlap between these two accounts, as the table below adequately shows: Table 3: The attributes of wujūd and mawjūd compared with the concomitants of quiddity (māhiyya) in Ilāhiyyāt, V, 1–2 The concomitants (lawāzim, lawāḥiq), properties (khawāṣṣ), accidents (a῾rāḍ, ῾awāriḍ), and states (aḥwāl) of “the existent” (mawjūd) and “existence” (wujūd) in the Ilāhiyyāt

The concomitants (lawāzim, lawāḥiq), properties (khawāṣṣ), accidents (a῾rāḍ, ῾awāriḍ), and states (aḥwāl) of quiddity (māhiyya) in the Ilāhiyyāt

TASHKĪK AL-WUJŪD AND THE LAWĀZIM IN METAPHYSICS

“the universal and the particular” (al-kullī wa-l-juz᾿ī) (I, 1, section 15; I, 2, section 13; V, 1 section 1)

123

“the universal/universality and particularity” (V, 1, p. 149. 10–11); “specific and general” (V,1, p. 149. 15–17; p. 152. 16–18; p. 153. 3–5; p. 154. 17–19; p. 155. 1; V, 2, p. 157. 15–16, etc.)

“potentiality and actuality” “actuality and potentiality” (V, 1, (al-quwwa wa-l-fi῾l) (I, 1, section 15; p. 149. 12–15; p. 152. 16–18; p. 153. 3–5) I, 2, section 13; IV, 1, section 20) “possibility and necessity” (al-imkān wa-l-wujūb) (I, 1, section 15); “the possible and the necessary” (almumkin wa-l-wājib) (I, 2, section 13); the state of the necessary and the state of the possible (ḥāl al-wujūb wa-ḥāl al-imkān) (I, 4, section 1) “the one and the many” (al-wāḥid wa-l-kathīr) (I, 2, section 13); “the one” (al-wāḥid) (I, 4, sections 5–6)

“oneness and plurality” (V, 1, p. 149. 12–15; p. 152. 16–18; p. 153. 11–12)

“priority and posteriority” (al-taqaddum wa-l-ta᾿akhkhur) (I, 4, sections 5–6; IV, 1, sections 1–2)

“priority” (V, 1, p. 153. 9; p. 156. 7)

“the cause and the caused” (al-῾illa wa-l-ma῾lūl) (VI, 1, section 1)

“cause and effect” (V, 1, p. 156. 8)

“the state of what is essential and what is accidental” (ḥāl alladhī bi-l-dhāt wa-alladhī bi-l-῾araḍ) (I, 4, section 1)

“in itself/essentially and accidentally” (V, 1, p. 149. 11; p. 152. 16; p. 153. 7–8; etc.)

That Avicenna refers to these concomitants as pertaining alternatively to essence and existence can presumably be explained in part by the common extensionality of shay᾿ and mawjūd – and to some extent also of māhiyya and wujūd – in his metaphysics. Inasmuch as these notions are co-extensional and co-implicative, the accidents and concomitants of mawjūd/wujūd of Ilāhiyyāt, I, 1–5 can also be said to belong to shay᾿/māhiyya. The upshot is that the aspects of tashkīk al-wujūd surveyed above will also pertain to māhiyya. These considerations apply not only to priority and posteriority, but to most of the other notions mentioned previously, such as possibility and necessity,

124

DAMIEN JANOS

actuality and potentiality, and particularity and universality. Moreover, it is noteworthy that Avicenna uses exactly the same terminology to describe these notions in relation to wujūd and to māhiyya: they are lawāzim, a῾rāḍ, ṣifāt, aḥwāl, etc. Yet, this convertibility of the lawāzim with regard to both wujūd and māhiyya is somewhat confusing. Is the idea here that existence is a concomitant of essence that itself has concomitants? Or that essence has various concomitants alongside one another, one of which is existence? Both of these claims seem to be correct and can be corroborated by what Avicenna himself says. The idea that a concomitant can itself have concomitants seems to be firmly rooted in his metaphysics.66 As we saw above, wujūd, technically speaking, is a lāzim, which itself has lawāzim, lawāḥiq, and a῾rāḍ. The same can be said about oneness (waḥda). Avicenna presents it both as a concomitant of essence and as one from which other concomitants derive. In Ilāhiyyāt, Book III, he explores the various accidental and essential applications of that notion. And at Ilāhiyyāt, VII, 1, he suggests that waḥda has accidents and concomitants of its own, which the metaphysician ought to investigate.67 Interestingly, this idea is borne out at the theological and cosmological levels as well: God’s effects are concomitants (lawāzim) that themselves possess and generate other concomitants (lawāzim). In fact, this seems to be how Avicenna accounts for the proliferation of intelligible multiplicity in the world: concomitants deriving from other concomitants.68 In light of this, there is no logical inconsistency in regarding wujūd as a concomitant of essence alongside many others; or as regarding wujūd as a concomitant of essence that itself yields various other concomitants.69 66. See AVICENNA, al-Ishārāt wa-l-tanbīhāt, ed. S. DUNYĀ, Cairo 1957–60, 4 vols., namaṭ 4, 17, pp. 30–31, where the shaykh explains that quiddity causes attributes (ṣifāt) that themselves can cause other attributes. In that same passage, however, he also cautions that nothing precedes existence. But the point remains that wujūd can be viewed as a lāzim or ṣifa that causes other lawāzim or ṣifāt. 67. AVICENNA, Ilāhiyyāt, VII, 1, pp. 236. 5–237. 2. Avicenna asserts something similar in the case of accidents: it is acceptable that some accidents themselves have other accidents. 68. See JANOS, Avicenna, pp. 653, 686, and 700–701. 69. The real question is whether, for Avicenna, wujūd is a special and unique concomitant of essence, as it is for the Bahshamiyya, who prioritize the ṣifat al-wujūd over other ṣifāt, except for the ṣifat al-dhāt, which has absolute essential and conceptual priority in their system (even over the ṣifat al-wujūd). Generally speaking, it is clear that wujūd

TASHKĪK AL-WUJŪD AND THE LAWĀZIM IN METAPHYSICS

125

In the final analysis, then, all of these notions (including existence) are concomitants (lawāzim, lawāḥiq) of the essence, and all of these notions are also modulated (mushakkika). When approached from this angle as well, a narrow correlation emerges between modulation and extrinsicity vis-à-vis essence. But here the challenge is to determine whether modulation is always and necessarily limited to the external concomitants of the essence, or whether it allows for cases of essential predication and can apply to the essence taken strictly in itself (al-māhiyya min ḥaythu hiya hiya).70 For example, when it comes to existence, does tashkīk al-wujūd treat wujūd always and in all cases as a lāzim of the essence? Or does it tolerate the concept of essential existence? The same question may be posed with regard to oneness. Avicenna typically describes oneness as a concomitant, as something predicated accidentally of things. But is there such a thing as essential oneness? In these potential cases, tashkīk al-wujūd and tashkīk al-waḥda would qualify the very essence or quiddity of a thing, as opposed to its lawāzim. Most of the relevant evidence pertaining to this discussion can be found in Ilāhiyyāt, V, 1. Among other interpretive difficulties, one intriguing aspect of the evidence contained in this chapter is that it seems to disrupt the normal application of the māhiyya-lawāzim paradigm, according to which only the external or extrinsic concomitants of essence can be the object of modulated predication. Moreover, it describes pure quiddity with notions that Avicenna is usually at pains to distinguish from it. What is of interest here is the possibility that the pure quiddity (al-māhiyya min ḥaythu hiya hiya) can be the object of modulation, thereby greatly complexifying Avicenna’s māhiyyalawāzim paradigm. Indeed, of the various notions that are modulated and/or participate in tashkīk al-wujūd in Avicenna’s metaphysics, some seem to be applied essentially to the pure quiddity, as opposed to being treated as external concomitants of it. This is the case in occupies a special status in Avicenna’s metaphysics, since it is the core notion investigated by metaphysics. More specifically, since Avicenna believes that wujūd qua concomitant itself has lawāḥiq and lawāzim that derive from it (i.e., that an object that actually exists has other concomitants, attributes, and accidents attached to it), its status is in fact comparable to that of the Bahshamite ṣifat al-wujūd. For a detailed comparison of these two views, see JANOS, Avicenna, pp. 591–612. 70. In the remainder of this paper, when I use the expression “essential predication,” I mean that modulation applies to the very essence or quiddity itself, as opposed to its external concomitants.

126

DAMIEN JANOS

particular of priority, existence, oneness, and universality. In Ilāhiyyāt, V, 1, Avicenna describes the special status of pure quiddity by resorting to these notions, in a way that makes it clear that they do not apply to it externally and accidentally, but rather essentially. Accordingly, the pure quiddity is essentially and ontologically “prior” vis-à-vis the contingent, particular thing in the concrete world and the complex universal form in the mind. Moreover, it has an essential kind of universality, which is distinct from that of the complex universals, such as the genus “animal” or the species “horse” that is formed in the human mind. Finally, it is said to possess a distinct ontological status and to have its own wujūd.71 Consequently, there are strong reasons to contend that what Avicenna describes as quiddity’s “special existence” (wujūd khāṣṣ), “divine existence” (wujūd ilāhī), and “essential existence” (wujūd dhātī) pertain to the unique ontological status of the quiddity taken strictly in itself.72 In addition to the evidence from the Shifā᾿, which I discussed in detail elsewhere, I wish to add here two relevant passages drawn from the Kitāb al-Ḥudūd and the Kitāb al-Najāt that had escaped my attention until now, and which provide interesting insight into the notions of essential existence and essential oneness. When explaining the relation between definition and quiddity in the Kitāb al-Ḥudūd, Avicenna explicitly glosses māhiyya as wujūd dhātī: Regarding the true definitions (al-ḥudūd al-ḥaqīqa), it is necessary, as we have learned from the art of logic, that they indicate the quiddity 71. See AVICENNA, Ilāhiyyāt, pp. 153. 9–11; 153. 16–154. 2; 155. 14; 155. 17–19; 156. 6–8; 157. 4–5. I will not reformulate here the matrix of arguments and problematics linked to this topic. For a detailed analysis of these issues, see JANOS, Avicenna. 72. Note that two of these expressions (wujūd ilāhī and wujūd dhātī) are already found in the metaphysics of the Arabic Christian philosopher and theologian Yaḥyā b. ῾Adī, whose influence on Avicenna is being increasingly emphasized in the scholarship. In Yaḥyā’s system, these formulae explicitly refer to the special ontological status of essence according to a Neoplatonic scheme; see the seminal study by M. RASHED, “Ibn ῾Adī et Avicenne: sur les types d’existants,” in: V. CELLUPRICA – C. D’ANCONA (eds.), with the collaboration of R. CHIARADONNA, Aristotele e i suoi esegeti neoplatonici: logica e ontologia nelle interpretazioni greche e arabe, Naples 2004, pp. 107–171, especially pp. 142–146; S. MENN – R. WISNOVSKY, “Yaḥyā ibn ῾Adī On the Four Scientific Questions Concerning the Three Kinds of Existence. Editio princeps and Translation,” in: Mélanges de l’Institut dominicain d’études orientales du Caire (MIDEO) 29 (2012), pp. 73–96; JANOS, Avicenna, pp. 359–366, and especially pp. 521–531; and YAḤYĀ B. ῾ADĪ, Trattato sull’ unità (Maqālah fī l-tawḥīd). L’uno, il molteplice e l’unità di Dio, Testo arabo a cura di S. KH. SAMIR sj. Introduzione, traduzione e note a cura di O. L. LIZZINI, Prefazione di G. ENDRESS, Bologna 2020, especially pp. 282, 315, and 333.

TASHKĪK AL-WUJŪD AND THE LAWĀZIM IN METAPHYSICS

127

of a thing, which is the perfection of its essential existence (wa-huwa kamāl wujūdihi l-dhātī), in a way that none of its essential attributes are excluded, be they in actuality or in potentiality.73

Defining quiddity as “the perfection of the essential existence” of a thing is reminiscent of what Avicenna says about māhiyya in Ilāhiyyāt, I, 5 and V, 1. Admittedly, however, the grammar of this passage is ambiguous, because the expression wa-huwa kamāl wujūdihi l-dhātī could be translated equally cogently as “which is the essential perfection of its existence,” with dhātī referring back to kamāl instead of wujūdihi. However, there are two reasons to favor the previous translation. The first is that wujūd dhātī appears to have been somewhat of a consecrated expression by the time Avicenna was writing. For example, it was already employed by Yaḥyā b. ῾Adī in his philosophical treatises.74 In the latter’s works, the connection between the definition, the quiddity, and essential existence is stressed in a manner akin to this passage of the Kitāb al-Ḥudūd. The second reason is the presence of a very similar statement that appears in the Kitāb al-Najāt, but which focuses this time on essential oneness rather than essential existence. In spite of this difference, the similarity in phraseology between the two passages is striking: He [the Necessary Existent] is also one on account of the fact that each thing has a oneness that is proper to it and through which [it obtains] the perfection of its essential reality (wa-wāḥid min jiha anna li-kull shay᾿ waḥda takhuṣṣuhu wa-bihā kamāl ḥaqīqatihi l-dhātiyya).75

The importance of this text for the present purposes should not be underestimated. First, it lends substantial weight to the suggestion that the Kitāb al-Ḥudūd passage is in fact referring to “the perfection of the essential existence,” in a manner akin to the way in which this passage of the Kitāb al-Najāt refers to “the perfection of the essential reality.” Moreover, and by implication, it shows beyond doubt that Avicenna reserved a special sense of waḥda to signify “essential 73. AVICENNA, Kitāb al-Ḥudūd, p. 3, section 5 of the Arabic text, my translation. 74. See, for example, YAḤYĀ B. ῾ADĪ, Trattato sull’ unità (Maqālah fī l-tawḥīd), p. 282, section 472 (= p. 400. 14‒18 in Saḥbān Khalīfāt’s edition). Ibn ῾Adī’s notion of essential existence is, under one aspect, closely linked to the definition (ḥadd) and quiddity (māhiyya) of a thing; see MENN – WISNOVSKY, “Yaḥyā ibn ῾Adī On the Four Scientific Questions,” p. 84. 6–7 for the Arabic text, and p. 94 for the English translation. 75. AVICENNA, Kitāb al-Najāt, ed. M. T. DĀNESHPAJŪH, Tehran 1985, p. 557. 6–7, my translation.

128

DAMIEN JANOS

oneness,” which he contrasted to various senses of “accidental oneness,” and which he associated closely with quiddity (māhiyya) and essential reality (ḥaqīqa). The quotation is taken from a list of various senses of wāḥid and waḥda that Avicenna enumerates and that he applies to God. It is therefore mentioned in a theological context and aimed at explaining how God’s being or reality is essentially one. Nevertheless, what is striking about this particular sense of oneness is that it is clearly meant to extend to Avicenna’s entire ontology. As the shaykh explains, this sense of oneness applies to “each thing” (kull shay᾿ ), inasmuch as it corresponds to the quiddity or essential reality of that thing.76 Accordingly, there is a sense of “the one” that is predicated essentially of each thing, and which contrasts with the various aspects of “the one” said accidentally, or of the one taken as an “external concomitant” of the essence, which Avicenna had discussed earlier in the same work.77 Finally, it reinforces the hypothesis that modulation (tashkīk) comprises instances of both accidental and essential predication and can predicate existence and oneness essentially of certain things, including the immaterial divine existents. Its scope is therefore both categorical and transcendental. Not only do tashkīk al-wujūd and tashkīk al-waḥda make possible a certain description of God’s essential oneness and essential existence; they also enable us to apply these very notions to the pure quiddities of things and therefore possess a relevance for the whole of Avicenna’s ontology. In all of the cases referenced above, it appears that certain notions that are subsumed under tashkīk (existence, oneness, priority, causativeness, and universality) can be applied essentially to the pure quiddity (māhiyya) and true nature or reality (ḥaqīqa) of a thing, as opposed to accidentally or in a manner that pertains to its external attributes and concomitants. This also attests to the relevance of the bi-l-dhāt/ bi-l-῾araḍ distinction that Avicenna includes in tashkīk al-wujūd, but 76. This sense of oneness seems related to what al-Fārābī describes as the oneness of “what is set apart by its quiddity” (῾alā l-munḥāz bi-māhiyyatihi); see ABŪ NAṢR AL-FĀRĀBĪ, Alfarabi’s On One and Unity (Kitāb al-Waḥid wa-l-waḥda), ed. M. MAHDI, Casablanca 1989, section 4, pp. 51–52. 77. See AVICENNA, Kitāb al-Najāt, pp. 514–515, where oneness is described as “a concomitant accident” (min al-a῾rāḍ al-lāzima), “a concomitant attribute” (ṣifa lāzima), and is said to have “an accidental nature” (ṭabī῾a ῾araḍiyya).

TASHKĪK AL-WUJŪD AND THE LAWĀZIM IN METAPHYSICS

129

this time with regard to the māhiyya-lawāzim cleavage, rather than the jawhar-῾araḍ one. It should be noted that the bi-l-dhāt/bi-l-῾araḍ distinction is crucial also to understand how tashkīk applies to God’s existence and oneness, since these notions are predicated of Him essentially and not accidentally, with reference to His māhiyya, and not to the lawāzim deriving from it. In fact, in the two passages dealing expressly with tashkīk where Avicenna introduces the disjunctions bi-l-dhāt/bi-l-῾araḍ and min dhātihi/min ghayrihi, namely Maqūlāt, I, 2 and Jadal, II, 2, it is quite clear that he intends them to apply to essence vs. concomitant. In other words, they refer to a kind of essential predication.78 Before concluding this section, one should note that Avicenna at Ilāhiyyāt, V, 8 hints at the fact that the term māhiyya itself can be regarded as somewhat equivocal (bi-shtirāk al-ism), because it can be ambiguously applied to species, genera, and even to the “particular individual” (al-mufrad al-juz᾿ī).79 Earlier in the same chapter he had explained that, Just as “the existent” and “the one” are among the things common to the categories, albeit by way of priority and posteriority [i.e., as modulated 78. In Jadal, II, 2, it is obvious that the shaykh intends this distinction to apply to essence vs. concomitant, and not to substance vs. accident, because the example he provides immediately afterwards to illustrate this point is the quiddity of triangle vs. its external concomitants. As for the Maqūlāt, I, 2 passage, both TREIGER, “Avicenna’s Notion,” p. 357, and DE HAAN, “The Doctrine,” p. 276, take the al-awlā wa-l-aḥrā notion that is presented there as one of the aspects of tashkīk al-wujūd, as well as the corresponding gloss Avicenna provides for it, i.e., the min dhātihi vs. min ghayrihi distinction, as marking the difference between the Necessary Existent and the contingent beings, and hence between what has existence essentially vs. what has existence accidentally or from another, rather than the difference between substances and accidents. I agree and think this is correct, but one should not lose sight of the fact that Avicenna sometimes applies these sets of distinctions to differentiate between substances (which exist not in a subject and hence have existence “in themselves”) and accidents (which exist in a subject and have existence “accidentally” and “from another” that renders them subsistent; see Ilāhiyyāt, II, 1). 79. AVICENNA, Ilāhiyyāt, V, 8, p. 188. 6‒9. The last notion could be a reference to the “vague individual,” which has been analyzed by D. BLACK, “Avicenna’s ‘Vague Individual’ and its Impact on Medieval Latin Philosophy,” in: R. WISNOVSKY – F. WALLIS – J. C. FUMO – C. FRAENKEL (eds.), Vehicles of Transmission, Translation, and Transformation in Medieval Textual Culture, Turnhout 2011, pp. 259‒292. In this connection, one should remember that, according to Avicenna, the quiddity taken strictly in itself (al-māhiyya min ḥaythu hiya hiya) does not presuppose the logical or ontological status of a thing. For instance, “blackness,” “horseness,” and “animalness” are three pure quiddities, which, conceived of strictly in themselves, do not presuppose accident, species, and genus respectively. In that

130

DAMIEN JANOS

notions], the case is also similar with things that have a quiddity and definition, for this is not on the same level in all things.80

This shows that equivocity and modulation, which are in any case two closely related notions in Avicenna, can, under certain circumstances, be applied to quiddity itself. According to the shaykh, most quiddities, namely, those of the natural species and genera, can be said to be univocal notions, and this for justifiable reasons. One human being is not more “human” or more intensely “human” than another, nor is a lion more “animal” than a tiger. “Human” and “animal” apply to, and are predicated of, various things in exactly the same way, that is, equally and homogenously. With that being said, Avicenna hesitates at times to maintain such a clear-cut interpretation of univocals, equivocals, and modulated terms. On the one hand, some modulated terms are said to be “like” (ka) genus or to “resemble” genus, even though they are not properly said to be genus. This is the case notably of existence (wujūd), which, Avicenna tells us, is like a “quasi-genus” that contains “quasi-species” under it.81 Conversely, certain notions such as quiddity that are typically regarded as univocal can be occasionally subjected to the process of modulation or equivocation, as we saw above. In fact, it would appear that the way in which these notions are to be construed depends to some extent on the context in which they are found. Their categorization is therefore somewhat relative and contextual. At al-Samā῾ al-ṭabī῾ī, II, 2, after having glossed kamāl as a term that is predicated “through modulation” (bi-l-tashkīk), Avicenna explains that, depending on the context, it can also be regarded as equivocal or even univocal, depending sense, quiddity, or rather, the pure quiddity, i.e., what the thing is when taken strictly in itself, seems to be somewhat intrinsically ambiguous or equivocal, as it does not convey information concerning its logical status and its relations to other things. In brief, the concept of the pure quiddity, as a distinct elaboration provided by Avicenna on earlier philosophical thought, does not entail information about the place of an essence within the grid of the categories inherited from Aristotle. To return to the aforementioned example, one can predicate quiddity (māhiyya) of animal, human, and even more specifically of an individual such as ῾Amr – inasmuch as ῾Amr has a specific quiddity that can be taken together with its concomitants and accidents, or, alternatively, inasmuch as there is such a thing as the essence of a “vague individual” for Avicenna – where quiddity is used in a modulated or even equivocal way to refer to these various things. 80. AVICENNA, Ilāhiyyāt, V, 8, p. 186. 5–7, translation slightly revised. 81. See, for example, AVICENNA, Ilāhiyyāt, I, 2, p. 10. 4–5.

TASHKĪK AL-WUJŪD AND THE LAWĀZIM IN METAPHYSICS

131

on the relations established between it and other notions.82 In the final analysis, the question of whether a notion is univocal, modulated, or equivocal is largely determined by its relation to other notions and the philosophical context at hand. Given these considerations, it is not unrealistic to hypothesize that quiddities can, under certain aspects, be regarded as modulated. While they are to be treated as univocal most of the time (i.e., when applied to particular instances that fall under them, such as the universal concept “horse” in its relations to concrete individual horses A and B), when related to other notions, such as priority, universality, and existence, quiddities themselves become potentially subjected to the process of modulation. This remark becomes more compelling when one remembers that Avicenna identifies various loci for the quiddities: in concrete beings, in the human intellect, in the separate intellects (at the very least in the Agent Intellect), and possibly also in God’s mind. When applied specifically to the quiddities, modulation does not treat such notions (priority, oneness, etc.) qua external concomitants, and it does not amount to a kind of accidental predication. Rather, it attributes these notions essentially to the pure quiddity itself.83 It is unnecessary to elaborate on these points. What should be stressed is the fact that, in such instances, and if this hypothesis is correct, core modulated notions and/or core notions of tashkīk al-wujūd – wujūd, taqaddum, waḥda, kulliyya, sababiyya – could be meant by Avicenna to pertain to the pure quiddity itself (“animalness,” “horseness,” etc.), rather than to the external concomitants of essence that become realized with the actualization of the contingent particular or universal thing (shay᾿). There are strong reasons to 82. AVICENNA, al-Samā῾ al-ṭabī῾ī, pp. 132–133, especially p. 133. 5–7 of the Arabic text. 83. Intriguingly, the notions that Avicenna appears to predicate essentially of pure quiddity do not clash with its constituents, nor do they introduce multiplicity in the pure quiddity. Instead, they help to explain how the pure quiddity can be conceived of as a distinct object in the mind. That the pure quiddity can be conceived of as a distinct object in the mind would be accepted, presumably, by all scholars of Avicenna. So, when I think of the pure quiddity “tigerness,” it is not unreasonable to ascribe to pure “tigerness” an essential and intelligible existence (the concept as such exists in my intellect), oneness (it is one concept), priority (it is essentially prior to specific apprehensions of concrete tigers and the act of relating them to a universal), and universality (it is an essentially universal concept that may coincide with the one present in the Agent Intellect, which is distinct, however, from the intentional, complex universal that arises in the mind as a result of relating “tigerness” to individual tigers in the concrete world).

132

DAMIEN JANOS

conclude that the framework of tashkīk in general – and of tashkīk al-wujūd and tashkīk al-waḥda in particular – extends to the pure quiddities themselves and thus allows for a mode of essential and intrinsic predication in addition to a mode of accidental and extrinsic predication. This hypothesis may also account for one of the major ways to interpret his philosophy that developed during the twelfth century. In any case, it is intriguing that this mode of essential predication of tashkīk in Avicenna’s ontology finds an important parallel in his theology. 5.2. Tashkīk and quiddity in Avicenna’s theology Having addressed the issue of tashkīk and the lawāzim in Avicenna’s general ontology, I now wish to discuss wujūd taken specifically as a concomitant of God’s essence, and hence shift the focus to a theological plane. On the assumption that tashkīk al-wujūd does apply to God, how is wujūd to be conceived of, and what is its relation to God’s essence?84 In his study devoted to this topic, Treiger convincingly argued that Avicenna extends the notion of tashkīk al-wujūd 84. The question of whether tashkīk al-wujūd in Avicenna is transcendental and applies to God is a controversial one; see T. MAYER, “Faḫr ad-Dīn ar-Rāzī’s Critique of Ibn Sīnā’s Argument for the Unity of God in the Išārāt and Naṣīr ad-Dīn aṭ-Ṭūsī’s Defence,” in: D. C. REISMAN (ed.) with the assistance of A. H. AL-RAHIM, Beyond and After Avicenna: Proceedings of the First Conference of the Avicenna Study Group (Leiden / Boston 2003), pp. 199‒218; TREIGER, “Avicenna’s Notion”; DE HAAN “The Doctrine”; WISNOVSKY, “On the Emergence,” p. 293; and JANOS, Avicenna, pp. 449–477. For that same issue in al-Fārābī, see P. VALLAT, Farabi et l’école d’Alexandrie: des prémisses de la connaissance à la philosophie politique, Paris 2004, pp. 347–365. In my view, some of the recent scholarship on tashkīk al-wujūd makes a compelling case for its being transcendental. In what follows, I will not engage with this question in detail and will merely recap the main points pertaining to this issue: 1/ there is explicit textual evidence in the Mubāḥathāt that includes God’s existence within tashkīk al-wujūd; 2/ many of the aspects or modes of tashkīk al-wujūd (priority, necessity, worth and relevance), as well as other modulated notions (especially tashkīk al-waḥda), clearly apply to God in Avicenna’s metaphysics; and 3/ Avicenna’s major commentators in the twelfth and thirteenth centuries agreed on the fact that tashkīk al-wujūd extends to God, although they disagreed on other related issues (i.e., whether wujūd is a concomitant of God or identical to His essence; whether the essence-existence distinction is purely mental or extends to the extramental plane; and whether existence generally construed is univocal or equivocal/modulated). In spite of these differences, they all believed that Avicenna intended tashkīk al-wujūd and tashkīk al-waḥda to have a transcendental scope (on this point, see BENEVICH, “The Necessary Existent,” and ZAMBONI, “Is Existence?”).

TASHKĪK AL-WUJŪD AND THE LAWĀZIM IN METAPHYSICS

133

to encompass the divine existence. That is to say, Treiger claimed that this notion has a transcendental application in addition to its categorial application (i.e., contingent substances and accidents). Whether or not Avicenna truly is the first thinker to have come up with this idea in the Arabic tradition, and whether or not some features of this theory had been anticipated by al-Fārābī, are irrelevant for my present concerns.85 What is of interest here is the question of the status of God’s existence in the modern reconstruction of tashkīk al-wujūd that has been proposed. As mentioned at the beginning of this paper, one noticeable trend in the scholarship on Avicenna and on post-Avicennian thinkers who engaged with his philosophy has been to regard God’s existence also as a kind of lāzim. This approach was implicitly adopted by Treiger and more explicitly by Benevich in recent studies.86 The argument, in a nutshell, runs as follows: tashkīk al-wujūd is both transcendental and univocal because it always deals with existence as a concomitant of the essence. It is transcendental, because the principle that wujūd is a lāzim that is external to the essence applies to all cases, including that of God. Here the status of wujūd as a lāzim is taken to its maximal extensionality and applied indiscriminately to all beings. And it is univocal, because if wujūd always relates to essence as a single external attribute, then it is predicated in a similar way of all essences, while potentially also preserving its modulated quality. As scholars have pointed out, this approach seems supported by some textual evidence that can be found in the late treatises al-Ta῾līqāt and al-Mubāḥathāt.87 Following this line of thought leads one to the conclusion that God’s existence is a lāzim of His essence. In my eyes, however, this interpretation is likely to be anachronistic and in part indebted to al-Rāzī’s reading of Avicenna. As Benevich lucidly showed in his article, tashkīk al-wujūd, the status of wujūd as a concomitant, and the univocity of being all go together in al-Rāzī’s mind. But in Avicenna, as we saw, the correlation between tashkīk al-wujūd and wujūd taken as a lāzim of the essence is only relative 85. See TREIGER’s (“Avicenna’s Notion,” pp. 351–352) response to Philippe Vallat regarding this point. 86. TREIGER, “Avicenna’s Notion,” pp. 362–363; BENEVICH, “The Necessary Existent,” pp. 150–151. 87. One such passage is AVICENNA, Mubāḥathāt, p. 169, section 479.

134

DAMIEN JANOS

and partial, not absolute. It can only be taken to indicate that wujūd is a lāzim in many or even most instances, but not all. Moreover, and by the same token, one cannot infer the univocal nature of wujūd from its being a lāzim. According to Avicenna, the fact that wujūd is (in most cases) a lāzim points not to its being univocal, but rather to its being mushakkik, which is not exactly the same thing. Moreover, the question remains – and has not been decisively answered – as to whether mushakkik in Avicennian parlance has a meaning closer to univocity or equivocity. Although one observes a tendency in the modern scholarship to conceive of tashkīk al-wujūd as expressing a kind of univocity of being, this approach poses some serious textual and conceptual problems, which have been evoked elsewhere.88 For one thing, Avicenna’s identification of essence and existence in God poses a serious obstacle for the univocal reading of tashkīk, for it undermines the presumed connection between tashkīk and wujūd taken as a lāzim. And it disrupts the hypothesis according to which existence is univocal because it is always to be regarded as an external concomitant of essence. Al-Rāzī, who was an acute reader and critique of the shaykh, clearly perceived this problem and proceeded to refute Avicenna’s thesis of the essence-existence identity in God (but not his theory of tashkīk). Hence, at best, one can say that the connection between tashkīk al-wujūd and the lawāzim in Avicenna is only partial, given that there is at least one case of essential existence (the difficult case of the pure quiddities notwithstanding). As for the evidence found in al-Ta῾līqāt and al-Mubāḥathāt that has been taken by some to suggest that wujūd is a lāzim of God, the relevant passages are rather difficult to interpret properly, given the absence of an adequate context (not provided in these works), and so in my view they remain highly ambiguous. In fact, the references to wujūd in these passages could also be taken to refer to the existence of God’s effects qua lawāzim of His essence, and hence to the wujūd caused by God, rather than to God’s wujūd itself. In other words, given that Avicenna frequently uses the term lawāzim to refer to the effects caused by God, the more straightforward interpretation in these rare instances where wujūd seems to be described as a lāzim would be to connote it as an effect of God’s causality. What is more, the claim 88. JANOS, “Avicenna on Equivocity and Modulation.”

TASHKĪK AL-WUJŪD AND THE LAWĀZIM IN METAPHYSICS

135

that Avicenna aligned himself with a kalām view according to which God’s existence is distinct from His essence is ultimately untenable in light of the overwhelming evidence contained in his core works concerning the essence-existence identity thesis in God. Hence, the urgency felt by some scholars (who in this regard follow al-Rāzī) to ground tashkīk al-wujūd fully and strictly in the notion of the concomitants (lawāzim) – including in its transcendental application – appears uncalled for. It is supported neither by the textual evidence related to tashkīk, nor by what the shaykh himself says in his canonical works about God’s wujūd and the essence-existence identity thesis. In view of this, another approach has been suggested, which is to reject the hypothesis of the transcendental scope of tashkīk al-wujūd.89 The main obstacle with that recommendation lies in the considerable evidence and philosophical reasons that point to the transcendental nature of tashkīk in Avicenna. Hence, the relevant question that emerges at this juncture is the following: is there a way to reconcile the transcendental scope of tashkīk al-wujūd and the essence-existence identity thesis? As it turns out, the only viable option remaining to alleviate this conundrum is to postulate that tashkīk allows for these notions to be predicated of God essentially and with regard to His very quiddity. In this case, God’s existence is not an added and external concomitant of His essence, but is rather to be identified with His essence taken in itself. Accordingly, Avicenna tells us that God’s quiddity is existence, or rather that He has existence essentially. Exactly the same reasoning can be applied to God’s oneness. It is not an external, accidental, and derivative concomitant of God’s essence, but represents a kind of essential oneness, as we saw above with regard to the passage from the Kitāb al-Najāt. As has been shown in the scholarship, Bahmanyār, one of Avicenna’s closest disciples, seems to have interpreted the theological scope of tashkīk al-wujūd precisely in this fashion.90 Recall that existence, necessity, priority, and oneness are routinely ascribed to God in Avicenna’s philosophy. Now, they also happen to be central features of tashkīk and tashkīk al-wujūd. Hence, whenever Avicenna attributes existence, necessity, priority, and oneness to God, he is presumably doing so within the framework established by his 89. This approach is recommended by WISNOVSKY, “On the Emergence,” p. 293. 90. JANOS, Avicenna, pp. 455–456; ZAMBONI, “Is Existence?” especially pp. 143–145.

136

DAMIEN JANOS

theory of modulation. Yet, in these instances, it is quite clear that the shaykh is not predicating these notions accidentally of God and as external lawāzim of His essence. Rather, he can only have essential existence, essential necessity, essential priority, and essential oneness in mind. After all, he did not follow earlier thinkers, such as his Christian philosophical counterpart Yaḥyā b. ῾Adī, nor did he anticipate later Muslim theologians such as al-Ghazālī and al-Rāzī, in allowing a certain intelligible multiplicity in the divine essence, either by way of a plurality of divine concomitants (lawāzim) or of a plurality of attributes (ṣifāt). For example, according to Yaḥyā, God is both one and multiple, a move that enables him to maintain the paramount doctrine of the Trinity, as well as to obviate the problem of God’s knowledge of a multiplicity of objects. And according to both al-Ghazālī and al-Rāzī, God’s essence cannot be said to be completely simple (basīṭa). Avicenna, in contrast, put forth the doctrine of God’s absolute oneness and simplicity, which means that any attribute or quality predicated of God will perforce be identical to His essence. In brief, as in the case of pure quiddity, any act of predication in this case will be bi-l-dhāt, essential and simple, which, it should be recalled, is also a feature of tashkīk. For these reasons, one may surmise that the connection between tashkīk and the lawāzim that Avicenna is typically inclined to emphasize co-exists with a kind of essential predication of modulation that is applied to the pure quiddities, including God’s special quiddity or essence. Transposing this argument to theology suggests a transcendental use of tashkīk that obeys the rule of essential predication and mirrors its application to the pure quiddities in Avicenna’s ontology. In both cases, this implies a rejection of the wujūdlāzim connection. Here one may pause and ask what may have prompted Avicenna to defend such a position. One consideration could be the need to elaborate an ontology that can accommodate objects as diverse as God, the pure quiddities, caused and contingent beings, mental existents, and other types of beings, while at the same time skirting the pitfalls associated with strict univocity and strict equivocity. Avicenna’s brilliance here was to devise a modulated ontological framework that salvaged the unity of metaphysics as a science, but at the same time was refined and flexible enough to address such diverse entities in a coherent and integrated way.

TASHKĪK AL-WUJŪD AND THE LAWĀZIM IN METAPHYSICS

137

One angle through which to further tease out this idea of essential modulation focuses on the notions of existence “through or by virtue of one’s self” (or “essential existence”) (bi-dhātihi), as opposed to “accidental existence” (bi-l-῾araḍ). As we saw earlier, Avicenna explicitly makes this distinction a criterion of tashkīk al-wujūd. It should be noted that this aspect of modulation is entirely consistent with the shaykh’s general description of the metaphysical investigation, which studies things that occur both accidentally and essentially, or, as Avicenna himself puts it in Ilāhiyyāt, I, 4, the state of the necessary and the state of the possible, as well as “the state of what is in essence and what is accidental” (ḥāl alladhī bi-l-dhāt wa-alladhī bi-l-῾araḍ). Although he did not elaborate his views, Marmura also took the latter statement to apply to tashkīk in his commentary on the Ilāhiyyāt. He writes: “it [what is said bi-l-tashkīk] is a word that has the same meaning but differs in priority and posteriority … [and] also in its essential and accidental usage.”91 Now, at first glance, this distinction between bi-l-dhāt and bi-l῾araḍ seems primarily aimed at distinguishing the existence of substances and accidents, and Avicenna indeed often mentions these notions together, as in Ilāhiyyāt, II, 1. Whereas substances have existence in themselves and do not need a subject in order to exist, accidents in contrast need a subject in order to exist, are made subsistent by substance, and have a purely accidental type of existence (such as “musical” in a certain human being, or “like the existence of Zayd as white” (mithl wujūd Zayd abyaḍ), which is the example Avicenna gives at the beginning of II, 1). However, when applied to God, it is clear that the formulae bi-l-dhāt and bi-dhātihi assume the narrower meaning of “essentially” or “in His essence,” given that, technically, God is not a substance (jawhar) for Avicenna, but rather a pure, abstract, and perfectly unitary essence. In this context, wujūd bi-dhātihi means something like wujūd dhātī or wujūd ilāhī, expressions which in turn correlate with Avicenna’s essence-existence identity thesis in God, i.e., that God’s wujūd is His māhiyya. The same can be said about divine oneness: it is waḥda bi-l-dhāt or waḥda dhātiyya. After 91. MARMURA, The Metaphysics, p. 391, in his comments on section III, 2. The view that tashkīk deals with the essential and the accidental is also acknowledged in Zamboni’s paper, which was cited previously.

138

DAMIEN JANOS

all, what would the essential predication (bi-l-dhāt) of oneness and existence pertain to, if not to a being that possesses these things essentially, as opposed to accidentally and externally? Hence, although these formulae apply to substances and accidents (and this may very well be their primary application in Avicenna’s ontology), it is also quite clear that the shaykh wants to extend these notions to God and to the pure quiddities in order to distinguish the essence and what is essential from the concomitant and what is accidental and external. That Avicenna intends tashkīk partially in this transcendental and essential way seems further corroborated by a statement that appears at Ilāhiyyāt, I, 6: “There are specific properties (khawāṣṣ) that belong individually each to the Necessary Existent and the possible existent.”92 That claim is quite committal, for there are only a limited number of properties that God can share with the contingent beings. In light of the foregoing analysis, it seems that Avicenna primarily has things such as existence, priority, necessity, and unity in mind here (i.e., the modulated notions), and that by “properties” he is referring to the lawāzim, lawāḥiq or ῾awāriḍ of wujūd that he elsewhere mentions in Book I of the Ilāhiyyāt. But while all of these notions fall under tashkīk and/or tashkīk al-wujūd, in the case of the First, these notions definitely do not apply qua concomitants or accidents; rather, they are predicated essentially of It. Ultimately, the observation that the bi-dhātihi and bi-l-῾araḍ distinction is subsumed under tashkīk al-wujūd seems highly relevant to understand Avicenna’s theology and ontology. The theory of tashkīk encompasses these notions as they apply both to contingent beings and God: in the first case, they apply accidentally and focus on the external concomitants; in the latter case, they apply essentially and focus on the divine quiddity. 6. Some implications for the post-Avicennian tradition: Bahmanyār, al-Ṭūsī, al-Rāzī, and al-Suhrawardī The two issues of whether wujūd is to be treated as a lāzim of the essence (both in the case of contingent beings and in the case of God) and of whether tashkīk applies to essences themselves in addition to the concomitants of essence, are foundational when one tries 92. AVICENNA, Ilāhiyyāt, I, 6, p. 29, Marmura’s translation.

TASHKĪK AL-WUJŪD AND THE LAWĀZIM IN METAPHYSICS

139

to understand the post-Avicennian metaphysical tradition. With regard to the first issue, there are various debates that flared regarding the extramental status of wujūd, whether wujūd is a purely mental consideration or possesses extramental reality, and, if so, what its relation to essence is, i.e., whether it is a concomitant or attribute external and superadded to essence, or whether it is to be identified with the essence of a thing itself. With regard to the latter issue, the question was how to interpret Avicenna’s theory of tashkīk al-wujūd and the various modulated notions it subsumes, as well as to identify the various objects to which that theory could be applied. In a theological context, these two issues merged, and the aim was to understand exactly what tashkīk modulates with regard to God: wujūd and waḥda as external lawāzim of His essence, or His very essence. For many of these thinkers, who were deeply involved in commenting on and interpreting the master’s works, the challenge was to determine whether God’s existence and oneness are essential to Him and identical with His essence or quiddity, or whether they relate to God qua external attributes or concomitants and remain fundamentally distinct from His essence. Put briefly, the philosophical issue at stake was nothing less than the proper interpretation of Avicenna’s māhiyya-lawāzim paradigm. When one connects these issues with the question of the meaning and scope of tashkīk and tashkīk al-wujūd, one is confronted with a cluster of problematics that defines much of the theological and metaphysical discussion that unfolded during the twelfth and thirteenth centuries. In this regard, it is remarkable that most of Avicenna’s commentators accepted the transcendental application of tashkīk and tashkīk al-wujūd. The point of contention was not whether these notions apply to God, but rather how and in what regard they apply to God, i.e., whether modulated notions such as existence, oneness, priority, and necessity are essential to God or external concomitants of His essence. Since very insightful studies on this topic have appeared recently, I will limit myself in what follows to delineating some important features of the reception of Avicenna’s theory of tashkīk in the works of some of his main commentators.93 My contention is that the interpretation of 93. See in particular WISNOVSKY, “On the Emergence”; BENEVICH, “The Necessary Existent”; and ZAMBONI, “Is Existence?” In the remainder of this study, I outline some

140

DAMIEN JANOS

Avicenna’s metaphysics that was articulated earlier in the present study allows one to better grasp three major developments that occur after the master’s death, all of which gravitate toward a particular interpretation of tashkīk and its relation to the lawāzim. The first trend materialized in the works of Bahmanyār and Naṣīr al-Dīn al-Ṭūsī. As scholars have shown, these two thinkers broadly endorsed Avicenna’s theory of tashkīk al-wujūd, which they applied both to their ontology and theology.94 Although they considered wujūd generally as a lāzim of the essence of contingent beings, and tashkīk as modulating wujūd in this capacity, in the case of God, it is quite clear that they held to a view of essential existence and to Avicenna’s essence-existence identity thesis. In his Sharḥ al-Ishārāt, al-Ṭūsī directly criticizes al-Rāzī for treating God’s existence on a par with that of contingent beings and for describing it as a concomitant distinct from His essence. According to al-Ṭūsī, al-Rāzī was profoundly “misled” or “confused” about this matter (iḍtaraba iḍtirāban): Since he [al-Rāzī] observed that the existence of contingent beings is something accidental to their quiddities (amran ῾āriḍan li-māhiyyatihā), and since he had ruled that the existence of the Necessary [Existent in Itself] is equal to the existence of the contingent [beings], he concluded that the existence of the Necessary is also accidental to Its quiddity.95

And al-Ṭūsī to argue that “the origin of this mistaken approach is his ignorance regarding predication by modulation.”96 This passage adequately illustrates the divergent interpretations of tashkīk and its application to theology that were put forth by some of Avicenna’s main commentators. The implications of al-Ṭūsī’s position are that tashkīk allows for both accidental and essential predication and that, while transcendental, it is applied differently to God and contingent beings. Yet, both tashkīk al-wujūd and tashkīk al-waḥda encompass general ontology as well as theology. Overall, Bahmanyār’s and al-Ṭūsī’s general considerations regarding the reception of Avicenna’s theory of tashkīk that rely in part on the discussions found in these papers. 94. In addition to the articles mentioned above, see RIZVI, Mullā Ṣadrā, pp. 42–44, 51; and JANOS, Avicenna, pp. 455–456, for the views of Bahmanyār and al-Ṭūsī. 95. NAṢĪR AL-DĪN AL-ṬŪSĪ, Sharḥ al-Ishārāt, in al-Ishārāt wa-l-tanbīhāt, ed. S. DUNYĀ, Cairo 1957–60, 4 vols., namaṭ 4, 17, vol. 3, p. 30. 15–22, my translation. This passage is discussed also in RIZVI, Mullā Ṣadrā, pp. 43–44. 96. AL-ṬŪSĪ, Sharḥ al-Ishārāt, namaṭ 4, 17, vol. 3, p. 31. 2, my translation.

TASHKĪK AL-WUJŪD AND THE LAWĀZIM IN METAPHYSICS

141

positions are therefore rather close to Avicenna’s views according to the reconstruction I proposed earlier. However, unlike in Avicenna, there is no clear indication in their philosophical systems that they would also apply tashkīk to the pure quiddities. At least to my knowledge, there is nothing that approximates Book V, 1 of the Ilāhiyyāt in their works. The second trend is grounded in the works of Fakhr al-Dīn al-Rāzī. Although al-Rāzī has little to say about tashkīk per se, it is clear from what he does say about it in the Sharḥ ῾Uyūn al-ḥikma that he does not oppose it on philosophical grounds. In fact, he regards it as being reconcilable with the strong univocity of existence he advocates in his works.97 Moreover, al-Rāzī elaborates on Avicenna and very strictly correlates univocal modulation with the state of wujūd as a concomitant. Yet, in his case, he extends this tashkīk-lāzim correlation to the entirety of his ontology and theology, and, hence, to God as well. As Benevich has shown, Fakhr al-Dīn al-Rāzī outlines three different positions regarding the issue of the univocity vs. equivocity of wujūd and includes Avicenna, together with his theory of tashkīk, in the clan of those who uphold a univocal interpretation of wujūd.98 But al-Rāzī’s portrayal of these groups seems misleading when one attempts to grasp the nuances of Avicenna’s reasoning on the matter. For there is no strict relation in Avicenna between tashkīk al-wujūd and the univocity promoted by al-Rāzī; nor, as we saw, is there an absolute correlation between tashkīk al-wujūd and existence taken as an external concomitant of essence, even though this would adequately describe the majority of cases to which that theory applies. Moreover, Fakhr al-Dīn al-Rāzī seems especially intent on connecting univocity of being with the essence-existence distinction, and equivocity of being with the essence-existence identity. Since, according to al-Rāzī, Avicenna defended two views that are irreconcilable – the univocity of wujūd and the essence-existence identity in God – he must be mistaken on some level. Overall, al-Rāzī seems eager to provide 97. See BENEVICH, “The Necessary Existent,” pp. 134–135. 98. The three positions outlined by al-Razī are the following: (1) existence is equivocal and identical to God’s essence; (2) existence is univocal and distinct from God’s essence; and (3) existence is univocal and identical to God’s essence. Al-Rāzī defends position (2), ascribes position (1) to al-Ash῾arī, and position (3) to Avicenna; see BENEVICH, “The Necessary Existent,” p. 125.

142

DAMIEN JANOS

a schematic classification that fits his outlook, but which inevitably proffers a simplified account of Avicenna’s metaphysics. Now, given that al-Rāzī endorses the view that wujūd is univocal, and that he regards tashkīk al-wujūd as being potentially reconcilable with univocity, what he proceeds to criticize in Avicenna is not his theory of tashkīk al-wujūd, but rather his view that, in God, essence and existence are one and identical. Although this fact seems firmly established, what is interesting here is al-Rāzī’s putative reason for doing so and the rationale behind his approach. The most convincing explanation is that al-Rāzī, in embracing Avicenna’s māhiyya-lawāzim model, held to a strict correlation between tashkīk and the lawāzim, brought it to its full logical potential, and allowed for no exceptions to this rule. This implies that, even in the case of God, existence must be regarded as external to His essence. It would seem that it is precisely this hard correlation between tashkīk and the lawāzim that – in al-Rāzī’s eyes – justifies the very univocity of existence: the latter is a single nature (ṭabī῾a) that is always to be construed as a distinct, posterior, and external concomitant of the essence of a thing, and it is always predicated as such. So al-Rāzī faults Avicenna for departing from this rule and for introducing exceptions to the application of his own māhiyya-lawāzim paradigm. In fact, for the Ash῾arite theologians and philosophers who flourished during the twelfth century – al-Ghazālī, al-Mas῾ūdī, and especially al-Rāzī –, Avicenna’s own theory of tashkīk al-wujūd and his māhiyya-lawāzim paradigm demand that we regard God’s existence as an external and posterior concomitant predicated of His essence. In any case, it is clear that al-Rāzī tried to co-opt Avicenna’s theory of tashkīk al-wujūd and his māhiyya-lawāzim model to suit his own agenda, while at the same time downplaying its complexity and ambiguity in Avicenna’s thought. In contrast, Bahmanyār and al-Ṭūsī both strove to maintain a nuanced interpretation of tashkīk and of its relation to the māhiyya-lawāzim distinction, which led them to defend the relative equivocity of existence, as well as God’s essential existence. Overall, their account of tashkīk is more discerning and also seems more faithful to Avicenna. It is understandable, then, that al-Ṭūsī responded critically to al-Rāzī’s characterization of Avicenna’s metaphysics, rectifying what he perceived to be errors and simplifications. In spite of these profound differences, all of these thinkers appear to have conceptualized these issues in an interconnected and partially

TASHKĪK AL-WUJŪD AND THE LAWĀZIM IN METAPHYSICS

143

overlapping way. From the point of view of Avicenna’s heirs and commentators, the issue of how to interpret tashkīk al-wujūd obviously overlapped with the issue of determining whether wujūd is univocal or equivocal or something other than that; this in turn called for an examination of Avicenna’s māhiyya-lawāzim paradigm and the question of whether tashkīk al-wujūd pertains only and always to wujūd taken as a lāzim of the essence; finally, the latter issue raised the question of how all of this applies to God’s existence, that is, whether God has essential existence or whether His existence should also be regarded as an external concomitant of His essence, with ensuing repercussions for the theses of univocity, equivocity, and modulation. For Bahmanyār, al-Rāzī, and al-Ṭūsī, the question of whether tashkīk and tashkīk al-wujūd apply to the very quiddity of a thing – and, hence, can designate a kind of essential existence – or whether these are limited to the external or extrinsic features of an essence was framed in terms of the God vs. contingent beings dichotomy. These thinkers did not consider applying tashkīk either to pure essences whatsoever (al-Rāzī) or to pure essences other than God’s essence (Bahmanyār and al-Ṭūsī). This brings us to the interesting and rather peculiar interpretation of Avicenna’s metaphysics proposed by al-Suhrawardī. Al-Suhrawardī also picks up Avicenna’s theory of tashkīk, but unlike the other philosophers, he believes that it can be used to modulate essences themselves. Granted, in his philosophy, the foundational principle is light (nūr), rather than the pure quiddities postulated by Avicenna. But the hierarchy of lights al-Suhrawardī posits is clearly intended as a counterpart to the Avicennian essences or quiddities, and he regards existence as a purely mental construct or consideration (i῾tibār) that has no real status in the extramental world. According to him, only hierarchies of light constitute the reality of the world. Al-Suhrawardī has therefore adopted and elaborated on Avicenna’s fundamental intuition that essence and existence are distinct, and he has offered his own interpretation of their relationship in the mind and in the concrete world. Although al-Suhrawardī rarely mentions the term tashkīk in his works, he clearly believes that modulation can be applied to the lights or pure essences that constitute the beings or real entities of the world. He makes ample use of some notions that Avicenna had associated with tashkīk, especially perfection (kamāl) and intensity and weakness (al-shidda wa-l-ḍu῾f ). For example, in Kitāb al-Mashāri῾

144

DAMIEN JANOS

wa-l-muṭāraḥāt, he describes the intensity of the lights in terms of degrees of perfection, and in the Ḥikmat al-ishrāq he explains that they are differentiated “through perfection and deficiency” (bi-l-kamāl wa-l-naqṣ).99 As Roxanne Marcotte put it, “what distinguishes quiddities is the degree of their luminous intensity within the ontological hierarchies or ‘grades’ (more or less) of light, from the noblest of light, the Light of Lights (God) to the absence of it (matter).”100 Now, as we saw earlier, Avicenna occasionally includes the criterion of al-shidda wa-l-ḍu῾f in his general framework of tashkīk al-wujūd, even though it does not seem to play an important role in his conception of it. Yet, it would be an exaggeration to claim, as is often done in the secondary literature, that the shaykh excluded it entirely from modulation. He seems rather to have limited “strength and weakness” mostly to the lawāzim of essence, and more specifically to accidents, such as blackness and whiteness, although he also occasionally mentions it with regard to substances and accidents more generally. Whatever the case may be, al-Suhrawardī appears to have adopted this aspect of tashkīk al-wujūd, amplified it, and linked it to Avicenna’s other claim that essences in themselves can be modulated (albeit using different aspects of tashkīk). Hence, there are fundamental ideas that al-Suhrawardī could have picked up from Avicenna’s works: that tashkīk can modulate the pure quiddities themselves; that the notion of bi-dhātihi/bi-l-dhāt, which Avicenna explicitly subsumes under tashkīk al-wujūd and applies in one sense to essence, can be taken to apply directly to the lights; and that, through tashkīk, some things are worthier and more deserving, prior, and more perfect than others. In addition, recall that Avicenna explicitly calls “perfection” (kamāl) a 99. AL-SUHRAWARDĪ, Majmū῾a fī l-ḥikma l-ilāhiyya, ed. H. CORBIN, Istanbul 1945, p. 465; Œuvres philosophiques et mystiques, Tome II, ed. H. CORBIN, Tehran 1993, p. 119. Recall that the term “differentiation” (ikhtilāf), which al-Suhrawardī uses in these two passages, is a keyword that appears in virtually all of Avicenna’s discussions of tashkīk al-wujūd. Although RIZVI, Mullā Ṣadrā, p. 44, gives a reference to the Ḥikmat al-ishrāq to support his claim that “Suhrawardī himself argued that the hierarchy of light is mushakkak,” I was not able to find the term mushakkik or mushakkak in the passage of the Ḥikmat al-ishrāq alluded to in his article. However, Rizvi’s point is well taken: al-Suhrawardī establishes his hierarchy of lights on the basis of a gradated or modulated scheme grounded in the notions of perfection and deficiency, as well as strength and weakness. 100. See R. Marcotte, “Suhrawardi,” The Stanford Encyclopedia of Philosophy (Summer 2019 Edition), Edward N. Zalta (ed.), URL = . ID.,

TASHKĪK AL-WUJŪD AND THE LAWĀZIM IN METAPHYSICS

145

modulated term in al-Samā῾ al-ṭabī῾ī.101 There seems therefore to have been crucial material in the Avicennian works that could have informed al-Suhrawardī’s theory of the modulation of lights or essences. This intriguing point, however, requires additional research. Over and beyond this matrix of issues and the disagreements that raged between these thinkers, two things appear clearly enough in the post-Avicennian period. First, all of these thinkers believed that Avicenna’s theory of tashkīk applies to God in addition to the contingent beings – i.e., that it amounts to a transcendental theory, as Treiger put it. With regard to tashkīk al-wujūd specifically, the point of contention for them was not whether it includes God’s existence or not, but whether it forms part of a univocal or equivocal framework, and whether it focuses on wujūd as an external concomitant of His essence or as His special essential being. The second point is that all of these thinkers proposed new and divergent formulations of Avicenna’s māhiyya-lawāzim paradigm in connection with wujūd and tashkīk. Some, like al-Rāzī, insisted on limiting existence to the periphery of the lawāzim, which explains why existence according to him is always external, posterior, and added to the essence, even in the case of God. Others, such as Bahmanyār and al-Ṭūsī, and in a strange way al-Suhrawardī and the ishrāqīs, reserved a special sense of wujūd that applies to quiddity or essence taken in itself. In the case of the Bahmanyār / al-Ṭūsī affiliation, this sense of essential existence is reserved for God alone; in the case of al-Suhrawardī and the ishrāqīs, who take this idea much further, light or essence is the only real and foundational principle in the exterior world, and, as such, becomes the main object of modulation. For al-Suhrawardī, the idea that existence qua concomitant is something real and distinct in the concrete world is fully abandoned, and wujūd as such is relegated to the status of a concept or a mere mental construct. Hence, the line dividing what Benevich called the “conceptualists” from the “realists”102 in the postclassical period hinged to some degree on the way in which tashkīk was thought to relate to Avicenna’s māhiyya-lawāzim paradigm. 101. AVICENNA, al-Samā῾ al-ṭabī῾ī, II, 2, p. 132 of the Arabic text. 102. F. BENEVICH, “The Essence-Existence Distinction: Four Elements of the PostAvicennian Metaphysical Dispute (11–13th Centuries),” Oriens 45 (2017), pp. 203–258, especially pp. 208–211.

146

DAMIEN JANOS

7. Conclusion The analysis has shown that tashkīk and tashkīk al-wujūd lie at the very heart of Avicenna’s conception of the metaphysical science. Tashkīk focuses primarily on two notions, two paradigmatic cases – wujūd and waḥda – that form the core of the subject matter of metaphysics and articulate its two main aspects, ontology and henology. The anatomy of tashkīk al-wujūd (i.e., the analysis of the main notions subsumed under it that “perform” modulation) reveals that these notions as well are modulated, chief among which are priority and posteriority. What is more, the various notions subsumed under tashkīk al-wujūd overlap considerably both with the attributes and accidents of wujūd and mawjūd as described in Ilāhiyyāt, I, 1–5 and with the concomitants of quiddity as described in Ilāhiyyāt, V, 1. This can be explained in part by the co-extensionality and co-implicativeness of shay᾿ and mawjūd in Avicenna’s philosophy. Furthermore, there is limited but strong evidence to suggest that Avicenna extended tashkīk to the pure quiddities themselves and allowed for an essential application of certain notions (existence, oneness, priority, causativeness, universality) to the quiddities taken strictly in themselves (including, of course, God’s essence, which represents the most eminent case). On the one hand, the previous analysis leads to a reconsideration of the subject matter of metaphysics according to Avicenna. It is often stated in the secondary literature in a somewhat cliché manner that Avicennian metaphysics consists in the study of being qua being or of “the existent inasmuch as it is existent.” Yet, as Amos Bertolacci has convincingly shown, Avicenna’s metaphysical project amounts to both an ontology and a henology: it consists in the study of existence and its concomitants as well as of oneness and its concomitants.103 In this connection, it is worth pointing out that the two core notions underlying these fields, namely, existence (wujūd) and oneness (waḥda), are modulated notions and are themselves also described as concomitants (lawāzim) of the pure quiddity. Given this fact, an alternative, and perhaps more precise, way of describing Avicenna’s metaphysical project would be to say that it aims to study the relation between quiddity and its concomitants, as well as the concomitants 103. See BERTOLACCI, The Reception, pp. 162–169, 176–180.

TASHKĪK AL-WUJŪD AND THE LAWĀZIM IN METAPHYSICS

147

of the concomitants of quiddity, so that the māhiyya-lawāzim distinction (which subsumes both existence and oneness) ultimately emerges as the central concept of Avicenna’s metaphysics. On the other hand, the application of modulation to the pure quiddities considerably complexifies the assumption that tashkīk applies solely to the concomitants of a thing (which, however, remains true for the most part in Avicenna’s metaphysics). It also suggests that wujūd should not be construed as a univocal notion in Avicenna (as al-Rāzī and some modern scholars advocate), but rather as a mildly equivocal notion (as Bahmanyār and al-Ṭūsī held). Additionally, the tashkīklāzim correlation represents only one aspect of Avicenna’s theory of modulation, and was not in any case intended as an absolute rule, although it was subsequently picked up and erected into an axiom by al-Rāzī. When it comes to theology and the First, tashkīk presupposes a transcendental and essential mode of predication of the notions associated with it. It is this theological exception, coupled with the tashkīklāzim connection as applied to contingent beings, that characterizes Bahmanyār’s and al-Ṭūsī’s reception of Avicenna’s metaphysics. Finally, the tantalizing Avicennian idea that there is such a thing as the modulation of pure quiddity announces al-Suhrawardī’s later metaphysical elaborations, which will emphasize the modulation of essences or lights through the criteria of kamāl and al-shidda wa-l-ḍu῾f. Although Avicenna explicitly claims that these notions are modulated – and in the case of al-shidda wa-l-ḍu῾f, that it even forms part of the conceptual framework of tashkīk al-wujūd – he seriously limits their implementation. It is therefore fascinating to realize that the main ontological and theological systems that were articulated during the twelfth and thirteenth centuries, even though they are drastically opposed to one another, all stem from Avicenna’s theory of tashkīk and his māhiyyalawāzim paradigm and may be regarded as different iterations of these seminal theories. Damien JANOS Université de Montréal, Canada Humboldt Fellow, LMU, Munich [email protected]

FREEDOM AND RESPONSIBILITY IN AVICENNA Jari KAUKUA Abstract It is still a matter of some debate whether Avicenna grounds moral responsibility in a robust notion of free will. In this contribution, I will first delve into Avicenna’s theory of voluntary agency, arguing that he holds voluntary agency to be responsive to reasons but also thoroughly determined by the agent’s beliefs concerning the relevant goals, instruments, and qualifying circumstances. Since these beliefs in turn are caused, it seems that there is little room for a causally undetermined will in Avicenna’s theory. I will conclude by considering the question of whether Avicenna is some kind of compatibilist concerning the relation between determinism and responsibility.

Many scholars today hold that Avicenna was a robust determinist, some even come close to interpreting him as a full-blown necessitarianist, who thinks that all that exists is necessarily as it is, and that nothing could have been otherwise than it is.1 This verdict is a logical consequence of Avicenna’s metaphysical commitments. Most importantly, he endorses the principle of sufficient reason, which is commonly taken to entail necessitarianism.2 Moreover, for theological reasons, Avicenna is compelled to hold that the actual world is the only 1. See, for example, A. GOICHON, La distinction de l’essence et de l’existence d’après Ibn Sīnā (Avicenne), Paris 1937, pp. 162-163; M. E. MARMURA, “The Metaphysics of Efficient Causality in Avicenna (Ibn Sīnā),” in: M. E. MARMURA (ed.), Islamic Theology and Philosophy: Studies in Honor of George F. Hourani, Albany 1984, pp. 172-187; J. R. MICHOT, La destinée de l’homme selon Avicenne, Leuven 1986, pp. 61-64; R. M. FRANK, Creation and the Cosmic System: Al-Ghazālī and Avicenna, Heidelberg 1992, pp. 23-24; C. BELO, Chance and Determinism in Avicenna and Averroes, Leiden 2007, pp. 91-120, 225-232; and M. DE CILLIS, Free Will and Predestination in Islamic Thought, London/New York 2014, pp. 36-40. 2. For PSR in Avicenna, see K. RICHARDSON, “Avicenna and the Principle of Sufficient Reason,” in: The Review of Metaphysics 67 (2014), pp. 743-768. On PSR and necessitarianism, see Y. Y. MELAMED – M. LIN, “Principle of Sufficient Reason,” in: E. N. ZALTA (ed.), Stanford Encyclopedia of Philosophy, Spring 2020 Edition, URL = https://plato. stanford.edu/archives/spr2020/entries/sufficient-reason/.

150

JARI KAUKUA

possible one: since there is only one creator whose goodness necessitates Him to create the best possible world, there is no room for alternative states of affairs. Some scholars have challenged the determinist interpretation on ethical grounds. In this regard, Jules Janssens’ stance is arguably the most explicit: since Avicenna is clearly committed to the grounding idea of meaningful ethics, namely that human beings are morally responsible for their own acts and that this requires those acts being free of causal restraint in some sufficiently robust sense, we must not interpret his metaphysics in a way that is incompatible with it.3 Along sympathetic lines, a recent paper by Anthony Ruffus and Jon McGinnis suggests that Avicenna’s theory of the will leaves room for freedom in at least some cases.4 Others have sought refuge in the fact that Avicenna’s modal metaphysics seems designed to clear room for genuine possibility or contingency (imkān) in the metaphysical makeup of the world,5 or in the indeterminacy left by Avicenna’s account of material and efficient causation.6 Whichever side we choose in the debate, we should recognise that the underlying dilemma of causal determination and responsibility is one Avicenna himself would have recognised. By the turn of the first millennium CE, the Mu῾tazilite and the Ash῾arite schools of Islamic theology had already been debating the foundation of human responsibility for almost a century. This debate hinged on the theological question of how to reconcile between God’s attributes of justice and omnipotence: if God is just, then He should distribute rewards and punishments based on acts His subjects have performed freely and on their own accord; yet if any of God’s creatures are free to act by their own choosing, then God has only limited power over them, 3. J. JANSSENS, “The Problem of Human Freedom in Ibn Sīnā,” in: P. LLORENTE – A. BOADAS – F. J. FORTUNY – A. GRAU – I. ROVIRÓ (eds), Actes del simposi internacional de filosofia de l’edat mitjana: Vic-Girno, 11-16 d’abril de 1993, Barcelona 1996, pp. 112-118. 4. A. RUFFUS – J. MCGINNIS, “Willful Understanding: Avicenna’s Philosophy of Action and Theory of the Will,” in: Archiv für Geschichte der Philosophie 97 (2015), pp. 160195. 5. L. E. GOODMAN, Avicenna, London/New York 1992, p. 81. 6. For contingency based on matter, see A. L. IVRY, “Destiny Revisited: Avicenna’s Concept of Determinism,” in: MARMURA (ed.), Islamic Theology and Philosophy, pp. 160-171. M. RASHED, “Théodicée et approximation: Avicenne,” in: Arabic Sciences and Philosophy 10 (2000), pp. 223-257, suggests that the celestial efficient causes leave room for individual variation, and thus indeterminacy and freedom, in their sublunary effects.

FREEDOM AND RESPONSIBILITY IN AVICENNA

151

which violates His omnipotence. Faced with these two options, the Mu῾tazilites endorsed human freedom and were prepared to limit God’s power as a consequence, whereas the Ash῾arites stuck to omnipotence, and attempted to ground human responsibility on the difficult idea that human beings somehow “acquire” (kasb, iktisāb) the acts that they are necessitated to perform.7 Avicenna would have been aware of this discussion, and thereby of the problem of responsibility and freedom. In the following, I begin with a concise overview of Avicenna’s theory of voluntary agency, arguing that he holds voluntary agency to be responsive to reasons but thoroughly determined by the agent’s beliefs concerning the relevant goals, instruments, and qualifying circumstances (section 1). I then argue that although voluntary acts are not straightforwardly determined by causes extrinsic to the agent, her motivating beliefs in turn are causally determined, and so it seems that there is little room for a causally undetermined will in Avicenna’s theory (section 2). After considering the question of whether Avicenna’s determinist theory of voluntary agency is reconcilable with ethical responsibility (section 3), I conclude by addressing possible objections to my interpretation that arise from existing scholarship, and by briefly reflecting on whether Avicenna might have opted for the sort of view that is now called compatibilism in the dilemma of determinism and responsibility (section 4). 1. Voluntary agency All physical activity is a kind of motion, which Avicenna divides into three types: forced, natural, and voluntary.8 In addition to this, he makes a further distinction within the class of voluntary motion between uniform and variable motion, with the uniform motion belonging to the celestial spheres, and the variable to human and 7. The standard account of this debate is D. GIMARET, Théories de l’acte humain en théologie musulmane, Paris 1980. The foregoing characterisation is a simplification, because many Mu῾tazilites actually held human acts to be determined by motives, and thus not free in the sense of being caused by an entirely undetermined will (see GIMARET, Théories, pp. 30-35, 47-49, 57-60). However, this does not change the nature of the debate, which connects freedom with responsibility, although it does cast doubt on the coherence of the Mu῾tazilite doctrine. 8. AVICENNA, al-Samā῾ al-ṭabī῾ī, I, 5, 1, ed. J. MCGINNIS, Provo 2010, p. 37.

152

JARI KAUKUA

non-human animals in the sublunary world.9 It is the last category that we shall focus on. In contemporary theory of action, actions are frequently distinguished from mere physical events by virtue of their being intentional, or directed to some end, but in the Aristotelian framework Avicenna endorses, teleology is not a differentiating factor. Here, all natural motion is directed to an end specific to the nature of the moving thing, starting from the most basic level of the elements, with earth and water tending downwards, and air and fire upwards. In the final analysis, even forced motion is teleological, insofar as it ultimately goes back to some natural or voluntary motion,10 although in this case the goal may not belong to the thing that is forced to move. Instead of teleology, the distinguishing feature of voluntary action is that the end is an actual object of the agent’s cognition (idrāk) and that the agent opts for the end by virtue of a deliberate will (irāda) or choice (ikhtiyār), even if the means to the end may not always be conscious.11 For instance, my typing this sentence is a voluntary act by virtue of my thinking of the sentence as something I want to bring about, although the way in which I carry out the act of typing through the muscles of my hands and fingers involves a great deal of unconscious physical activity. To put the goal’s central role in another way, Avicenna says that the cognised end of a voluntary act is the most important one of the four causes that enter into the production of the act, for an actual belief about and a drive towards the act’s goal are necessary conditions for initiating the act, and thus the principle that first actualises the other causes, and in particular the efficient cause of the extraction and contraction of muscles.12 Interestingly, Avicenna extends this model to concern also those trivial acts (῾abath) that we seem to perform without any apparent end, such as a man’s fiddling with his beard, for even in such cases, close analysis 9. AVICENNA, al-Samā῾ al-ṭabī῾ī, I, 5, 2-3, pp. 38-39. 10. Cf. AVICENNA, Ilāhiyyāt, X, 1, 6, ed. M. E. MARMURA, Provo 2005, p. 360. 11. AVICENNA, al-Samā῾ al-ṭabī῾ī, I, 14, 11, pp. 97-98; cf. AVICENNA, Ilāhiyyāt, VI, 5, 3, pp. 220-221; VI, 5, 7, p. 222; and AVICENNA, Nafs, IV, 4, ed. F. RAHMAN, London 1959, pp. 194-195. 12. AVICENNA, Ilāhiyyāt, VI, 5, 28-31, pp. 228-229; for an extended analysis of this point, see R. WISNOVSKY, “Notes on Avicenna’s Concept of Thingness (šay᾿iyya),” in: Arabic Sciences and Philosophy 10 (2000), pp. 181-221.

FREEDOM AND RESPONSIBILITY IN AVICENNA

153

will reveal that there is some imagined end, however faint, that the activity is aiming at.13 On the basis of this brief review, we can define an act as voluntary insofar as the agent’s faculty of volition or choice initiates the act by opting for an end that the agent cognises. Thus, such a voluntary act arises from within the agent and is not understandable without reference to the agent’s cognition. To put this another way, voluntary agency is based on reasons, and because of this, we can justify our acts by appealing to those reasons. In a justification of my act, I simply give expression to the goal (or goals) that I actually entertained in performing the act. Finally, because voluntary acts are performed for reasons, they are responsive to reasons, that is, we can influence a person’s voluntary agency by directing her attention at reasons relevant to her act. Let us look into these reasons and their relation to the will at some more detail. 2. Absolute and determined will For Avicenna, will is a faculty that differentiates a voluntary agent from a naturally moving creature. When it comes to sublunary things, this difference manifests in the variability of voluntary agency: naturally moving things move in a uniform way, whereas voluntary agents act in complex, even seemingly idiosyncratic ways depending on a variety of factors. For instance, unless forced upwards or sideways, a rock always falls down because of its natural drive towards the ground and the centre of the world, and given an open space with sufficient sunlight, a pine grows upwards because of the innate tendency of its vegetative soul. By contrast, a person is able to move in all six directions, to perform a wide variety of kinetic and acoustic gestures, and so forth. This is because her faculty of will, considered in isolation, does not determine the goal she will adopt in each given circumstance. Instead, the will of a voluntary agent acquires its goal from one of the cognitive faculties of the soul: The will’s desire is a voluntary matter. It is a volition of something sensible that is sought, such as pleasure, of something pertaining to estimation and imagination, such as winning, or of something pertaining to 13. AVICENNA, Ilāhiyyāt, VI, 5, 10-16, pp. 223-224.

154

JARI KAUKUA

opinion, namely, the good according to opinion. That which seeks pleasure is the appetite, that which seeks winning is the irascible, that which seeks the good according to opinion is opinion, and that which seeks the true, pure good is the intellect. This seeking is called choice (ikhtiyār).14

This indeterminacy of the will allows Avicenna to speak of a certain contingency with respect to it. It can be directed to goals of very different kinds, such as sensible pleasure and the avoidance of sensible pain, which are related to the faculty of appetite and its co-operation with the internal senses. Alternatively, the will can be directed to goals that pertain to things like the interrelations between conspecific creatures, such as winning or overpowering another individual, obtaining the respect of one’s peers, or avoiding shame and dishonour in one’s community. These come from the irascible faculty and its co-operation with the internal senses, in particular with the estimation. Finally, the will’s goal may be concerned with the good in a more general sense, pertaining to the question of how best to lead one’s life as a whole. Avicenna further specifies that this sort of goal may be due to different epistemic processes, ranging from mere opinion to a true intellectual conception of the goal proper to one’s essence or nature. Opting for one of these goals is an act of choice by the will. Such a range of goals that one can choose might seem to leave conceptual space for a notion of genuinely free will: faced with a variety of choices, the will can freely, or without any extrinsic determination, choose from among the alternatives presented by the different cognitive faculties. However, once we look into Avicenna’s theory of the process underlying the choice, things begin to look different. A crucial distinction he makes here is between absolute or merely inclined will, that is, will in the sense of a faculty that is not determined to act in any specific way by its nature but can adopt a number of different goals depending on the circumstances, and determined will, or will that has adopted a specific goal. He writes: Some of these faculties (quwan) that are principles for movements and acts are associated with reason and imagination […]. [1] Those that are associated with reason and imagination are of the same genus as reason 14. AVICENNA, Ilāhiyyāt, IX, 2, 12, p. 312; cf. DE CILLIS, Free Will and Predestination, pp. 38-39.

FREEDOM AND RESPONSIBILITY IN AVICENNA

155

and imagination, for it is almost the case that man and non-man are known by one faculty, and it belongs to one faculty to estimate the matter of pleasure and pain and, in general, to estimate something and its opposite. Likewise, each of these faculties is a faculty over something and its opposite. [2] However, in reality it will not be a complete faculty – that is, a complete and actual principle of change from one thing to another, insofar as it is other – unless will (irāda) is associated with it, triggered (munba῾itha) either by an estimative belief that follows an appetitive or irascible [act of] imagination or by an intellectual view that follows an intellectual thought or a conception of an intellectual form. [3] Hence, when that will is associated with it and it is no longer a [merely] inclined will (irāda mumayyala) but a decisive will (irāda jāzima), namely that which is a resolution (al-ijmā῾) that necessitates the movement of the organs, then it inevitably becomes the principle of a necessitated act (ṣārat lā maḥālata mabda᾿an li-l-fi῾li bi-l-wujūb), for we have shown that unless a cause becomes a necessitating cause (῾illatan bi-l-wujūb), such that something is necessitated by it, no effect will come into existence by virtue of it. Before this state, the will was only weak, the resolution not having taken place. When these faculties associated with reason are taken in isolation, then even if that upon which they are to act were present and stood in such a relation to them that were they to perform an act upon it, they would act through [that relation], one would not act necessarily by virtue of them, and they would still be [mere] potencies (quwwa).15

Let us pay attention to three things in this rich passage. First, in [1], Avicenna states that the faculties that initiate acts are cognitive faculties. These consist of perceptions of one’s circumstances as well as estimates and beliefs related to those perceptions, such as that the dog that I perceive in front of me has the capacity and the intention to hurt me. Here it is important to note that each of these faculties is capable of entertaining opposite kinds of objects. For instance, although I may initially have had a belief that the dog poses a threat, I may be persuaded to approach it by the dog’s owner who insists on its friendliness, and consequently change my belief about the dog’s intention. This confirms that the cognitive principles of agency are responsive to reasons. Second, in [2], Avicenna claims that a cognition does not suffice to initiate an act on its own; for this, willing is required to tip the balance from not acting to acting. However, the will does not make 15. AVICENNA, Ilāhiyyāt, IV, 2, 10, p. 133.

156

JARI KAUKUA

its choice in a vacuum, but is in turn triggered by a cognitive act. This may be an estimation, perhaps based on something perceived earlier, or a belief that results from rational deliberation. For instance, I may have prior traumatic experience with dogs, due to which I have an irresistible urge to flee from them, and consequently the persuasion by this dog’s owner fails to convince me. This brings us to the third point, which Avicenna makes in [3]: once the present cognition and the will that inclines one way or another are combined, a resolute decision to act emerges in the agent, and only this combination is sufficient to initiate the act. Based on this passage, Avicenna’s will is not a faculty of free choice between alternatives, but rather boils down to a tendency to react in new circumstances based on instinctive disposition, temperament, or acquired habit. As such, willing consists of a cognitive element, a belief, which includes an evaluation concerning a certain type of object. This means that as all cognitive content in the human mind, our individual wills are determined by their causal histories. Avicenna is an empiricist in the sense that he does not recognise any innate ideas in the mind: all that we know is either perceived by the senses, or the result of a complicated processing of perceptual material that he calls abstraction (tajrīd). Each episode in this process of obtaining cognitive content is causally determined. Sensation takes place because extrinsic objects excite the sense organs in an appropriate manner, and the soul is drawn to pay attention to the event.16 Similarly, the first steps in the abstractive process follow naturally once the senses have provided the initial input. Aided by the active intellect, the human soul forms the first intelligible concepts and the axiomatic first principles of intellection without any voluntary effort, and further knowledge accumulates on this basis according to the same principles.17 Thus, in a person who has developed sufficiently to function 16. AVICENNA, Nafs, II, 2, pp. 66-67; cf. AVICENNA, Ta῾līqāt, 10, ed. S. H. MOUSATehran 2013, pp. 30-31. 17. Avicenna’s theory of abstraction is a matter of scholarly debate, especially concerning the question of whether the active intellect participates in it. For clear presentations of the two opposite points of view, see D. N. HASSE, “Avicenna on Abstraction,” in: R. WISNOVSKY (ed.), Aspects of Avicenna, Princeton 2001, pp. 39-72; and D. L. BLACK, “How Do We Acquire Concepts? Avicenna on Abstraction and Emanation,” in: J. HAUSE (ed.), Debates in Medieval Philosophy: Essential Readings and Contemporary Responses, New York / London 2014, pp. 126-144. For attempts to situate the debate in the broader VIAN,

FREEDOM AND RESPONSIBILITY IN AVICENNA

157

as an autonomous agent in her environment, each new mental event is brought about either by the environment or by prior mental events: bodily processes of digestion may give birth to a perceived need to eat something, a perception of a dog may awaken an association of threat, or a thought of starving children may result in a willingness to donate to a charity. However, regardless of whether the immediate cause of the mental content is extrinsic or intrinsic, each mental event, including volitions, emerges by virtue of a necessitating cause: The volitions (irādāt) that belong to us have come into being after not being, and everything that comes into being after not being has a cause. Hence, every volition that belongs to us has a cause. The cause of this volition is not a volition, regressing to the infinite, but things terrestrial and celestial that accede from outside. The terrestrial things terminate in the celestial, and the combination of all that necessitates the existence of the volition.18

What is more, Avicenna declares that this causal determination of the initial impulse of our voluntary agency can be traced back to God, the necessitating ground of all existence: The power (qudra) in us is potency (quwwa), and when one of the two alternatives [of] a potency is not preponderant by virtue of some preponderating factor, it is not more appropriate than the other alternative. Hence, in our power there is inevitably something that comes to us from the outside, and that which comes to us determines the act. It is preordained by God (bi-l-taqdīri min Allāh), such that the preordination has initiated that determining factor. What comes to us from outside is, for instance, motives and volitions concerning necessities but [also] other [things]. Our power is not devoid of contingency, and so all our acts are by virtue of the decree, and all our acts are coerced (bi-l-jabr). Hence, when [something] coming from outside does not preponderate upon our potency, the act is not possible.19

In the final analysis, that which determines our volition to execute one act instead of another comes from without us, and ultimately from God Himself, who has determined this world in such a way that framework of Avicenna’s philosophy, see T. ALPINA, “Intellectual Knowledge, Active Intellect and Intellectual Memory in Avicenna’s Kitāb al-nafs and Its Aristotelian Background,” in: Documenti e studi sulla tradizione filosofica medievale 25 (2014), pp. 131-183; and J. KAUKUA, “Avicenna’s Outsourced Rationalism,” in: Journal of the History of Philosophy 58 (2020), pp. 215-240. 18. AVICENNA, Ilāhiyyāt, X, 1, 12, p. 362; cf. X, 1, 6, p. 360. 19. AVICENNA, Ta῾liqāt, 4, pp. 23-24; cf. AVICENNA, Ilāhiyyāt X, 1, 12, pp. 362-363.

158

JARI KAUKUA

it could not be improved upon or changed in any regard. In the end, Avicenna even strikingly admits that this undermines the difference between voluntary and forced agency. Thus, instead of a ground of freedom, the indeterminacy of the absolute will turns out to designate a mere potency to act. Any actualisation of that potency, just as with other things that come to be after not having been, requires a cause. In the case of volitions, those causes are reasons that the agent entertains through her cognitive faculties. As episodes of cognition, the reasons in turn come to be in time, and thus require causes. In corroboration of the foregoing, Avicenna presents us with an analysis of voluntary agency, which explicitly denies that it presupposes a free choice between two alternative courses of action. The context is theological and concerned with the question of how we should understand God’s act of creating the world – is God determined to create by virtue of His essence, or does He create in a more radically free fashion, by virtue of a choice that is not determined by anything, including His own essence? Some anonymous opponents of his had claimed that acting voluntarily, or having the power (qudra) to act, entails a capacity to either act or not to act that is completely undetermined. Thus, they would agree that an agent’s having power is accurately captured in a pair of conditional propositions: (P1) if the agent chooses to act, she acts, and (P2) if the agent does not choose to act, she does not act. The opponents’ view is that in the case of free agency, the truth of the consequent in both propositions varies with the truth of the antecedent. Thus, a description of free agency requires us to commit to both propositions, and consequently, to two states of affairs: God’s not choosing to act, as a result of which the world does not yet exist (from P2), and His choosing to act, as a result of which the world comes to be (from P1). In his counterargument, Avicenna emphasises that the propositions must be understood in their conditional form. This means that they may both be true even if the truth-values of their antecedents remain constant, as long as the truth-value of the consequent appropriately follows that of the antecedent. Thus, if we consider God as the agent

FREEDOM AND RESPONSIBILITY IN AVICENNA

159

of an eternal creation, which of course is Avicenna’s own view, we find that the antecedent of P1 is always true, as is the consequent – and this is a distribution that renders the conditional proposition true. On the other hand, the antecedent of P2 is always false, as is its consequent, and these truth-values also render the conditional true. Thus, on the opponents’ own principles, God can be said to act voluntarily even though His agency does not vary.20 Although the focus here is theological, Avicenna’s conclusion concerning the criterion of voluntary agency applies generally: all that is required for an act’s being voluntary is that an appropriate causal relation, expressed in P1 and P2, obtains between the agent’s will and her act, and this does not entail any capacity of doing otherwise. The outcome is that from a metaphysical point of view, Avicenna’s theory of voluntary agency does not have to be an exception to his necessitarianism, whereas the appeal to reasons in the causal explanation of voluntary agency suffices to distinguish it from mere physical events. Nevertheless, the question still remains of whether such a reconciliation is a sufficient ground for an ethically viable notion of responsibility. 3. Responsibility An optimistic suggestion to the question concerning ethical responsibility would be that even if Avicenna’s metaphysics leaves no room for a radical concept of freedom, perhaps the responsibility of a human agent for her acts can be grounded in her capacity to respond to reasons. Thus, we would be morally responsible for acting out of the right motives, or for failing to do so once we are presented with them. In this sense, a person who is made aware of the moral standards of her community is subsequently responsible for abiding by those standards in her activity, and should she fail to do so, the community (or God) is justified in subjecting her to consequential retribution. The problem, of course, is that even the capacity to respond to reasons in the right way, or being responsive to the right reasons, is causally determined. If a person is temperamentally (or genetically, as we would now say) so determined, an obsessive desire for alcohol, say, may frequently overwhelm concerns of health and conceptions of a 20. AVICENNA, Ilāhiyyāt, IV, 2, 8-9, pp. 132-133.

160

JARI KAUKUA

harmonious life without alcohol, even if she sincerely endorsed those concerns. By the same token, abuse and neglect in childhood may adversely affect a person’s capacity to endorse values central to her community. Thus, once we trace back the causes of why someone is or is not properly responsive to reasons, we find that they were never really up to the person herself. One might think that Avicenna’s adherence to essentialism complicates the matter. In any case, realism about natural essences allows him to say that in every human agent, there is an intrinsic natural drive towards a goal specific to human beings, namely to the acquisition of knowledge and good life in accordance with the acquired knowledge. Since it is concomitant to a person’s essence, this drive remains in her, regardless of how serious her dependence to a substance is or how horrible the abuse she has suffered.21 And since the intrinsic drive to the true and the good is inherently responsive to the right reasons, perhaps Avicenna can say that deviations from the norm are the responsibility of individual agents, and thus justly punishable. To put this another way, human individuals are responsible for cultivating their innate drive towards the humanly good. The normative idea of acting according to one’s nature tallies well with the concept of freedom that I have outlined, and thereby with Avicenna’s determinism: insofar as one acts according to the innate drive of one’s essence, one acts freely, and no radically free will is required.22 As we have seen, this implies no capacity to ever choose otherwise than one is determined to do. Thus, Avicennian freedom is exclusively negative: it is freedom from extrinsically induced motives to act contrary to one’s nature. Or to put the same in positive terms, freedom is the innate drive’s power over contrary extrinsic motives.23 But even if we approve of this way of reconciling freedom with determinism, the question of responsibility returns. It does not take 21. On the resulting conflicts within the person’s soul, see, for instance, AVICENNA, Ilāhiyyāt, IX, 7, 8-9, pp. 349-350; and IX, 7, 14-15, p. 351. The state proper to one’s nature is a balanced state, along the lines of Aristotelian ethics; see Ilāhiyyāt IX, 7, 21-22, pp. 354-355. 22. This also fits well with Avicenna’s theological analysis of God’s freedom, which he takes to mean nothing but the lack of extrinsic coercing factors; see AVICENNA, al-Ishārāt wa-l-tanbīhāt, namaṭ 7, ed. J. FORGET, Leiden 1892, pp. 159-160. 23. Cf. DE CILLIS, Free Will and Predestination, pp. 77-78; and RUFFUS – MCGINNIS, “Willful Understanding,” p. 187.

FREEDOM AND RESPONSIBILITY IN AVICENNA

161

a great deal of reflection to see that the constant presence of the natural drive towards the good has little impact on the general theory. Despite its constancy, the drive remains one factor among many in the complete causal determination of an individual act, and since the cases, in which circumstantially induced motives overpower the natural drive, are all causally determined, the incapacity to follow one’s nature is never up to the person who is doomed to fail. Just as one cannot be held responsible for having the nature one has, one cannot be held responsible for being causally determined to deviate from it. When we consider the human individual, she is determined to act (or react) in a certain way in certain circumstances by the complex complete causes that include her nature and the subsequent qualifications to it. Little ground remains for personal responsibility. In light of the foregoing, it is not surprising that Avicenna resorts to an allegorical interpretation, when faced with the problem of how to justify the Qur᾿ānic account of reward and punishment: Perhaps you will also say: If preordination pertains, why is there punishment? Reflect on the answer, namely that the punishment belongs to the soul by virtue of its mistakes. As you will come to know, it is like the sickness the body obtains from over-indulgence, for it is a concomitant bestowed upon it by past states, the occurrence of which is inevitable, as is the occurrence of what follows them. If this took place in some other way and involved an extrinsic principle, the story would be different. Besides, even if one did grant an extrinsic executor for the punishment, this would also be appropriate (ḥusn), because a deterrent must be among the causes that are established, for it is beneficial in most cases, and assent reinvigorates the deterrent. Should it happen, by virtue of the causes included in the preordination, that someone goes against the result of both the deterrent and consideration, commits an error, and falls prey to sin, the assent would still be necessary by virtue of the common objective, although it is unfavourable for that individual and unnecessary for a merciful chooser (min mukhtārin raḥīm). If there were only the side of the one afflicted by the preordination, the particular affliction would not yield him a great common and universal good. However, one does not pay attention to the particular for the sake of the universal, just as one does not pay attention to the part for the sake of the whole, but amputates and hurts a limb in order to heal the body as a whole.24 24. AVICENNA, Ishārāt, namaṭ 7, pp. 188-189; cf. AVICENNA, Ta῾līqāt, 141, p. 110; AVICENNA, Ilāhiyyāt IX, 7, 16-18, pp. 352-353; and IX, 7, 23-26, pp. 355-357.

162

JARI KAUKUA

In fact, Avicenna here presents two solutions to the problem of responsibility. The first interprets the revealed account of final punishment allegorically: the alleged punishment for vicious acts is their natural consequence according to principles of habituation. This best possible world that we inhabit is construed in such a way that what may seem like a reward in this life, like the material benefits acquired by a morally corrupt agent, actually reinforces the habits that result in a miserable afterlife, where the person accustomed to corporeal pleasures will suffer from their absence. However, there is also another way of thinking about the punishment – one that completely sidesteps the question of responsibility, instead of solving it. Here the religious story of divine punishment is morally appropriate, because it acts as a deterrent, guiding people towards virtue by means of their fear of the punishment. Towards the end of the passage, when considering this justification of the Qur᾿ānic account, Avicenna displays striking disregard for the concern of responsibility: even if some individuals turn out to be at the receiving end, without ever having been able to choose otherwise, this is justifiable in light of the great amount of common good the deterrent yields for mankind as a whole. The cards just had to be dealt in this way in the best possible world. Such nonchalance at the plight of the person determined to sin raises the question of just how seriously Avicenna took the ethical concern. Regardless of our answer to this question, one thing our passage clearly shows, I believe, is that Avicenna recognised the tension between his metaphysics and the ethical idea of responsibility. I also believe that it betrays his reluctance to revise his deterministic metaphysics because of this. But before judge the case closed, let us look more closely at possible objections. 4. Objections and the question of compatibilism Let us begin with the objections that target Avicenna’s metaphysical determinism, and by so doing, attempt to clear room for robust contingency, genuine freedom, and consequently, human responsibility. In this vein, Lenn Goodman has suggested that Avicenna’s modal metaphysics leaves room for genuine indeterminacy, for according to Avicenna’s famous formula, all things other than God are contingent

FREEDOM AND RESPONSIBILITY IN AVICENNA

163

by virtue of their essence (or by virtue of themselves), and necessary only by virtue of their complete cause, which consists of God’s activity but also various contingent factors. In the case of a human act, this means that the agent’s will enters into its complete cause, and in Goodman’s view, this reconciles determinism with the sort of freedom required for responsibility.25 As I have tried to show in the foregoing, it is true that Avicenna considered the human will to be indeterminate, but only when the will is considered in isolation from the motives, which determine it to act in one way or another and without which it remains in potency. In this regard, Avicenna’s analysis of voluntary agency is in perfect consonance with his general modal metaphysics: there is room for contingency in his necessitarian universe, but only when we consider the limits of the causal power an essence has over the full range of factually existing features of its individual instantiations. For instance, a dog’s essence does not determine the exact height of the animal, the colour of its hair, or its temperament. This does not mean that these features are not causally determined; on the contrary, the complete cause of the existence of the individual dog includes factors beyond the essence, such as the designated material substrate that the essence informs, and once all such factors enter into the equation, nothing in the individual dog remains undetermined. Similarly, in the case of a determined act of the will, there will always be a cognitive reason that determines the will to choose in a particular way, and in case there are conflicting reasons in the agent’s mind, there will always be a sufficient difference that tips the balance in favour of one or the other. Since the reasons are causally determined in turn, the contingency and the consequent freedom are not properties of any factually functioning will but reduce to our point of view that focuses on the will in abstraction from the reasons and motives that determine the will to function in the way that it does. By contrast to Goodman, Alfred Ivry locates indeterminacy, and thus freedom, in the material substrate of human being. As the bearer of the possibility of generated and corruptible things, matter in its core is receptive to any form, and thus radically undetermined. To put this another way, it introduces an element of unforeseeability to the 25. GOODMAN, Avicenna, pp. 88-89, 92-93.

164

JARI KAUKUA

unfolding of the necessary causation that begins from God. Thus, somewhat paradoxically, human beings are free insofar as they are material creatures.26 Ivry is right to point at a difficulty in Avicenna’s theory of matter, namely matter’s dual role as the receptive cause and as the bearer of possibility. As the receptive cause of forms that partially determines the individual properties of an enmattered instantiation of an essence, matter is already informed, if not by anything more complex, then at least by elemental forms. For instance, when a human soul comes to animate an embryo, it does not connect to prime matter completely divested of forms, but to a volume of matter that has been preliminarily organised by the interplay of the semen and the ovum in the mother’s womb.27 The situation is different with prime matter as the bearer of pure possibility. When it first comes to receive elemental forms, it cannot determine the reception in any way, given that as a principle of pure potentiality, it has no actual features that could be responsible for this. Avicenna seems to think that prime matter is prepared to receive the elemental forms by the motions of the celestial spheres, which have different effects on different parts of prime matter.28 This is not an entirely unproblematic theory,29 but we can accept it for sake of the argument, because it allows us to point at the central problem in Ivry’s claim. Consider the two aspects of matter I have just outlined. In the first case, as a receptive cause, matter is no longer completely indeterminate, but determined by some forms for the reception of other forms. Consequently, matter is just as tightly part of the network of necessitating causes as the essence, which Goodman proposed as a ground of freedom, and the same argument 26. IVRY, “Destiny Revisited,” pp. 167-168. 27. For the generation of the human soul in Avicenna, see Th.-A. DRUART, “The Human Soul’s Individuation and Its Survival after the Body’s Death: Avicenna on the Causal Relation between Body and Soul,” in: Arabic Sciences and Philosophy 10 (2000), pp. 259-273. 28. See K. RICHARDSON, “Avicenna and Aquinas on Form and Generation,” in: D. N. HASSE – A. BERTOLACCI (eds.), The Arabic, Hebrew and Latin Reception of Avicenna’s Metaphysics, Berlin/Boston 2012, pp. 251-274, esp. pp. 265-266. 29. As Shihāb al-Dīn al-Suhrawardī (d. 1191 CE) points out, it is not clear how prime matter, purely without forms, can have distinct parts that are variously receptive to the influence of celestial motion, and yet this assumption is necessary for Avicenna’s explanation of the emergence of the elements. See SUHRAWARDĪ, Ḥikmat al-ishrāq, I, 3, 4, 79, ed. J. WALBRIDGE – H. ZIAI, Provo 1999, pp. 57-58.

FREEDOM AND RESPONSIBILITY IN AVICENNA

165

holds: regardless of which of the possibilities allowed by the matter’s receptivity gets realised, its realisation is necessary by virtue of its complete cause. On the other hand, in the second case, as bearer of pure possibility, prime matter lacks the power of determination altogether, and the question of which elements are realised in which part of prime matter is entirely determined by the celestial motions. Thus, in view of neither aspect of matter is there any room for radical indeterminacy, and consequently for any sense of freedom. In an interesting paper concerning the problem of evil in Avicenna and Leibniz, Marwan Rashed has argued that according to Avicenna, the relations between the motions of the celestial spheres cannot be expressed by means of rational numbers, and that this is important because it suffices to reject the possibility of the eternal return of temporal cycles.30 The constellations of celestial motion never recur, and consequently each individual in world history is unique. Now, Leibniz applies a very similar insight to tackle the problem of evil by grounding indeterminacy and freedom in this individuality, which he takes to fall outside of the scope of celestial determination of sublunary affairs: evil in the best of all worlds is the responsibility of individual human beings, whose acts are not thoroughly determined by extrinsic causes. Although Rashed does not say that Avicenna makes the same move, we can again, for sake of the argument, consider this as a possible objection to my determinist interpretation. Here, I would challenge the underlying assumption according to which the lack of identical cycles entails gaps in causation. It seems to me that all that would result from the irrational relations between the celestial motions is that there is no return of identical cycles, but this does not have to affect the principles of causation in any way. The celestial motions would still be part of the complete cause of each sublunary individual, and together with other parts of the complete cause, would determine that individual down to the smallest details of her existence. Thus, I believe Rashed shows commendable charity of interpretation when he refrains from attributing this attempted solution to Avicenna. In their study on Avicenna’s theory of voluntary agency, Ruffus and McGinnis argue that there might be “an option to will or not to 30. RASHED, “Théodicée et approximation.”

166

JARI KAUKUA

will in some, perhaps extremely limited, cases”.31 This is because in exceptionally gifted human beings, such as prophets and saints, the essential innate drive towards the true and the good predominates unhindered by extrinsic causes that could make them deviate from the norm. If free agency is possible for some human beings, it is in principle possible for all, and so our shortcomings with regard to what is in principle possible to us are our own responsibility. At first glance, this is an especially attractive interpretation, because it builds on the same concept of freedom that Avicenna applies to God: in both cases, freedom is unhindered activity in accordance with one’s essence.32 Although Ruffus and McGinnis focus on freedom, it is clear that this way of conceiving the capacity to act without an extrinsically caused motive also corroborates a libertarian understanding of responsibility: as long as there is a capacity to choose otherwise, we have a sufficient ground for responsibility. However, as I have argued in the foregoing, Avicenna’s analysis of the power to act explicitly rejects the capacity to choose otherwise. The will is undetermined on its own, but as soon as a sufficiently strong motive is introduced, it is necessitated to choose the corresponding objective.33 Acting according to one’s innate drive does not require such a capacity either, for it can be perfectly well understood as the constant predomination of its objectives over all alternatives. And with the freedom goes the ground for responsibility: just as I have not chosen my nature, I have not chosen the circumstantial causes that lead me to deviate from it. Finally, Jules Janssens has argued that remarks endorsing human freedom and responsibility in some of Avicenna’s brief ethical treatises preclude a determinist interpretation of his metaphysics. For instance, in a treatise concerning responsibility (῾ahd), Avicenna writes that one “has to make a strong personal effort in order to increase (the powers) of his soul, according to the measure of force which was given to the soul”, and that “[w]hen a moral quality is not occurring to him, it is possible for man to acquire it by himself, or, even when he himself 31. RUFFUS – MCGINNIS, “Willful Understanding,” p. 187. 32. RUFFUS – MCGINNIS, “Willful Understanding,” pp. 185-189. 33. Ruffus and McGinnis recognise that most human activity is causally determined (see RUFFUS – MCGINNIS, “Willful Understanding,” pp. 185-186). The difference between our views is that I do not think Avicenna’s system allows for any exceptions to this rule.

FREEDOM AND RESPONSIBILITY IN AVICENNA

167

agrees with the existing moral quality, change by his will to the contrary of that quality”.34 According to Janssens, the personal effort required of an individual implies freedom, albeit limited by the given power of the soul, and in order for a person to be able to actively acquire and change her moral qualities, she must be free. Janssens also refers to the general religious idea that one’s destiny in the hereafter is one’s moral desert, and concludes that in Avicenna’s theory, “[m]an has some freedom of choice in developing his moral and intellectual capacities”.35 Janssens’ case is a lucid formulation of the dilemma: Avicenna should be an indeterminist, if we take the claims of his ethical treatise at face value. And yet, he does not seem to have made any of the required concessions in any of his central works – on the contrary, we have seen him try to explain away the moral desert in the afterlife in a way that leaves his metaphysics intact. The interpretative challenge posed by the seemingly voluntarist ethical texts can also be turned around: given Avicenna’s explicit formulation and use of the principle of sufficient reason, as well as his analysis of voluntary action as governed by causally determined reasons, it seems that if we insist on the literal truth of the ethical claims, his metaphysics collapses. As far as I can see, the most charitable interpretation of the ethical remarks is to read them along the same lines as we read Aristotle’s theory of the cultivation of moral virtues: a human being can and should do her share in becoming a better person, but she is only able to do this if she already has a sufficiently strong prior motivation to do so, and since such a motivation is a product of one’s temperament and upbringing, no one can acquire it from scratch and on her own. Furthermore, when it comes to the problem of reconciling ethics with metaphysics, Avicenna is in good company: the dilemma of necessity and responsibility has haunted all robust determinists at least since the Stoics, and it is not without reason that Kant doomed the postulates of the theoretical and the practical reason to remain irreconcilable. 34. The translations are from JANSSENS, “The Problem of Human Freedom,” p. 115, with a minor correction of a misprint in the second text. For the Arabic texts, see AVICENNA, Risāla fī l-῾ahd, in: AVICENNA, Tis῾ rasā᾿il fī l-ḥikma wa-l-ṭabī῾iyyāt, ed. ANON., Cairo n. d., pp. 142-151, esp. pp. 142 and 146. The same material is discussed, with conclusions similar to Janssens’, in RUFFUS – MCGINNIS, “Willful Understanding,” p. 186. 35. JANSSENS, “The Problem of Human Freedom,” p. 117.

168

JARI KAUKUA

To sum up, the picture conveyed by the foregoing study resembles a theory one might today characterise as a form of compatibilism. There is room for a concept of freedom in Avicenna’s universe, but only for freedom in the sense of acting according to one’s nature, free from, or powerful over, extrinsic factors. This sort of freedom rests on a robust theory of essence: each entity has an essence that comes with a specific drive to a specific end, and those instantiations of the essences are free, in which that drive rules unhindered by opposing forces. In the case of human beings, the specific end is inseparable from our cognitive capacities, and consequently, cognitive causes (or reasons) function prominently in all activity that is specifically characteristic of us. Like most modern versions of compatibilism, Avicenna thus rejects the concept of a radically free will, and tries to explicate voluntary agency as responsiveness to reasons. Also like most versions of compatibilism, his theory ultimately boils down to straightforward determinism: our responsiveness to reasons is caused in part by our essence, and in part by the circumstantial factors of our prior existence, in neither of which there is any room for us to intervene. Thus, even if we accepted this analysis of freedom, the question remains of whether it can ground our ethical concept of responsibility. Since the debate that Avicenna inherited from the Mu῾tazilites and the Ash῾arites was primarily concerned with responsibility, and only secondarily with freedom, insofar as it was perceived as a viable ground for responsibility, we have reason to think that this was also Avicenna’s main concern. In light of the foregoing, it is difficult to avoid the conclusion that like so many before and after him, he failed to quite hit the target.36 Jari KAUKUA Department of Social Sciences and Philosophy - University of Jyväskylä Opinkivi Keskussairaalantie 2 P.O.Box 35 Fi-40014 University of Jyväskylä [email protected] 36. I would like to express my sincere gratitude to the anonymous referee of the journal as well as the audience who heard and commented on a previous version of the paper in a University of Helsinki web conference. The shortcomings that remain despite their perspicacious comments are due exclusively to me. The research that went into this paper was generously funded by the European Research Council (grant agreement no. 682779).

ITTISĀL, CONTACT, OF THE HUMAN AND SUPERNAL INTELLECTS IN AVICENNA: THE GOAL OF ETHICS, THE FOUNDATION OF EPISTEMOLOGY Dimitri GUTAS Abstract In his works, Avicenna refers to the moment of intellection by the human rational soul as a contact, ittiṣāl, (not “junction” or “union”) between it and the intellects of the celestial spheres which think the specific intelligible. Avicenna derives this concept, which forms the foundation of his epistemology and the goal of his ethics, from Aristotle’s description of the same process as the “touching” of the intelligible by the intellect in Metaphysics Lambda 1072b14-26, which Avicenna explains in his commentary on the passage. An appendix includes the edition of the Greek text of this passage as it was found in the MS which was used by Usṭāth for his Arabic translation, and the edition of this Arabic translation as it was read by Avicenna.

Among the more recent of his numerous and invaluable contributions to the study of Avicenna – practically a one-man library of scholarship on the philosopher to act as foundation for future work – is our honoree’s edition and translation, with his colleagues Geoffroy and Sebti, of the commentary on Book Lambda of the Metaphysics.1 Upon re-reading it with the same care and attention as those expended on it by him and his colleagues, I was led to the realization of the origins and extreme significance of the concept of ittiṣāl in Avicenna, which I offer here in gratitude and appreciation.2 1. M. GEOFFROY – J. JANSSENS – M. SEBTI, Avicenne (Ibn Sīnā), Commentaire sur le livre Lambda de la Métaphysique d’Aristote (Chapitres 6-10), Paris 2014, henceforth cited as GJS. 2. I offered a brief preliminary report on the subject in my article on “Ibn Sina [Avicenna],” in: The Stanford Encyclopedia of Philosophy, September 15, 2016: http://plato. stanford.edu/archives/fall2016/entries/ibn sina/, note 7, which was then used with much needed elaboration by M. A. RAPOPORT in his discussion of ittiṣāl in the Ishārāt: “Sufi Vocabulary, but Avicennan Philosophy. The Sufi Terminology in Chapters VIII–X of Ibn Sīnā’s al-Išārāt wa-l-tanbīhāt,” Oriens 47 (2019), pp. 145–96, at pp. 176-86.

170

DIMITRI GUTAS

The imperative to know, and to know rationally, which is the motivation behind Avicenna’s conception and then realization of his scientific system, is based on Aristotle’s concept of happiness as the activity of that which differentiates humans from all other organic life, of the mind: The activity of the intellect is thought to be distinguished by earnest effort (σπουδή, ijtihād), since it employs contemplation (θεωρία, ra᾿y), and it does not desire to have any other end at all except itself; and it has its proper pleasure (ἡδονή, ladhdha) ... Complete happiness (εὐδαιμονία, sa῾āda) is this.”3

Avicenna subscribed fully to this view of human happiness in this world, and extended it to make it also the basis for happiness in the next – as a matter of fact, he made it a prerequisite for happiness in the next. Only the contemplative life, while with the body, prepared the intellect, which has to use the corporeal external and internal senses to acquire knowledge and gain the predisposition for thinking the intelligibles, for the contemplative life after death. In understanding the goal of human life in this manner Avicenna was again being true to the Aristotelian view of divine and human happiness as the contact between thinking and thought in that fateful passage of the Metaphysics Book Lambda (1072b14-26). Avicenna read this passage in the Arabic translation by Usṭāth, as follows:4 Nature is for us like a thriving state for a short time. Now this state, when it comes to that one, it is forever, but for us it is impossible, since pleasure is an activity also of that one, and for this reason wakefulness, sense-perception, and understanding are pleasant. As for hope and memory, they take the place of these. As for understanding, the one in itself, it is of what is best in itself, and that which is most is of that which is most. That which understands itself is the intellect as it acquires the intelligible, because it becomes intelligible when it touches and understands. Therefore the intellect/intellection and the intelligible are one and the same because that which is receptive of the intelligible and the substance is intellect. And it is active just because it possesses, and 3. ARISTOTLE, Nicomachean Ethics X, 7, 1177b19-25, cited in the Arabic translation which Avicenna read, in: The Arabic version of the Nicomachean Ethics, ed. A. A. AKASOY – A. FIDORA, with an introduction and annotated translation by D. M. DUNLOP, Leiden 2005, p. 561. 4-8. 4. The Greek text, as read by Usṭāth, and the Arabic text, as translated by Usṭāth, and the references, are given in the Appendix.

ITTISĀL OF THE HUMAN AND SUPERNAL INTELLECTS

171

therefore it is thought that the intellect possesses that divine [entity] more than this one. [Having a] theory also is something very pleasant and excellent. And if the deity is always like the state in which we sometimes are, then this is marvelous; and if it is more, then it is even more marvelous.

In the Inṣāf, Avicenna gives a running commentary of this as follows (in bold showing the words which Avicenna quotes directly from Usṭāth’s text):5 Aristotle said: Nature is for us like a thriving state, etc. … We say that we, despite the weakness with which we form conceptions of the powerful intelligibles and our immersion in a corporeal nature, we do manage inconspicuously to have a contact with the First Real appear to us, which then becomes like a marvelous happiness for a very short time. This state it possesses always, but for us it is impossible because we are corporeal and it is not possible for us to scrutinize [the contents of] the divine radiation except [for whatever we can snatch away] hastily and inconspicuously. Since pleasure is an activity also of that one, that is, it is being in actuality that is not a state of being affected, it is a perfection of perception and being in actuality that is perceptive precisely of the suitable specific [object]. And for this reason it derives pleasure from being awake because it is a perception, and from sense-perception because it is being used and not left idle. And understanding [i.e.], forming concepts, is pleasant in itself because it is complete life, for life is, as it were, the ability for this. Hope and retention [in memory] are pleasant in the place of understanding, because hope makes what is potential as if it were an [actually] perceived existent from whose image we immediately derive pleasure, while retention, that is, recollection of things that produce joy, makes it as if it were [actually] existent; but as for understanding, it gives pleasure only because of itself. Then he says: As for the foremost understanding in itself, it is of what is best in itself; and as for what understands itself, it is the substance of the intellect when it acquires the intelligible, because it becomes intelligible right away as soon as it touches it, as it were. And intellection, that which intellects,6 and the intelligible become one and the same only with regard to the essence of the thing as it relates to itself, and so there is then a single essence, which is the intelligible form, just 5. GJS p. 57, 142, p. 59, 151-167. Square brackets include words that extrapolate the implications of the Arabic and facilitate the English idiom. Emphasis is added. 6. The Arabic for nous is ῾aql, as rendered by Usṭāth, but ῾aql is also the verbal noun, “intellecting, intellection.” Avicenna takes this meaning as the one intended here and thus adds the third item, ῾āqil, “that which intellects.”

172

DIMITRI GUTAS

as it may be said that it possesses its essence, that is, its essence is not separable insofar as it is intelligible in itself. Then he said: And if the deity is always like the state in which we sometimes are, then this is marvelous; and if it is more, then it is even more marvelous.

In this passage from the Metaphysics, the metaphor of “touching”, or contact, appears as the nodal point of the experience which constitutes the telos, goal, of ethics as well as the means of intellection, the goal of epistemology. The state of bliss through intellective knowing as the goal of ethics, as described here by Aristotle, and fully accepted and endorsed by Avicenna in his commentary, needs little elaboration. Avicenna made it also the basis of bliss in the afterlife, as already mentioned, but in this too he was making explicit what was implied in Aristotle’s understanding of the immortal “agent intellect” (De anima, 3, 5) whose posthumous existence, since by definition it must always be active and thinking, can only indicate the posthumous bliss of the individual. The metaphor of “touching” and contact as the means and goal of intellection, on the other hand, needs some elucidation. Aristotle uses this metaphor to describe all apprehension, whether physical (as in 1051b24-25) or mental, as in this passage. What the metaphor means in Aristotle is explained as follows by W. D. Ross: The metaphor of contact in the description of simple apprehension [in 1051b24] recurs in Λ, 1072b21. Its implications are (1) the absence of any possibility of error, which characterizes the apprehension of the ἴδια αἰσθητά (cf. De An. 430b29), (2) the apparent (though on Aristotle’s view only apparent, De An. i. 11) absence of medium in the case of touch. τὸ θιγεῖν [touching] means an apprehension which is infallible and direct.7

Avicenna found the concept useful and employed it throughout his works. The metaphor of contact, ittiṣāl, was particularly apt in describing mental apprehension, something that can be described only metaphorically, since both the human rational soul and the active intellect, as well as the other celestial intellects all the way to the “First Real,” are immaterial substances. There is no word to 7. W. D. ROSS, Aristotle’s Metaphysics. A Revised Text with Introduction and Commentary, 2 vols., Oxford 1958, vol. 2, p. 277. See also S. H. Rosen, “Thought and Touch: A Note on Aristotle’s De Anima,” in: Phronesis 6 (1961), pp. 127-137.

ITTISĀL OF THE HUMAN AND SUPERNAL INTELLECTS

173

describe the “contact” of two immaterial substances – that is, how to describe, in concrete terms, how the human soul relates to the active intellect when the former acquires an intelligible, which by its very nature is actually thought of a-temporally by the active intellect, or, to use another metaphor, is “stored” in it since the active intellect is the only location where intelligibles can actually exist in perpetuity when we are not thinking them. Avicenna thus actually explains, in the very passage in the commentary on Lambda just cited, what this “contact” entails as he paraphrases Aristotle; he says, “and as for ‘what understands itself, it is’ the substance of ‘the intellect’ when ‘it acquires the intelligible, because it becomes intelligible’ right away as soon as ‘it touches it’ (yulāmisuhu), as it were.” Here the word “touching”, yulāmisuhu, which renders the Aristotelian θιγγάνων, explains “contact,” ittiṣāl. The apprehension of the intelligible, as in Aristotle, is instantaneous and “direct” (in Ross’s terms). The Greek circumstantial participle θιγγάνων, normally taken by modern translators to indicate here manner or means (“through touching, in touching”), is interpreted temporally by Usṭāth, who translates, “when it touches” (ḥīna yulāmisu). Avicenna specifies ḥīna to make it more precise, and interprets it as “right away as soon as it touches” (fī l-ḥāli, ka-mā yulāmisuhu).8 Establishing the instantaneity of the contact enables Avicenna to avoid the pitfall generated by what Aristotle next says: “And intellection, (that which intellects), and the intelligible become one and the same,” which to an uncritical reader would indicate the identity of the two or three entities involved in the contact, or, in plebeian terms, the identity or junction of the human intellect and the supernal intellects up to and including the first intellect. Unfortunately, this is precisely how the word ittiṣāl was understood and accordingly was mistranslated as “junction, conjunction, or even union” by Davidson, whose influential study led to the prevalence of this understanding in subsequent literature.9 8. It is to be noted that ka-mā here means “as soon as” (see M. ULLMANN, Wörterbuch der klassischen arabischen Sprache, Wiesbaden 1970, vol. 1, p. 9a 17) and not “as if”, which would render a hypothetical sense to the touching (as interpreted in GJS, p. 58, “de la même manière que s’il touchait”), which would be the opposite of what Avicenna wants to (and Aristotle does) say: the contact is real and instantaneous. 9. H. A. DAVIDSON, Alfarabi, Avicenna, and Averroes, on Intellect, Oxford / New York 1992, pp. 89, 103-105.

174

DIMITRI GUTAS

However, the implications of such an identity or junction would be disastrous. These terms mean becoming one, and thus the human intellect would become one with the active intellect, or, as in the case of the text in Avicenna’s commentary above, with the First Real. This in turn would imply that the human rational soul loses its identity as it unites with the active intellect (or, blasphemously, with the First), and that it knows all intelligibles, like the active intellect, once it becomes one with it — for how would then the human soul, now undifferentiated from the active intellect, be able to acquire only one of the myriad intelligibles that the latter thinks always? And most egregiously, all this – essentially the human soul becoming divine once it has united with the active intellect or the First – would happen while the human soul is still with the body! Avicenna avoids the pitfall by specifying in which way the two or three entities involved in the contact become one and the same, and that is through the particular intelligible form which is being acquired at the very instant of contact, and only at that instant, at the moment of thinking (κατὰ μετάληψιν τοῦ νοητοῦ, and, as Avicenna interprets it, fī l-ḥāli, ka-mā yulāmisu [al-ma῾qūla]). He says, and I emphasize: “And intellection, (that which intellects), and the intelligible become one and the same, only with regard to the essence of the thing as it relates to itself (wa-innamā yakūnu l-῾aqlu wa-l-῾āqilu wa-l-ma῾qūlu wāḥidan bi-qiyāsi dhāti l-shay᾿i ilā nafsihī) and so there is then a single essence, which is the intelligible form.” This is merely to say, that at the moment of intellection, of the contact, and only for that instant, what is identical in all entities involved in the contact is the one intelligible form – which would almost go without saying: obviously when the human intellect seeks an intelligible, following the processes just recently minutely described by Alpina,10 and acquires it as it is being thought by the active intellect (which thinks it and all the other intelligibles a-temporally), the intelligible it acquires and the intelligible thought by the active intellect are one and the same. And mutatis mutandis, since the First, as the supreme intellect-in-actuality also thinks ceaselessly all the intelligibles, including the one that is sought by the human intellect at any given moment, the human intellect can also be said to come into contact, at the instant of intellection, with 10. T. ALPINA, Subject, Definition, Activity, Berlin / Boston 2021.

ITTISĀL OF THE HUMAN AND SUPERNAL INTELLECTS

175

the First as well. Avicenna says, in a statement expressed indirectly, “we do manage inconspicuously to have a contact with the First Real appear to us,” the indirection and the adverb intending to indicate that the First is neither the primary source of the intelligible nor the first recourse for it. This is indicated by the term he uses, ikhtilās, in ῾alā sabīl al-ikhtilās, repeated later in the sentence in the hendiadys khaṭfatan wa-khulsatan (the two words being practically synonymous), which means to snatch something hastily away when the owner is unaware. The further metaphor of the lightning or rain bearing clouds, which the observer scrutinizes to detect where and how the lightning or rain might fall (nashīmu l-bāriqata), drives the indirection fully home: the human intellect cannot scrutinize the full arsenal of intelligibles in the First Real seeking something, but it has to gain access to a particular intelligible through contact with the chain of supernal intellects and only indirectly and inconspicuously, during the fleeting moment of intellection, have the contact with the First Real appear to it (fa-yaẓharu lanā ttiṣālun bi-l-ḥaqqi l-awwali).11 Ittiṣāl, then, is an epistemological term, and not ontological or theological, describing metaphorically the contact (and decidedly not “the junction, conjunction, or even union”) between the human intellect and the intelligible thought by the celestial intellects at the very instant of intellection. To round out the description of intellection in Avicenna, it suffices to add here that this contact between the human intellect and the celestial intellects is possible only because they all belong to the same genus, they are congeneric, mujānis, as Avicenna says in a number or places.12 “We are made of star stuff,” to quote Carl Sagan.

11. Avicenna refers numerous times and in similar terms to the instantaneous and ephemeral nature of this contact with the First Real. See the discussion and references in RAPOPORT, “Sufi Vocabulary,” pp. 166-167. 12. See, for example, the discussion in D. GUTAS, “Intellect Without Limits: The Absence of Mysticism in Avicenna,” in: M. C. PACHECO – J. F. MEIRINHOS (eds.), Intellect et imagination dans la philosophie médiévale (Actes du XIe Congrès International de Philosophie médiévale de la S.I.E.P.M., Porto, du 26 au 31 août 2002), Turnhout 2006, pp. 351-372, at pp. 358-359.

176

DIMITRI GUTAS

APPENDIX: THE TEXTS

I Avicenna read the passage from the Metaphyics discussed here (1072b14-26) in the translation of Usṭāth, who in turn translated the Greek text which he found in his source MS, Σ (now not extant). This is the Greek text which Usṭāth read and translated, which I punctuate as he read and parsed it, as indicated by his translation. The apparatus that follows compares the readings in Σ with those in the other Greek MSS, as recorded by Werner Jaeger in his edition, whose MS sigla I have followed, and explains Usṭāth’s translation choices. The superscript numbers are those of the lines in Bekker’s Greek text. The references in the apparatus follow these lines. ¹4Καὶ ἡ φύσις διαγωγή ¹5ἐστιν οἵα ἡ ἀρίστη μικρὸν χρόνον ἡμῖν. Οὕτω γὰρ ἀεὶ ἐκεῖνό ἐστιν, ¹6ἡμῖν μὲν γὰρ ἀδύνατον, ἐπεὶ καὶ ἡ ἡδονὴ ἐνέργεια τούτου, ¹7καὶ διὰ τοῦτο ἐγρήγορσις, αἴσθησις, νόησις, ἥδιστον. Ἐλπίδες ¹8δὲ καὶ μνῆμαι διὰ ταῦτα. Ἡ δὲ νόησις ἡ καθ’ αὑτήν, ¹9τοῦ καθ’ αὑτὸ ἀρίστου, καὶ ἡ μάλιστα τοῦ μάλιστα. Αὑτὸν 20δὲ νοεῖ ὁ νοῦς κατὰ μετάληψιν τοῦ νοητοῦ· νοητὸς γὰρ 21γίγνεται θιγγάνων καὶ νοῶν, ὥστε ταὐτὸν νοῦς καὶ νοητόν, 22τὸ γὰρ δεκτικὸν τοῦ νοητοῦ καὶ τῆς οὐσίας νοῦς. Ἐνεργεῖ δὲ 23ἔχων, ὥστ’ ἐκεῖνο μᾶλλον τούτου [ὃ] δοκεῖ ὁ νοῦς θεῖον ἔχειν. 24Καὶ ἡ θεωρία τὸ ἥδιστον καὶ ἄριστον· εἰ οὖν οὕτως ἔχει ²5ὡς ἡμεῖς ποτὲ ὁ θεὸς ἀεί, θαυμαστόν, εἰ δὲ μᾶλλον, ἔτι ²6θαυμασιώτερον. 14 Σ, like Π, omitted δ’ after διαγωγή, and thus the translator began the new sentence with καὶ ἡ φύσις. 15 Σ, like Π, read ἐκεῖνό ἐστιν, with the verb. 16 Σ, like Π, read ἡ ἡδονή for ἡδονὴ ἡ. 17 The Arabic translated the superlative ἥδιστον as ἡδύ. 23 ἐκεῖνο μᾶλλον τούτου is the reading of the Greek MSS. Following Alexander of Aphrodisias, Ross corrects it to ἐκείνου μᾶλλον τοῦτο (in his edition cited in note 6), followed by Jaeger. 23 ὃ is not translated; either it was missing in Σ, or Usṭāth skipped it or paid no attention to it. 24 Σ and J omit εὖ before ἔχει.

ITTISĀL OF THE HUMAN AND SUPERNAL INTELLECTS

177

II The Arabic translation by Usṭāth of this very text was the following. This is the text in Bouyges,13 which follows MS B (in Bouyges’ sigla); I correct it to reflect the original translation, as indicated by the quotations in Avicenna. I record the details in the apparatus that follows. The superscript numbers are those of the lines in Bekker’s Greek text. The references in the apparatus follow these lines. ،‫ فهذه الحال أ ّما ذاك فهو أبد ًا‬.‫صالحة زمان ًا قليل ًا‬15 ‫والطبيعة لنا كحال‬14 ‫والحس‬ ‫ولهذه الع ّلة اليقظة‬17 ،‫وهو لنا غير ممكن فإنّ اللذة ِف ْع ٌل لذلك أيض ًا‬16 ّ ٌ ِ 18 ‫ أ ّما الرجاء‬.‫لذيذ‬ ِ ‫فللذى‬19 ‫ وأ ّما الفهم الذى بذاته‬.‫مكان هذه‬ ‫والذكر ف ِل‬ ‫والفهم‬ 20 ‫ والذى يفهم ذات ََـه هو‬. ‫أكثر‬ ‫أكثر فللذى هو‬ ‫ والذى هو‬،‫هو أفضل بذاته‬ َ َ 21 ً ُ‫ فإذ ًا العقل‬،‫معقولا حين ُيلامس ويفهم‬ ُ ‫العقل باكتساب المعقول؛ فإنّه يصير‬ ُ ٌ ‫إذ‬23 ‫ وإنّما يفعل‬.‫عقل‬ ‫لأنّ قابل المعقول والجوهر هو‬22 ٌ‫شئ واحد‬ ٌ ‫والمعقول‬ 24 ً ‫شئ‬ .[‫أكثر من هذا ]أيضا‬ َّ‫ظن أنّ ِللعقل ذلك الإلاهى‬ ُ ٌ ‫والرأى أيض ًا‬ ّ ‫ فإذ ًا ُي‬،‫له‬ َ ُ ٌ ٌ ْ ‫وفاضل؛‬ ‫كحالنا فى وقت ما فذلك‬25 ‫>حال< الإلاه أبد ًا‬ ‫فإن كان‬ ‫لذيذ جدّ ًا‬ 26 ْ ، ‫عجيب‬ .‫فأكثر َع َجب ًا‬ ‫أكثر‬ ‫وإن كان‬ َ َ 15 reading ‫ زمان ًا قليل ًا‬for ‫ زمان قليل‬in B. Ibn Sīnā has ‫ فى زمان قليل‬which seems to be his own syntagma rather than reflect the text of the translation he was using. 22 reading ‫( يفعل‬ἐνεργεῖ) for ‫يعقل‬. 23 reading ‫( للعقل‬ὁ νοῦς ... ἔχειν) for‫ أيض ًا | العقل‬should be secluded; it is not in the Greek and does not appear in most of the witnesses used by Bouyges for the Arabic. 24 The transmitted Arabic text in Usṭāth’s translation here reads, fa-in kāna l-ilāhu abadan ka-ḥālinā fī waqtin mā, which must be supplemented as follows to make it acceptable Arabic, fa-in kāna l-ilāhi abadan ka-ḥālina fī waqtin mā. Without this addition the Arabic would mean, “if the deity is always like our state sometimes,” which makes little sense literally. The addition of the word ḥāl stands for the Greek οὕτως ἔχει, and the sentence then corresponds perfectly with the original Greek in MS Σ, εἰ οὖν οὕτως ἔχει ὡς ἡμεῖς ποτὲ ὁ θεὸς ἀεί (omitting εὖ before ἔχει), and Themistius’ paraphrase, 13. M. BOUYGES, Averroès. Tafsīr mā ba῾d aṭ-ṭabī῾at, 3 vols., Beirut 1938–52, vol. 3, pp. 1609-1614.

178

DIMITRI GUTAS

which Avicenna was following closely, fa-in kāna mā huwa li-llāhi dā᾿iman bi-manzilati mā huwa la-nā fī ba῾ḍi l-awqāti,14 where the repeated mā corresponds to the repeated ḥāl in Usṭāth’s translation. Since the word ḥāl is missing from all the available witnesses of Usṭāth’s translation (assuming the available editions are accurate), it must have fallen out of the text very early in its transmission (as a primitive error). By contrast, the word εὖ in the Greek is not translated in the Arabic, but this is due to its absence in the Greek manuscript used by Usṭāth, as it is also absent in the Greek manuscript J and in Themistius’ paraphrase as well, in whose tradition the Greek exemplar Σ for Usṭāth’s translation and the manuscripts of Themistius must have belonged.

Dimitri GUTAS Near Eastern Languages and Civilizations - Yale University 320 York Street New Haven, CT 06511, USA [email protected]

14. Arisṭū ῾inda l-῾Arab, ed. ‘A BADAWĪ, Cairo 1947, pp. 17-18. Cf. R. Brague, Thémistius. Paraphrase de la Métaphysique d’Aristote (Livre Lambda), Paris 1999, esp. pp. 26-27, 90-94, and 142-143.

AVICENNA ON THE HOUSEHOLD? Thérèse-Anne DRUART Abstract In this paper I examine whether the attribution of On Governance or Management of the Household, an economic text, to Avicenna is sound. The examination of its content reveals differences from Avicenna’s views on ethics in other texts: 1. it is not grounded in prophecy or eschatology; 2. it does not show Aristotelian influences but rather Galenic and, of course, Bryssonian ones. A close comparison with Metaphysics, X, 3-5 in the Shifā’, which deals with similar issues, makes it clear that on several of them the two texts disagree. Besides, whereas Metaphysics, X, 3-5, argues carefully, On Governance often offers no justification for its views or poor arguments, and even exhibits some inconsistencies. I conclude that the attribution of this text to Avicenna is most unlikely.

We all know Jules Janssens’ dedication and extensive contribution to the study of Ibn Sīnā’s works in Arabic and Persian as well as in the Medieval Latin translations. We also know how much his detailed and critical bibliographies support the work of all Avicennian scholars. We can never be grateful enough for them. As a token of my appreciation and gratitude for his work I am offering a new examination of a text attributed to Ibn Sīnā. At times, this text is known as On Governance or even as Politics (al-Siyāsa) and at times as Management of the Household (Tadbīr al-manzil).1 For 1. There are multiple editions of this text: 1. editio princeps by L. MALOUF, as “Kitāb al-siyāsat,” in: al-Machriq 9 (1906), pp. 967-973, 1037-1042, 1073-1078; reprint in L. MALOUF – C. EDDE – L. CHEIKHO (eds.) Traités inédits d’anciens philosophes arabes musulmans et chrétiens, Beirut 1911, pp. 2-17 (this latter book itself was reprinted in Cairo in 1980); 2. by S. J. AL-NAQDĪ, in his al-Marshad al-῾Irāqiyya li-qirā’a, Baghdad 1929, of unknown origin according to T. Hoff, p. 23 (see n. 2); 3. by F. AHMAD, with an introduction, in his Majmū῾ fi l-siyāsa, Alexandria 1982, pp. 61-111 [seems to be based on the editio princeps]; 4. a more critical ed. by SHAMS AL-DĪN in his Al-madhhab al-tarbawī, Beirut 1988, pp. 232-260, reprint in ZAY῾ŪR, Mayādīn al-῾aql al-῾amalī, Beirut 2001, pp. 142-166; 5. an uncritical ed. by H. NASHSHĀBAH in his Al-Turāth al-tarbawī, Beirut

180

THÉRÈSE-ANNE DRUART

easiness of reference I shall simply call it On the Household as this is its main theme. In 1993 after a new look at the Leiden Mss. 1020 Tineke Hoff translated this text into Dutch and provided a commentary.2 As Janssens indicates “she does not really put into question the authenticity of the attribution to Ibn Sīnā, but, at the same time, indicates elements which sound rather un-Avicennian.”3 In his 2nd edition of Avicenna and the Aristotelian Tradition, Dimitri Gutas lists this text under “Pseudepigraphs, to be authenticated” and refers to Janssens’ note.4 Hoff pointed to the unusual Inshā’ style5 of this treatise but she also was aware that in his autobiography Ibn Sīnā claimed he could imitate the style of eminent writers and so may have amused himself in writing one text in this style.6 She also indicated that several later falāsifa, such as al-Tūsī, who uses this text in his Nasirean Ethics, attribute it without hesitation to Ibn Sīnā.7 Though the Leiden manuscript does not specifically indicate its author, it does include On the Household among other texts clearly attributed to Ibn Sīnā. Taking all these points into account, it seems that arguments against its authenticity can only be based on its content: first, by examining how it fits Ibn Sīnā’s conception of the practical sciences and of its second discipline in particular, that is to say management of the household or economics; secondly, by examining its use of sources; and 1988, pp. 25-45; and 6. an edition without apparatus criticus indicated by J. JANSSENS, An Annotated Bibliography on Ibn Sīnā, Second Supplement (1995-2009), Tempe, Arizona 2017, p. 132, n. 545. I have used both Malouf’s 1911 edition and that (1988) of Shams al-Dīn. 2. T. HOFF, Avicenna, Over het beheer van huis en haard, Kampen 1993. J. MCGINNIS – D. C. REISMAN translated the Arabic text into English under the title On Governance in their Classical Arabic Philosophy: An Anthology of Sources, Indianapolis 2007, pp. 224-237, 390, which I shall quote, except otherwise indicated. I shall also refer to its division in numbered paragraphs. There is also a French translation by Y. SEDDIK, Bryson et Ibn Sînâ, Penser l’économique, Tunis 1995, pp. 35-55 with Malouf ’s 1906 Arabic text on pp. 25-40 of the Arabic. 3. J. JANSSENS, An Annotated Bibliography on Ibn Sīnā: First Supplement (1990-1994), Louvain-la-Neuve 1999, p. 36, n. 156. 4. D. GUTAS, Avicenna and the Aristotelian Tradition: Introduction to Reading Avicenna’s Philosophical Works. Second Edition, Leiden 2014, pp. 500-501. 5. The Inshā’ style is a very ornate style used in the composition of some formal letters, documents and state papers. It is quite different from the style for ordinary or scholarly prose. 6. HOFF, Over het beheer, p. 27, n. 19. 7. Ibid., p. 26.

AVICENNA ON THE HOUSEHOLD?

181

thirdly, by comparing it with a certainly authentic Avicennian text, dealing with some of the same issues, i.e., the Metaphysics of the Shifā’, X, 3-5. 1. Ibn Sīnā’s Conception of the Practical Sciences Many scholars indicate that Ibn Sīnā wrote little on practical philosophy. Gutas, for instance, claims that “the last sections of the Metaphysics of the Shifā’ constitute just about the best and most basic extant expression of his ideas on practical philosophy in all its branches, both the traditional three and the new fourth one on the ‘discipline of legislating’….”.8 Jon McGinnis states “providing a systematic account of Avicenna’s conception of practical philosophy presents something of a challenge for his interpreters” and points to “the scarcity of explicit writings on practical philosophy.”9 Yet, recently Amos Bertolacci highlighted important points in examining its grounding in the metaphysical proof of prophecy found in the Metaphysics of the Shifā’, book X.10 He indicates, as do McGinnis and Gutas, that practical philosophy is grounded in prophecy and eschatology. Bertolacci adds that early on, in his summae, at the end of his presentation of metaphysics, Avicenna ended with eschatology that followed prophecy. A turn occured with the Shifā’. From then on eschatology preceded prophecy, which itself directly preceded practical philosophy in order to better highlight the grounding of practical philosophy in prophecy. On the other hand, On the Household does not make any reference to prophecy or eschatology. That text rests all its presentation on the importance of ordering (tadbīr) and governance (siyāsa) showing how intellect in everything perceives that there is ruler and ruled and that so it should be. It also indicates that discernment (tamyīz) allows to see the difference between good and evil. Hence, that text ignores Avicenna’s mooring of practical philosophy in prophecy and eschatology and proceeds using reason alone, even if it shows its Islamic 8. GUTAS, Avicenna and the Aristotelian Tradition, p. 497. 9. J. MCGINNIS, Avicenna, Oxford 2010, p. 209. 10. A. BERTOLACCI, “The Metaphysical Proof of Prophecy in Avicenna,” in: A. PALAZZO – A. RODOLFI (eds.), Prophecy and Prophets in the Middle Ages, Florence 2020, pp. 68-72.

182

THÉRÈSE-ANNE DRUART

context by two Qur’ānic allusions (the wife as a source of rest for man in 7:189 and 30:21,11 and God as possessor of majesty and munificence in 55:27 and 7812) and also recommends that, as soon as the young boy is ready for it, he starts to learn the Qur’ān and the basics of religion.13 In an early text, the Divisions of Philosophy or On the Divisions of the Rational Sciences, Avicenna gives a brief overview of the three disciplines of practical philosophy: ethics, governance of the household, and politics according to the traditional order. Therein Avicenna relates ethics to eschatology and politics to both prophecy and eschatology and so makes clear the link any practical science has to prophecy and eschatology. As for what concerns the specific contents of governance of the household, Avicenna there tells us the following: The second science pertains to the second division [of philosophy, i.e., practical philosophy]. Through this science one knows how man ought to conduct the ordering (tadbīr) of his household – which is common to him, his wife, his children, his servants, and his slaves – so as to lead a well-ordered life that enables him to gain happiness. It is contained in Bryson’s On the Governance (or Ordering, tadbīr) of the Household and in books by others.14

On the Household certainly addresses all these topics with a strict focus on governance and ordering, but it also adds a section of man’s governance of himself, which at first blush would rather pertain to ethics, the first practical discipline. Besides, it includes a section on the need for most men to have a profession and on the proper use of the money so gained not specified in Avicenna’s description but not excluded either. Avicenna in his description had indicated that the contents of this discipline could be found in Bryson’s On the Governance of the Household and, indeed, Bryson’s text (the Greek is lost but the Arabic version 11. Arabic, Malouf, p. 5; English, n. 7. 12. Arabic, Malouf, p. 10; English, n. 20. 13. Arabic, Malouf, p. 13; English, n. 30. 14. M. MAHDI’s translation with a modification in J. PARENS – J. C. MACFARLAND (eds.), Medieval Political Philosophy: A Sourcebook, 2nd ed., Ithaca, NY 2011, p. 75; Arabic ed., M. KADYVAR, “Ibn Sīnā and the Classification of Philosophy,” in: Jāvīdān Kherad, Sophia Perennis. The Journal of Sapiential Wisdom and Philosophy 5.1 (2009), p. 108.

AVICENNA ON THE HOUSEHOLD?

183

has recently been critically edited) did influence On the Household but mainly on the sections concerning the family in general, the wife and the son and not always extensively. It certainly did not influence the long introduction on governance and ordering.15 Besides, Avicenna himself indicates that there are other sources. Pointing at least to some of them may help us to better understand On the Household and evaluate the authenticity of its attribution to Ibn Sīnā. 2. The Sources of On the Household As Tineke Hoff already said, On the Household shows dependence on Bryson’s Management of the Estate,16 though Bryson’s name does not appear in the text. It is no summary of it or slavish adaptation. The ordering of the sections is not exactly the same in the two texts. Bryson plunges immediately in his subject, which he says comprises four topics: property, servants, the woman, and the child.17 On the Household begins with a fairly lengthy introduction of its own on ordering and governance, which is followed by a general view of the family depending on Bryson. Then come: 1. man’s governance of his soul or of himself,18 a theme missing from Bryson’s text; 2. man’s professional life and how he should use his money, which is rather remote from Bryson’s section on property; 3. man’s governance of his wife; 4. man’s governance of his son; and, finally, 5. man’s governance of his slaves. On the Household has changed the order in which to examine the governance of the members of the household to follow the proper ordering and governance of the family. It goes from highest to lowest beginning with man himself, then the wife, who is the second in command, third the boy ruled by his parents, and finally 15. S. SWAIN, Economy, Family, and Society from Rome to Islam: A Critical Edition, English Translation, and Study of Bryson’s Management of the Estate, Cambridge 2013. Swain provides extensive comments and indicates parallels to Greek sources and various Arabic texts, including On the Household but without hinting at the possible inauthenticity of its attribution to Avicenna. 16. HOFF, Over het beheer, pp. 32-34 and passim; SWAIN, Economy, gives an overview of “Islamic Economics,” pp. 87-94 and points to specific influences on On the Household in his commentary. 17. Swain’s ed., Arabic and English, pp. 430-431. 18. The Arabic is ambiguous as nafs can be used both as reflexive pronoun and as the term for soul in philosophical texts, Ma’louf, p. 6; Shams al-Dīn, p. 240.

184

THÉRÈSE-ANNE DRUART

the slaves who occupy the lowest rank and so fits the ordering indicated by Avicenna himself in his Divisions of Philosophy. It also adds to both Bryson and Avicenna man’s self-governance. What may explain this addition is a sentence from another important source, Galen. Beginning with man’s self-reform seems to take its origin, as Tineke Hoff already pointed,19 from a sentence in Galen’s Character Traits, a text that survived only in Arabic: The aim of the virtuous person is that he should first reform his own soul and then, after that, the souls of all other people over which he has influence, [beginning with] all the closest [to himself]; the closest are the people of the household, then those to whom he is related, then those of his city, then all humankind…20

In On the Household the very first sentence of the section on man’s governance of his soul echoes Galen’s view: Of all the varieties of governance, man would do best to begin with the governance of his soul [or of himself] since his soul [or himself] is the closest thing to him, … and because once he governs his soul [or himself] well, he will not falter in what comes at the next level, namely, governance of the city.21

Curiously, this whole section on man’s governance of himself does not in any way rests on Aristotle’s Nicomachean Ethics, even if in his Divisions of Philosophy Avicenna had indicated that the first practical discipline, ethics as self-governance, should rest on it. It does not speak of the virtues, of virtue as a mean or of means to acquire the virtues, but rather focuses on how to discover one’s moral weaknesses so one can reform one’s character. On the other hand, in the Metaphysics of the Shifā’ at the very end, X, 5 (10-11), Ibn Sīnā speaks of justice as 19. HOFF, Over het beheer, p. 102. 20. The Arabic of Περί ἠθῶν was edited by P. KRAUS, “The Book of Ethics by Galen,” in: Bulletin of the Faculty of Arts of the University of Egypt, 5, part 1 (Arabic) (1937), pp. 1-51; J. N. MATTOCK gave a first English translation in “A translation of the Arabic Epitome of Galen’s Book ΠΕΡΙ ΗΘΩΝ,” in: S.M. STERN – A. HOURANI – V. BROWN (eds.), Islamic Philosophy and the Classical Tradition: Essays presented to Richard Walzer, Columbia, SC 1972, pp. 235-260; D. DAVIES gave a new English translation augmented by supplementary material and an introduction by P.N. SINGER, “Character Traits,” in: GALEN, Psychological Writings, ed. P.N. SINGER with D. DAVIES – V. NUTTON, Cambridge 2013, pp. 109-201. I am quoting Davies’ translation, p. 156. 21. English, McGinnis – Reisman’s transl., n. 11; Arabic, Malouf, p. 6, Shams al-Dīn, p. 240.

AVICENNA ON THE HOUSEHOLD?

185

a mean (10) and presents a sketchy Aristotelian conception of the virtues (11).22 Once again, Tineke Hoff pointed to the influence of Galen’s Character Traits.23 Galen explores the same topic in another text Affections and Errors, also called The Diagnosis and Treatment of the Affections and Errors Peculiar to Each Person’s Soul,24 the first part of which, Ḥunayn tells us, was translated into Arabic.25 This text early on has sections on the necessity and difficulty of self-knowledge and on the need to consult someone else to discover one’s weaknesses.26 Another text by Galen, Good Men Profit by their Enemies, also known in Arabic,27 addressed the same topic and is referred to by Abū Bakr al-Rāzī in his Spiritual Medicine, ch. 4: “Of How a Man May Discover his Own Vices”.28 On the Household develops at length this favorite Galenic theme. The importance of Galenic influences on On the Household gives us pause. Ibn Sīnā was well-aware of Galen’s medical works but does not seem to have paid much attention to his philosophical works. Yet, On the Household restructures Bryson’s text and adds to it precisely under the influence of Galen, while ignoring Aristotle’s Nicomachean Ethics. Besides, the introduction of the whole text pays no heed to prophecy or eschatology but highlights two of God’s attributes, intellect and compassion, which are two of the three attributes on which Abū Bakr al-Rāzī, whom Galen much influenced, grounds his ethics in his Philosopher’s Way of Life.29 Avicenna knows of al-Rāzī’s medical 22. AVICENNA, The Metaphysics of The Healing, A parallel English-Arabic text, translated, introduced, and annotated by M. E. MARMURA, Provo, Utah 2005. Marmura divided each chapter into numbered paragraphs. I shall refer to these numbers, and, except otherwise indicated, I am quoting his translation. 23. HOFF, Over het beheer, p. 104; an English translation of this passage can be found in DAVIES, “Character Traits,” in: GALEN, Psychological Writings, pp. 166-167. 24. English translation from the Greek with introduction by SINGER in: GALEN, Psychological Writings, pp. 203-331. 25. See J. C. LAMOREAUX, Ḥunayn ibn Isḥāq on His Galen Translations, Provo, Utah 2016, p. 120, n. 128. 26. GALEN, Diagnosis, pp. 240-250. 27. See LAMOREAUX, Ḥunayn ibn Isḥāq, p. 120, n. 129. 28. Arabic, Abi Bakr Mohammadi Filii Zachariae Raghensis (Razis), Opera philosophica, ed. P. KRAUS, Cairo 1939, pp. 33-35; English transl. by A. J. ARBERRY, The Spiritual Physick of Rhazes, London 1950, pp. 36-37; French translation by R. BRAGUE, Razi, La médecine spirituelle, Paris 2003, pp. 82-85. 29. For an English translation see MCGINNIS – REISMAN Classical Arabic Philosophy, pp. 36-44; Arabic, Kraus’ ed. of Opera philosophica, pp. 99-111.

186

THÉRÈSE-ANNE DRUART

works but neglects his philosophical texts. That Galen’s works and maybe one of Abū Bakr al-Rāzī rather than some of Aristotle’s works lead to major modifications to Bryson’s text put into doubt that On the Household stems from Avicenna’s pen. 3. Comparison with Avicenna’s Metaphysics of the Shifā’, X, 3-5 As we said earlier, the only Avicennian text we can rely on for a comparison with On the Household is the Metaphysics of the Shifā’, X, 3-5. Let us first look at specific issues, then at the use of sources, and finally at the quality of the argumentation. 1/ Specific issues a. Almsgiving by means of taxes: On the Household tells us: The just way for man to behave with what he has acquired is to spend some of it on taxes, alms and those in need of a favor…He should surrender the costs of taxes and alms happily, with a pure intention, an open heart, and a sense of surety that they are insurance for a day of want… [He should] do that for the sake of God, Lord of splendor and munificence [allusion to Qur’ān 55:27 & 78], expecting neither thanks nor recompense.30

The text, using the term zakāt and so referring to one of Islam’s five pillars, simply states what to do and how to do it but does not explain or justify why it should be done. On the other hand, Ibn Sīnā in Metaphysics, X, IV, (2), though not using the term zakāt, deals with the same issue in referring to Plato’s views in the Republic, while qualifying them: There must exist in the city a common fund, part of it consisting of duties imposed on acquired goods and natural profits, such as fruit and agricultural produce… the fund will serve to meet the exigencies of the common good, to meet the needs of the guardians who are not employed in a craft, and as expenditure on those prevented by maladies and chronic diseases from earning their livelihood. Some people have held the opinion that those among them whose recovery is not to be expected should be killed. But this is bad; for their sustenance will not hurt the city.31 30. English, p. 231, n. 20; Arabic, Malouf, p. 10, Shams al-Dīn, pp. 247-248. 31. Marmura’s transl. and ed., pp. 370-371.

AVICENNA ON THE HOUSEHOLD?

187

The Metaphysics while making the “zakāt” obligatory does not use this term nor ground this “almsgiving” in scripture but in Plato’s views, which for this purpose need to be seriously qualified, as Plato was in favor of expelling from the city those suffering from chronic diseases and so leading to their death. Ibn Sīnā adopts a more compassionate view consistent with that of Islam. The two texts promote “almsgiving” but in completely different ways and in using a different vocabulary. b. The need to work: On the Household explains that there are people who have an easy livelihood by inheritance or collection but that others need to have a profession.32 For the latter the section on child’s education speaks at length about the choice of a profession for making enough money and then getting married.33 Metaphysics, X, IV, (1) tells us: Idleness and unemployment must be prohibited. [The legislator] must leave the way open to no one for acquiring from another that share of livelihood necessary for man while exempting himself from any effort in return. Such people he must vigorously restrain. If they do not refrain [from such practice], he must then exile them from the land. But should the cause here be some physical malady or defect, [the legislator] must set aside a [special] place for them and those like them, under someone in charge.34

The former text speaks of work as something necessary only for those who have no other means to support themselves, but the latter insists that voluntary idleness cannot be tolerated and every man who is able bodied should work. It even prescribes expelling from the city any man who, though able to do so, does not work. On the other hand, illness and disability dispense from this obligation. Such people will be taken care of by means of the common fund spoken about in the previous section. As we will see in the next section, women too are dispensed of this obligation and, therefore, the husband must provide for them. 32. English, n. 18; Arabic, Malouf, pp. 14-15, Shams al-Dīn, p. 246. 33. English, nn. 33-35. 34. Marmura’s transl. and ed., p. 370.

188

THÉRÈSE-ANNE DRUART

Again, on the necessity to work the two texts are not fully compatible. c. Treatment of wives: Both texts claim that the wife should be secluded (this common claim is totally absent from Bryson’s Management of the Household). On the Household, section V, n. 27, indicates that the wife should be strictly secluded, but does not explain why. By contrast Metaphysics, X, 4 (9) gives a lengthy explanation: Since woman by right must be protected, inasmuch as she can share her sexual desire with many; [and since she] is much inclined to draw attention to herself and, in addition to that, is easily deceived and is less inclined to obey reason; and since sexual relations on her part [with many men] cause great disdain and shame, which are well-known harms – whereas [sexual relations] on the part of man [with other women] only arouse jealousy, which one should ignore as it is nothing but obedience to the devil – it is important to legislate that the woman should be veiled and secluded from men. Thus, unlike the man, she should not be a bread-earner.35

The point made in this text that husbands who have sexual relations with other women should ignore their wife’s jealousy contrasts with what is said in On the Household, section V on the governance of the wife, at the end of n. 27, which recommends that a man does not give rise to his wife’s jealousy (presumably by taking another wife).36 On the Household begins the section on man’s governance of his wife with a lengthy list of the qualities of the best woman. As first quality it lists “being intelligent” (al-ʻāqila), whereas Metaphysics, X, 4, insists that women are not very intelligent or rational. It says that she is really of feeble intellect (6) (bi-l-ḥaqīqa wāhiyat al-ʻaql) and quick to follow passion and anger. She is also the less rational of the two [i.e., man and woman] (7) (‘anqas al-shakhṣayni ʻaqlan) and, as we 35. Mamrura’s transl. and ed., p. 373. 36. Here I am following Hoff’s translation, pp. 72-73, “zorgen dat ze volledig gesluierd is en ervan afzien haar jalousie op te wekken [door een andere vrouw te trouwen]” and that of Youssef Seddik, p. 49, “de lui éviter ce qui peut la rendre jalouse,” rather than that of McGinnis – Reisman, p. 233, n. 28, “that she avoid[s] making her husband jealous;” Arabic, Malouf, p. 12, Shams al-Dīn, p. 251.

AVICENNA ON THE HOUSEHOLD?

189

saw earlier, she is easily deceived and is less inclined to obey reason (ʻaql) (9). It is obvious that on some issues, such as the necessity to work and woman’s intelligence, the two texts do not hold the same views. On other issues, the two texts seem to agree, as in the case of wives’ seclusion and almsgiving. Yet, for both almsgiving and the wives’ seclusion, On the Household does not give any reason whereas Avicenna’s Metaphysics does and grounds it in women’s lack of intelligence and their being easily swayed by passion. The two texts do not exhibit much agreement in the positions they defend or in their way of defending them when they happen to hold similar positions. 2/ The use of sources The two texts rely on different sources and use them differently. As I explained, On the Household at times clearly depends on Bryson’s text, though not following it slavishly, but also relies on Galenic views as well as the Islamic context, but ignores Aristotle. The various sources, as we shall see later on, are not very well integrated. For instance, Bryson gives women a much more active role but On the Household’s section on the wives strips them of much of this active role. By contrast, Metaphysics, X, 3-5, does not show any Brysonian influence, even if Ibn Sīnā in his classification of the sciences had spoken of Bryson, but rather depends on Plato’s Republic directly or indirectly and at the end on Aristotle. It is also an attempt to give rational justification for the views in Islamic law on marriage, divorce, etc., and, therefore, is much closer to Islamic views and law stemming from prophetic revelation than is On the Household. 3/ The arguments Finally, if we consider the way the two texts defend their positions, we notice that On the Household often does not much justify the views it presents, except for its insistence on hierarchical ordering and governance, which pervades the whole text, but does not at all constitute a main theme in Metaphysics, X, 3-4. On the Household gives much direction and advice but often does not explain the reason for such direction and advice and at times is not even very consistent. The issue on which

190

THÉRÈSE-ANNE DRUART

the two texts have most in common, i.e., on governance of the wife, illustrates the difference in the quality of argumentation. On the Household at times is not very consistent. In the section on family at the beginning, which relies much on Bryson (7), the author claims that as the husband is busy gathering food, he needs a wife to protect their supply. Not much further (9) he repeats this claim: “He needs a wife to protect his dwelling and watch over his earnings.” But a few lines further on he no longer follows Bryson but moves to consider the husband as a herdsman. Then it becomes the husband’s responsibility to protect the herd by using watchmen and sheepdogs. And the author spells out the analogy to the herdsman in stating that “the man who has wife, children, servants, and followers – in addition to being obliged to protect and care for them….”. It does not seem the wife is any longer protecting anything, except if one were to assimilate her to a watchman or a watchdog. The author here seems to make a patchwork of two different sources. Again, in the section on the wife On the Household says nothing about protecting the supply or the dwelling. In that section different points are made but the connection between them remains unclear. For instance, in the list of some twenty qualities the good wife should have, the first one is being intelligent (24), but in what follows nothing is said about how the wife should use this intelligence and what is the link between being intelligent and strictly secluded (27) as well as “keeping her outward appearance beautiful.” The text hops from one point to another and only in the passage on the need for the wife’s deference to her husband does this whole section alludes to the main theme of the long introduction, i.e., the need to distinguish ruler from ruled. If there is no such deference, then the proper order gets subverted, as the wife becomes the manager (mudabbira) and the husband the one managed (mudabbar).37 On the other hand, Metaphysics, X, 4 (end of [4] to [9]) presents a continuous argument to justify the different prescriptions about the wives. As we saw, after explaining the need for every man to work, paragraphs nn. (1) to (4) end with the following: Also, [those] acts must be prohibited which, even if permitted, would be detrimental to the city’s growth – like fornication and sodomy, 37. English, p. 232, n. 26; Arabic, Malouf, p. 22, Shams al-Dīn, p. 205.

AVICENNA ON THE HOUSEHOLD?

191

which dispense with the best foundation on which the city stands: that is marriage.38

Ibn Sīnā immediately moves to a justification for marriage’s importance: The first of the legislator’s acts must pertain to marriage resulting in issue. He must call and urge [people] to it. For [by marriage] is achieved the continuity of the species, the permanence of which is proof of the existence of God…39

He then requires that wedding and marriage be a manifest affair and explains why. This done he argues for permanence of marriage not only in order not to disturb “the bond between the children and their parents and renew the need of marriage for everyone” but also situates marriage in the framework of the city’s good: …because what is most conducive to the general good is love. Love is achieved only through friendship. Friendship through habit [and] habit is only produced through long association.40

For him it follows that divorce should not be encouraged, though permitted, and its initiative should not be in the hands of the wife “for, in reality, she is of feeble intellect and is quick to follow passion and anger.” (6) and so “the means of separation must not be placed in the hands of the less rational of the two” (7), but there are some regulations to discourage husbands from mistreating their wives or hastily asking for a divorce (7-8). As woman is less rational, other prescriptions follow: Since woman by right must be protected, inasmuch as she can share her sexual desire with many; [and since she] is much inclined to draw attention to herself and, in addition to that, is easily deceived and is less inclined to obey reason; and since sexual relations on her part [with many men] cause great disdain and shame, which are well-known harms – whereas [sexual relations] on the part of man [with other women] only arouse jealousy, which one should ignore as it is nothing but obedience to the devil – it is important to legislate that the woman should be veiled and secluded from man.41

38. Marmura’s 39. Marmura’s 40. Marmura’s 41. Marmura’s

transl. transl. transl. transl.

and and and and

ed., ed., ed., ed.,

(4), (5), (6), (9),

p. p. p. p.

371. 372. 372. 373.

192

THÉRÈSE-ANNE DRUART

From there follows that men may be polygamous, but women may not. The whole section rationally justifies prescriptions of Islamic law about marriage and family life. There is a continuous argument for encouraging the propagation of the human species and, even if one may wish to dispute the premise that women are less rational than men, once one accepts this premise, the argument follows nicely. It also shows that reason and prophetic prescriptions contained in Islamic law are in agreement. The stark contrast between the paucity of arguments in On the Household and the quality of the argumentation in Avicenna’s Metaphysics, X, 4, once again raises doubt about the authenticity of the attribution of On the Household to Avicenna. 4. Conclusion On the Household does not follow closely the criteria Avicenna himself gives for practical philosophy as it does not ground its views on prophecy. Its sources include Galen’s ethical works, which do not seem to be in Avicenna’s purview. It also shows no Aristotelian influence, even in its section on man’s governance of himself. The comparison between On the Household and a certainly authentic Avicennian work, Metaphysics, X, 3-5, shows that on several issues the two texts hold different views and even for similar views the justification may be quite different. Besides, On the Household, is poorly argued and philosophically weak,42 whereas Metaphysics, X, 4, presents a continuous argumentation. Therefore, the authenticity of the attribution of On the Household to Ibn Sīnā does not seem tenable. Thérèse-Anne DRUART School of Philosophy – The Catholic University of America 620 Michigan Ave., N. E. Washington DC 20064, USA [email protected]

42. Bryson’s text is more carefully argued.

ON THE AUTHENTICITY OF THE THRONE EPISTLE (AL-RISĀLA AL-῾ARSHIYYA) ASCRIBED TO AVICENNA Frank GRIFFEL Abstract Judged by the great number of manuscripts, Avicenna’s short Throne Epistle (al-Risāla al-῾arshiyya) is one of the more popular works ascribed to this author. The text explains Avicenna’s understanding of divine attributes and divine actions. This is an interesting subject because Avicenna argues that God has no positive attributes other than pure existence. All predications of the divine are either negative (He is not multiple, not spatially extended, or not ignorant) or they describe relations that He has with His creation. Avicenna seldom wrote about this subject in a systematic way, which is why this text seems to be a rare example where the Eminent Master lays down his views on theology. Certain features of the text, however, suggest that the Throne Epistle was not written by Avicenna but rather by one of his followers of the mid-6th/12th century. This article discusses these features and explains how a genuine text by Avicenna with the title Throne Philosophy (al-Ḥikma al-῾arshiyya) got lost during the Master’s lifetime and likely triggered the production of the well-known Throne Epistle two centuries later. While the latter text is unlikely authored by Avicenna it still gives a faithful presentation of his philosophical views and engages in polemics against adversaries that argue from a background in Ash῾arite and Mu῾tazilite Islamic theology. It is this polemic that betrays the pseudo-epigraphic character of the text.

1. Introduction I am not much in favor of debates about the authenticity of premodern texts. To me they seem like dinner parties, where one makes the acquaintance of a person and then after a few hours of conversation judges that certain statements this person made are out of character or not what we expect from that person. How can one possibly say such things after just a few hours of acquaintance? When it comes to many philosophers of the pre-modern period, particularly of the Arabic

194

FRANK GRIFFEL

and the Islamic tradition, I tend to think that we are in a situation where we just know them for a short period of time and can hardly say what their philosophical personalities – or rather the range of their thought – is and what one should expect from them. In many cases, the books that are available to us are a mere fraction of what these authors have written. It also seems to me that our knowledge of these authors is greatly inferior to that of their students or to readers who lived closer to them in time. Yet these often had no problem accepting works as authentic that we sometimes suspect are pseudo-epigraphies. Disputing the authenticity of such works means claiming to know them better than the author’s close students or early readers. A drastic example of such a misjudgment is the Western reaction to the publication of al-Ghazālī’s The Niche of Lights (Mishkāt al-anwār) in 1904–5.1 This book was largely unavailable in European manuscript libraries and only known through a Hebrew translation. When its Arabic text was published in Cairo, readers in Europe understood that there was a strong “Neo-Platonic” influence on this text, which manifests itself in the use of light-metaphors, for instance, and a style that was much closer to certain philosophical texts of the Arabic tradition than those of kalām, the tradition that al-Ghazālī was associated with.2 Articles and books by William H. T. Gairdner (1873–1928), Arent J. Wensinck (1882–1939), and Miguel Asín Palacios (1871– 1944) discussed what Gairdner had called “the Ghazālī problem.” This was the assumption that the great Muslim thinker had two kinds of teachings, an Ash῾arite exoteric one meant for a larger group of Muslim readers and a Neo-Platonic esoteric one for an elite of close followers.3 1. AL-GHAZĀLĪ, Mishkāt al-anwār, ed. F. Z. AL-KURDĪ, Cairo 1322/1904–05. 2. “Neo-Platonic” here refers “to the school of al-Fārābī and Ibn Sīnā,” see W. M. WATT, “A Forgery in al-Ghazālī’s Mishkāt?”, in: Journal of the Royal Asiatic Society (1949), p. 14 n. 1. 3. W. H. T. GAIRDNER, “Al-Ghazālī’s Mishkāt al-Anwār and the Ghazālī Problem,” in: Der Islam 5 (1914), pp. 121–153; A. J. WENSINCK, “Ghazālī’s Mishkāt al-anwār (Niche of Lights),” in: ID., Semietische Studien uit de Nalatenschap, Leiden 1942, pp. 192–212; ID., “On the Relation Between Ghazālī’s Cosmology and his Mysticism,” in: Mededeelingen der Koninklijke Akademie van Wetenschappen, Afdeeling Letterkunde Deel 75, Serie A, No. 6 (1933), pp. 183–209; M. ASÍN PALACIOS, La espiritualidad de Algazel y su sentido christiano, 4 vols., Madrid / Granada 1934–1941, vol. 3, pp. 247–254; vol. 4, pp. 242–250. I give a report of this stage in the research on al-Ghazālī in F. GRIFFEL, “The Western Reception of al-Ghazālī’s Cosmology from the Middle Ages to the 21st Century,” in:

ON THE AUTHENTICITY OF THE THRONE EPISTLE

195

One of the most drastic reactions was William Montgomery Watt’s (1909–2006) rejection of the authenticity of the so-called VeilSection in Mishkāt al-anwār.4 Watt rejected Gairdner’s esoterism as well as “explicitly Neoplatonic doctrines” in the great Muslim thinker’s works and he argued that the last third of The Niche of Lights, the so-called Veil Section, must be a forgery not authored by al-Ghazālī and that it was unduly inserted into the text of the book.5 Watt’s implicit assumption that a doctrinal enemy of al-Ghazālī must have taken the authentic parts of Mishkāt al-anwār and added to that the problematic Veil Section, led to more radical questions. The assumed possibility of such blatant counterfeiting led Watt and other Western scholars to a more thorough examination of texts ascribed to al-Ghazālī – or parts of texts and problematic passages in texts – and whether they could be forgeries. Earlier Asín Palacios had already identified six texts as candidates for forgeries.6 To these Watt added a list of another five, so that the full number of spurious texts falsely ascribed to al-Ghazālī amounts, in Watt’s mind, to thirteen books plus major sections in another three works.7 This Western obsession with forgeries in the œuvre of al-Ghazālī, as I may call it, continued into the second half of the twentieth century and led to contributions such as Hava Lazarus-Yafeh’s (1930–98) in 1966 and 1975, who dismissed all works ascribed to al-Ghazālī as inauthentic that employ philosophical terminology.8 Yet about the same time when Watt wrote his article on Mishkāt al-anwār, Franz Rosenthal published his study The Technique and Dîvân: Disiplinlerarası Çalışmalar Dergisi / Dîvân: Journal of Interdisciplinary Studies 16 (2011), pp. 33–62, at pp. 51–52. 4. WATT, “A Forgery in al-Ghazālī’s Mishkāt?”, pp. 5–22. 5. Ibid., pp. 5, 21–22. 6. These are al-Durra al-fākhira, Minhāj al-῾ārifīn, Mukāshafat al-qulūb, Rawḍat al-ṭālibīn, al-Risāla al-laduniyya, and Sirr al-῾ālamayn; ASÍN PALACIOS, La espiritualidad de Algazel, vol. 4, pp. 385–390. 7. W. M. WATT, “The Authenticity of Works Ascribed to al-Ghazālī,” in: Journal of the Royal Asiatic Society (1952), pp. 24–45, at pp. 30, 32–42, adds the so-called Hebrew Ajwiba, the Arabic Kīmiyā᾿ al-sa῾āda, al-Maḍnūn al-ṣaghīr, Minhāj al-῾ābidīn, Mi῾rāj al-sālikīn, and Mīzān al-῾āmal, and sections in Bidāyat al-mujtahid, al-Imlā᾿, and Mishkāt al-anwār. 8. H. LAZARUS-YAFEH, “Philosophical Terms as a Criterion of Authenticity in the Writings of Al-Ghazzālī,” in: Studia Islamica 25 (1966), pp. 111–121, reprinted in EAD., Studies in al-Ghazzālī, Jerusalem 1975, pp. 249–263.

196

FRANK GRIFFEL

Approach of Muslim Scholarship, where he explains how respectfully classical Muslim scholars treated the texts of their tradition and how they compared and collated different manuscripts of any given text.9 Rosenthal’s insights make it hard to imagine how anybody of a later generation could have changed a text that was published during the lifetime of the author, that circulated in various manuscripts, and that was continuously studied throughout the centuries. To assume that one could simply insert several pages into an already existing book – and that such changes would remain undetected until discovered by a scholar in the West who had not worked with any of the manuscripts – shows a significant degree of disdain for the seriousness of indigenous Islamic scholarship. After all, major experts on al-Ghazālī such as Tāj al-Dīn al-Subkī (d. 771/1370), al-Wāsiṭī (d. 776/1374), al-Nawawī (d. 676/1277), Ḥājjī Khalīfa (d. 1067/1657), or al-Murtaḍa al-Zabīdī (d. 1205/1791) list the Mishkāt al-anwār among his authentic works and were apparently not disturbed by its content. Watt’s attitude together with that of Asín Palacios and LazarusYafeh was rather typical for Western scholarship on Islam during the mid- and late-20th century. Western scholarship on Islam was like the dinner guest who had just made al-Ghazālī’s acquaintance. His Mishkāt al-anwār was available for just forty years and overall no more than a few dozen scholars in the West had read it. Al-Ghazālī’s overall œuvre was known for about a century but this Western knowledge was based entirely on written texts and detached from his original teaching tradition in the madrasas of Nishapur, Baghdad, or Damascus. The range of teachings that al-Ghazālī’s works support, for instance, were unknown to Gairdner, Wensinck, Asín Palacios, or Watt. Their knowledge of al-Ghazālī was vastly inferior to that of al-Subkī, al-Wāsiṭī, or al-Nawawī. Yet still, Western scholars hotly debated which books of al-Ghazālī were authentic or inauthentic – works that those with superior knowledge had never called into question. Today, Watt’s doubts about the Veil Section’s authenticity (and about many other works) are put to rest and this part of Mishkāt al-anwār is now

9. F. ROSENTHAL, The Technique and Approach of Muslim Scholarship, Rome 1947, pp. 22–27.

ON THE AUTHENTICITY OF THE THRONE EPISTLE

197

seen as key evidence for al-Ghazālī’s clever appropriation of philosophical ideas. 2. How Avicenna’s Throne Philosophy (al-Ḥikma al-῾arshiyya) Got Lost It is this dismal record of earlier Western scholarship that make me be reluctant to engage in discussions about the authenticity of works by pre-modern Arabic and Islamic authors. Yet, despite these reservations I suggest that a work long ascribed to Avicenna that the Arabic tradition itself has never, as far as I know, doubted as being authentic was, so I have come to conclude, not written by him, at least not in the textual form that has come down to us. I would like to stress that this text ascribed to Avicenna contains no un-Avicennan ideas or teachings that cannot be regarded as close to Avicenna. Quite the opposite, the teachings in this text are those of Avicenna or at least extremely close to his. This is not a case where one of Avicenna’s doctrinal enemies foisted this text on him. I rather suggest that there was indeed once an authentic text by Avicenna, or the fragment of a text, that got lost. That loss was well known among readers of Avicenna. Yet the desire to hold in hand this work of Avicenna was so great that the book was indeed “found.” The “finder,” however, was a forger who created a book that could pass as Avicenna’s lost one. The motivation for “finding” – that is forging – this book was most probably the money that book dealers and hence readers of Avicenna would pay for the lost book. Let’s tell that story step by step and let’s begin with the loss of Avicenna’s Throne Philosophy (al-Ḥikma al-῾arshiyya). This part of my detective story relies heavily on research done earlier by David C. Reisman and Dimitri Gutas. In his landmark study on Avicenna’s book al-Mubāḥathāt, Reisman noted that in one of his letters to a student Avicenna describes how he lost a work in progress titled al-Ḥikma al-῾arshiyya. Al-Mubāḥathāt is an important book of Avicenna where later caretakers of his Nachlass gather some of the letters he addressed to students together with what appears to be transcripts from questionand-answer-sessions between the Eminent Master and his students. The book was well known among readers of Avicenna and is quoted, for instance, by Fakhr al-Dīn al-Rāzī (d. 606/1210). He understood its value as a book where Avicenna addresses the most problematic

198

FRANK GRIFFEL

areas of his philosophical system.10 Reisman assumes that the material in al-Mubāḥathāt went through several stages of editing and re-editing during the second half of the fifth/eleventh century. By the time of Fakhr al-Dīn al-Rāzī in the second half of the sixth/twelfth century there might have been a fixed text that likely resembled the one we find in Muḥsin Bīdāfar’s edition of al-Mubāḥāthāt, published in 1992–93.11 Avicenna’s al-Mubāhathāt includes the copy of a letter that he wrote to an unnamed student whom both Reisman and Yahya Michot identified as Bahmanyār ibn al-Marzubān (d. 458/1066).12 He was Avicenna’s master student and at the time of this letters’ composition more a colleague of the Eminent Master and a close follower than a disciple. Already in 1988, in the first edition of his Avicenna and the Aristotelian Tradition, Dimitri Gutas understood the importance of this letter and translated it into English.13 The letter mentions the loss of certain books to hostile troops. Based on that, Gutas dated the letter to shortly after 425/1033–34 when the troops of the Ghaznawid governor of Rayy, Abū Sahl al-Ḥamdūnī (or: al-Ḥamdawī) pillaged Isfahan.14 At this time Avicenna was vizier in the service of Alā᾿ alDawla Muḥammad Ibn Kākūya (d. c. 433/1041), the local ruler of Isfahan. Both Avicenna and Ibn Kākūya left that city and returned two years later after a Kākūyid reconquest. Reisman, however, suggests that Gutas’ dating relies on confusing and ultimately mistaken information from the historian Ibn Funduq al-Bayhaqī (d. 564/1169– 70) and he corrected the dating of the letter to shortly after Muḥarram 421 / January 1030.15 In the second edition of Avicenna and the 10. See e.g. FAKHR AL-DĪN AL-RĀZĪ, al-Mabāḥith al-mashriqiyya, ed. M. AL-MU῾TAṢIM 2 vols., Beirut 1410/1990, vol. 1, p. 122; see F. GRIFFEL, Formation of PostClassical Philosophy in Islam, New York 2021, p. 400. 11. IBN SĪNĀ, al-Mubāhathāt, ed. M. BĪDĀRFAR, Qom 1371/1413 (1992–93). 12. D. REISMAN, The Making of the Avicennan Tradition, Leiden 2002, pp. 208–213; Y. MICHOT, “Le riz trop cuit du Kirmânî. Présentation, édition, traduction et lexique de l’Épitre d’Avicenne contestant l’accusation d’avoir pastiché le Coran,” in: F. DAELEMANS et al. (eds.), Mélanges offerts à Hossam Elkhadem par ses amis et élèves, (Archives et bibliothèques de Belgique 83 [2007]), pp. 81–129 at pp. 91–92. 13. D. GUTAS, Avicenna and the Aristotelian Tradition. Introduction to Reading Avicenna’s Philosophical Works, Leiden 1988, pp. 57–60. 14. Ibid., pp. 56, 136. On Abū Sahl see Ḡ. Ḥ. YŪSOFĪ’s article in Encyclopedia Iranica, vol. 1, ed. E. YARSHATER, London / New York / Costa Mesa 1982, pp. 369–370. 15. REISMAN, Making of the Avicennan Tradition, pp. 208–210. AL-BAGHDĀDĪ,

ON THE AUTHENTICITY OF THE THRONE EPISTLE

199

Aristotelian Tradition, Gutas accepts this correction.16 The loss of these books did not happen when Abū Sahl al-Ḥamdūnī conquered Isfahan but four years earlier when the Ghaznawid Sultan Mas῾ūd ibn Maḥmūd (reg. 421–32 / 1031–40) himself entered the city with his troops. Some historical background might be necessary. The Ghaznawid military conquest of Western Iran (Jibāl) began in 420/1029 when Mas῾ūd ibn Maḥmūd took the city of Rayy when his father Maḥmūd ibn Sebüktegin (reg. 388–421 / 998–1030) was still alive. Whereas that city remained under Ghaznawid control for the next decades, others nearby, such as Isfahan, slipped in and out of their rule. Alā᾿ al-Dawla Muḥammad Ibn Kākūya took control of Isfahan in 398/1008 as a Būyid governor. Later, he made himself independent and expanded toward Hamadan and into Kurdistan.17 Around 414/1024 he offered Avicenna the vizierate of his principality. Before that Avicenna had been vizier to the local ruler of Hamadan, who was Alā᾿ al-Dawla’s enemy. When he decided to switch sides, Avicenna together with his disciple and secretary al-Jūzjānī, his brother, and two slaves sneaked out of Hamadan in the disguise of a group of Sufis.18 After the Ghaznawids had installed themselves in Rayy in 420/1029 they made two attempts of conquering Isfahan, one in 421/1030 led by Maḥmūd ibn Mas῾ūd of Ghazna and a second time in 425/1033– 34 led by Abū Sahl al-Ḥamdūnī, the Ghaznawid governor of Rayy. In both cases, Ghaznawid troops conquered the city and Alā᾿ alDawla Muḥammad Ibn Kākūya fled with his vizier Avicenna. Both managed, however, to return a few years later and re-take control of the city. Whereas Gutas initially believed the loss of several of Avicenna’s books happened during the latter conquest, Reisman argued persuasively that it happened during the earlier one in 421/1030. 16. D. GUTAS Avicenna and the Aristotelian Tradition. Introduction to Reading Avicenna’s Philosophical Works. Second Revised and Enlarged Edition, Including an Inventory of Avicenna’s Authentic Works, Leiden 2014, p. 149. On pp. 49, 484, however, Gutas maintains the dating of the raid to 425/1034. 17. On the Kākūyid’s long and eventual successful struggle against the Ghaznawids see D. DURAND-GUÉDY, Iranian Elites and Turkish Rulers. A History of Iṣfahān in the Saljūq Period, London / New York 2010, pp. 54–65. 18. IBN SĪNĀ, The Life of Avicenna. A Critical Edition and Annotated Translation, ed. and trans. W. E. GOHLMAN, Albany 1974, pp. 62–65. On Ibn Sīnā’s life see D. GUTAS’s article in Encyclopaedia Iranica, vol. 3, pp. 67–70.

200

FRANK GRIFFEL

In his biography of Avicenna, al-Jūzjānī reports that the Eminent Master wrote a book with the title Fair Judgment (al-Inṣāf ), “but on the day when Sultan Mas῾ūd arrived at Isfahan, his troops rifled the Master’s baggage which contained the work and it was not found afterwards.”19 The letter to Bahmanyār, which became part of the book al-Mubāḥathāt, refers to this event. Here Avicenna responds to his companion’s concerns about the loss of books. These concerns likely have been voiced in an earlier letter from Bahmanyār, which is lost. Avicenna’s response begins: You expressed sorrow at the loss of Pointers and Reminders (al-Ishārat wa-l-tanbīhāt), but I think that a copy of this book has been preserved. As for the Eastern Questions (al-Masā᾿il al-mashriqiyya) I had packed them up – no, most of them – [as they were] in their quires so as no one would look at them, and I had also written down on slips of paper some passages from The Throne Philosophy (al-Hikma al-῾arshiyya); it is these which got lost. They were not voluminous, however, although they did treat many subjects and were quite comprehensive, and it would be easy to rewrite them. But yes, the Fair Judgment (al-Inṣāf ), on the other hand, could not but be extensive, and rewriting it would require work. But then, who is going to rewrite it (…)?20

The books that were lost in Muḥarram 421 / January 1030 in Isfahan were four in number: al-Ishārāt wa-l-tanbīhāt, al-Masā᾿il almashriqiyya, al-Ḥikma al-῾arshiyya, and al-Inṣāf. Avicenna informs his student of his believe that a copy of al-Ishārāt wa-l-tanbīhāt is preserved (maḥfūẓ), whereas he has no such conviction about the other three. The book al-Ḥikma al-῾arshiyya may have been a work in progress. Avicenna writes that he “put down some of [it]… on slips of paper ” (athbattu ashyā᾿ min al-Hikma al-῾arshiyya fī juzāzāt). He continues: “these are the ones that got lost” (fa-hādhihi hiya allatī ḍā῾at) but also says that “these were not voluminous” (illā annahā lam takun kathīrat al-ḥajm). There is a certain ambiguity as to what “these” refers to. As a pronoun of the close deixis, it certainly refers to the “slips of paper” (juzāzāt) mentioned just in the phrase before and hence to al-Hikma al-῾arshiyya. The deixis, however, may also extend and include 19. IBN SĪNĀ, The Life of Avicenna, pp. 80–81. 20. IBN SĪNĀ, al-Mubāḥathāt, pp. 49–50 (§§ 33–35); Engl. trans. from GUTAS Avicenna and the Aristotelian Tradition (2nd. ed.), pp. 50–51, who makes emendations to Bīdārfar’s text based on his own study of manuscripts. Cf. also the translation in REISMAN, Making of the Avicennan Tradition, pp. 207–208.

ON THE AUTHENTICITY OF THE THRONE EPISTLE

201

al-Masā᾿il al-mashriqiyya, mentioned a sentence earlier. If so, this book was also lost. Finally, Avicenna adds a further description of what the slips of paper of al-Ḥikma al-῾arshiyya contained: “they [covered] much content and [were] very comprehensive” (kānat kathīrat al-ma῾nā kulliyya jiddan). Note the opposition between negating kathīrat al-ḥajm and affirming kathīrat al-ma῾nā. While al-Ḥikma al-῾arshiyya had a lot of ma῾nā, which we here translate as “content,” it was small in size (ḥajm). As for being “comprehensive” (kulliyya), the reader would assume that, like many of Avicenna’s other books, it would cover the three major fields of his philosophy, namely logic, the natural sciences, and metaphysics. The title al-Ḥikma al-῾arshiyya likely derives from passages in the Qur’ān that associate God with a “throne” (῾arsh) and other furniture like a “pedestal” or “footstool” (kursī). In a number of verses God is described as sitting “upright on the throne.”21 The understanding of these verses has always been a contested issue between rationalist movements in Islam (such as Mu῾tazilism, Ash῾arism, and falsafa), who rejected the idea that God could be spatially extended and “sit” or be “upright,” and between scripturalist movements (such as Ḥanbalites or Karrāmites) who somehow acknowledged a literal truth of these descriptions. In philosophical literature these verses were subject to allegorical interpretation (ta᾿wīl) and the “throne” was together with “the tablet” (al-lawḥ) and the “pedestal,” or “footstool” (al-kursī) often understood as references to the highest creations in the celestial universe.22 Philosophers most often identified the “throne” with the most remote celestial sphere in the Aristotelian-Ptolemaic understanding of a geocentric universe, in Arabic the “sphere of the spheres” (falak al-aflāk), or primum mobile in the Latin tradition.23 The “footstool,” in turn, was the next to highest sphere. Whether philosophers agreed with these identifications or not, using the word “throne” evokes the idea that one is talking about the most elevated realms of the cosmos, those parts of the universe that are closest to the First Principle or God. A “philosophy related to the throne” (ḥikma ῾arshiyya) 21. istawā ῾alā l-῾arsh; Q 7: 54, 10: 3, 13: 2, 20: 5, 25: 59, 32: 4, 57: 4. 22. E. WAKELNIG, Feder, Tafel, Mensch. Al–῾Āmiris Kitāb Fuṣūl al-Ma῾ālim al-ilāhīya und die arabische Proklos Rezeption im 10 Jh., Leiden 2006, pp. 84–85. 23. Ibid., pp. 86–87, 159–160, 167, 392.

202

FRANK GRIFFEL

would be one that talks either about cosmology or about the cause of that cosmology, namely God. In his study of al-Mubāḥathāt, Reisman observed that the word “related to the throne” (῾arshī) was, like “eastern” (sharqī) used by Avicenna’s students in a way that suggests “that these [words] were already a part of Ibn Sīnā’s philosophical vocabulary.”24 From comments made by Bahmanyār that were collected in al-Mubāḥathāt, Reisman concludes that he contrasts a “throne demonstration or an eastern one” with those that Avicenna presents in his al-Shifā᾿.25 In response to Bahmanyār’s suggestion that a “throne demonstration” might provide a better explanation why the human soul is not imprinted on the body, Avicenna reiterates in al-Mubāḥathāt the arguments he wrote down in al-Shifā᾿. Should he find any better ones in the future, Avicenna adds, he shall write them down in al-Ḥikma al-῾arshiyya.26 This exchange tells us that for Avicenna and his students, “related to the throne” (῾arshī) held certain methodological connotations or ones related to how philosophy is presented, similar to how “eastern” (mashriqī) is used in the introduction of al-Mashriqiyyūn (The Easterners), one of the few extant parts of this otherwise lost book. There, the mashriqī way is contrasted with how Avicenna presents his philosophy in his most popular works, of which al-Shifā᾿ would be the prime example.27 The exchange documented in al-Mubāḥathāt also reveals that for Avicenna, teachings on the human soul and a proof for the latter’s substantiality and its ability to exist without connection to the body was one of the subjects to be treated in al-Ḥikma al-῾arshiyya. To repeat writing al-Ḥikma al-῾arshiyya would be easy, Avicenna says in his letter to Bahmanyār, but who shall do that? There is no information in any of Avicenna’s works as to whether he or someone 24. REISMAN, Making of the Avicennan Tradition, p. 229. 25. burhān ῾arshī aw sharqī; IBN SĪNĀ, al-Mubāḥathāt, p. 121, § 298; REISMAN, Making of the Avicennan Tradition, pp. 229–230. 26. IBN SĪNĀ, al-Mubāḥathāt, p. 121, § 299; REISMAN, Making of the Avicennan Tradition, pp. 229–230. 27. On the complex use of mashriqī in al-Mashriqiyyūn, see GRIFFEL, Formation of Post-Classical Philosophy, pp. 103–104. The sense of “῾arshī” as a higher kind of philosophy is also alluded in the title of Yaḥyā al-Suhrawardī’s al-Talwīḥāt al-lawḥiyya wa-l-῾arshiyya, where on several occasions “῾arshiyya” is used as a qualifier of a certain kind of argument or “point” (nuqṭa), see e. g. AL-SUHRAWARDĪ, al-Talwīḥāt al-lawḥiyya wa-l-῾arshiyya, ed. N. ḤABĪBĪ, Tehran 1388/2009, pp. 188. 15, 219. 11, 229. 7.

ON THE AUTHENTICITY OF THE THRONE EPISTLE

203

else did or did not rewrite this lost book. The title does not appear in the text of Avicenna’s autobiography and not in its continuation by al-Jūzjānī.28 It does, however, appear within the list of Avicenna’s works compiled by Ibn Funduq al-Bayhaqī, Avicenna’s earliest biographer after al-Jūzjānī. Ibn Funduq wrote his Tatimmat Ṣiwān al-ḥikma in the middle of the sixth/twelfth century, more than a hundred years after Avicenna’s death and he includes what has become known as the Shorter Bibliography of Avicenna’s works.29 The list includes a text titled The Sacred Philosophy (al-Ḥikma al-qudsiyya) as no. 18 and described it as a full book (kitāb) – as opposed to an epistle (risāla) – of a single volume length. This title is, according to Gutas, “manifestly a scribal corruption or misreading of ῾aršiyya as qudsiyya in unpointed Arabic.”30 Indeed, these two words are written in a very similar rasm and given that there is no other independent information for al-Ḥikma al-qudsiyya among Avicenna’s works, one should read it as al-Ḥikma al-῾arshiyya. The misreading and resulting confusion might be the result of Ibn Funduq al-Bayhaqī’s own information that the autograph of al-Ḥikma al-῾arshiyya got lost. By his time there was a considerable interest in the Avicennan corpus that also included concern about works of Avicenna that might have been lost. Al-Bayhaqī knew Avicenna’s al-Mubāḥathāt and he was keenly aware of the passage from the letter to Bahmanyār that we just looked at.31 In his biography of Avicenna, al-Bayhaqī gives a report of the several campaigns of warfare between the Ghaznawids and Alā᾿ al-Dawla Ibn Kākūya, Avicenna’s patron in Isfahan. After telling the story of the last campaign 424/1033–34, the one conducted by Abū Sahl al-Ḥamdūnī, he continues: The governor Abū Sahl together with a group of Kurds plundered the possessions of the Master and among them were his books. There are only parts of The Fair Judgment (Kitāb al-Insāf ) left. ῾Azīz al-Dīn 28. IBN SĪNĀ, The Life of Avicenna, pp. 16–89. The works mentioned there are listed in GUTAS, Avicenna and the Aristotelian Tradition (2nd. ed.), pp. 401–410. 29. AL-BAYHAQĪ, Tatimma Ṣiwān al-ḥikma of ῾Alī b. Zaid al-Baihaqī. Fasc. I-Arabic text, ed. M. SHAFĪ῾, Lahore 1935, pp. 46–47. The list is analyzed in GUTAS, Avicenna and the Aristotelian Tradition (2nd. ed.), pp. 401–410. The latest historical date mentioned in Tatimmat Ṣiwān is 553/1158–69, see GRIFFEL, Formation of Post-Classical Philosophy, p. 166. 30. GUTAS, Avicenna and the Aristotelian Tradition (2nd. ed.), p. 485. 31. On al-Bayhaqī’s knowledge of al-Mubāḥathāt (which he refers to as al-Mabāhith) see REISMAN, Making of the Avicennan Tradition, pp. 106, 120, 209.

204

FRANK GRIFFEL

al-Fuqqā῾ī al-Rayḥanī claimed: “In the year 545 [1150–51] I bought a copy of this book in Isfahan and brought it to Merv.” God knows best! As for the Eastern Philosophy (al-Ḥikma al-mashriqiyya) in its entirety and the Throne Philosophy (al-Ḥikma al-῾arshiyya), the Imām Ismā῾īl al-Bākharzī says that they were in the library of Sultan Mas῾ūd ibn Maḥmūd in Ghazna until al-Ḥusayn, who is the king of the Ghur, together with the army of Ghūr and some ghuzz burned it down in the year 546 [1151–52].32

We have already said that al-Bayhaqī got the occasion wrong when these books were plundered. It happened four years earlier, when Sultan Mas῾ūd ibn Maḥmūd himself and not his governor Abū Sahl al-Ḥamdūnī conquered Isfahan. About the fate of the books, Ibn Funduq adds that some parts of al-Inṣāf could have been retrieved and those may be in the ῾Azīziyya library in Merv, founded by ῾Azīz al-Dīn al-Fuqqā῾ī al-Rayḥanī (or: Zanjānī).33 Al-Bayhaqī suggests that the autograph copies of two other books, al-Ḥikma al-mashriqiyya and al-Ḥikma al-῾arshiyya, got lost. After stealing them from Avicenna’s possessions, Ghaznawid troops took them to Ghazna where they were in the royal library until it burned. The historians Ibn al-Athīr as well as Ṣadr al-Dīn al-Ḥusaynī, who both wrote seventy years after al-Bayhaqī, repeat this information. In both cases we must assume that their source is al-Bayhaqī.34 During the more than one century when these books were in a library in Ghazna, however, someone could have taken copies from the autographs. In fact, about thirty to fifty years before al-Bayhaqī wrote these lines a manuscript was copied most probably in Merv that includes parts of al-Ḥikma al-mashriqiyya and al-Inṣāf. This is the 32. AL-BAYHAQĪ, Tatimmat Ṣiwān, p. 56. Cf. the English trans. in GUTAS, Avicenna and the Aristotelian Tradition (2nd ed.), p. 122, and GRIFFEL, Formation of Post-Classical Philosophy, pp. 42, 56. 33. The library is mentioned in YĀQŪT, Jacut’s Geographisches Wörterbuch, ed. F. WÜSTENFELD, 6 vols., Leipzig 1866–73, vol. 4, p. 509. 15. Cf. M. SHĀFI῾ in his edition of al-Bayhaqī, Tatimmat Ṣiwān al-ḥikma, p. 194, and GUTAS, Avicenna and the Aristotelian Tradition (2nd ed.), p. 122. 34. IBN AL-ATHĪR, al-Kāmil fī l-ta᾿rīkh, ed. C. J. TORNBERG, Chronicon quod perfectissimum inscribitur, 14 vols., Leiden 1851–1876, vol. 8, pp. 211–212; AL-ḤUSAYNĪ AL-SARAKHSĪ, Zubdat al-tawārīkh. Akhbār al-umarā᾿ wa-l-mulūk al-saljūqiyya, ed. M. NŪR AL-DĪN, Beirut 1405/1985, p. 35. See the English translation by C. E. BOSWORTH, The History of the Seljuq State: A Translation With Commentary of the ‘Akhbār al-dawla al-saljūqiyya, London / New York 2011, p. 12. See also GUTAS, Avicenna and the Aristotelian Tradition (2nd ed.), pp. 122–123.

ON THE AUTHENTICITY OF THE THRONE EPISTLE

205

so-called Ṣighnākhī-Codex, now MS ḥikma 6M at the Egyptian National Library (Dār al-Kutub) in Cairo. ῾Abd al-Razzāq al-Ṣighnākhī, who was a student of al-Lawkarī (d. after 503/1109) in Merv and who was also involved in disputations with al-Īlāqī (d. 536/1141), copied the codex either around the turn of the sixth/twelfth century or during the two decades after. The Ṣighnākhī-codex is one of the most precious manuscripts of works by Avicenna and incudes, for instance, all extant parts of al-Inṣāf, a recension of al-Mubāḥathāt, a recension of al-Ta῾līqāt, Avicenna’s commentary on al-Uthūlūjiyā of pseudo-Aristotle, as well as the beginning part of al-Mashriqiyyūn.35 Other manuscripts preserve some parts of this latter book on the natural sciences.36 The full text of The Easterners, however, is not available. Al-Bayhaqī reports the loss of the autograph of al-Ḥikma al-῾arshiyya but he must have been aware that its burning need not mean that the text is irretrievably lost. Copies might have been made, which is why he most likely includes it in his list of Avicenna’s works without himself having ever seen a copy of the text. Very early on in the manuscript tradition, however, a well-meaning reader or scribe of al-Bayhaqī’s Tatimmat might have drawn the conclusion from his text that al-Ḥikma al-῾arshiyya got lost and no longer exists, hence reading al-῾arshiyya, which made no sense to him, as al-qudsiyya. A Lost Book Found Al-Bayhaqī included al-Ḥikma al-῾arshiyya in his list of Avicenna’s works because he had justified hope that someone might have copied the recently burned autograph in Ghazna and that such a copy might soon become available. After the word ῾arshiyya had changed into qudsiyya, however, The Throne Philosophy had disappeared from al-Bayhaqī’s lists of Avicenna’s works. It did not take long to reappear. Another 35. D. GUTAS, “Notes and Texts from Cairo Manuscripts, II: Texts from Avicenna’s Library in a Copy by ῾Abd-ar-Razzāq aṣ-Ṣiġnāḫī,” in: Manuscripts of the Middle East 2 (1987), pp. 8–17, at pp. 11–14; REISMAN, Making of the Avicennan Tradition, pp. 77–82. The texts in this manuscript have been copied extensively and provided the archetype for numerous editions of works by Ibn Sīnā produced mostly in Cairo. The most important edition of texts from this manuscript is ‘A. BADAWĪ’s Arisṭū ῾inda l-῾arab. Dirāsa wa-nuṣūṣ ghayr manshūra, 2nd ed., Kuwayt 1978, where the manuscript’s content is described on pp. 43–51. 36. GUTAS, Avicenna and the Aristotelian Tradition (2nd ed.), p. 128.

206

FRANK GRIFFEL

annotated list of the Eminent Master’s works is preserved in several manuscripts, the earliest of which was copied in 588/1192.37 It is known as Avicenna’s Longer Bibliography and includes 94 works as opposed to al-Bayhaqī’s 39. This list includes al-Ḥikma al-῾arshiyya but describes the work in different terms from what Avicenna wrote about it in his letter to Bahmanyār. Number 61 in this list is: “Throne Philosophy, which is an elevated (or: advanced) text on divine matters (or: metaphysics).”38 The Sacred Philosophy (al-Ḥikma al-qudsiyya) of al-Bayhaqī’s list does not appear in this Longer Bibliography. It reappears, however, in the next stage of bibliographical study on Avicenna, namely in the so-called Extended Bibliography found at the end of the tarjama on Avicenna in some manuscripts of al-Bayhaqī’s Tatimmat Ṣiwān al-ḥikma. The earliest of these manuscripts was copied in 639/1241.39 In this Extended Bibliography we find al-Ḥikma al-qudsiyya as no. 2 mentioned even before Avicenna’s al-Shifā᾿ and as no. 116, as the last title at the very end of this list, al-Ḥikma al-῾arshiyya. We do not know whether the author of the Extended Bibliography positively determined that all books listed therein existed. It seems rather that he compiled this list cumulatively, i.e. adding titles that one could find in other lists of this kind without truly verifying that those books circulated. The Extended Bibliography seems to be the result of this latter technique, which is why al-Ḥikma al-qudsiyya appears there alongside al-Ḥikma al-῾arshiyya, most probably on account of its appearance in al-Bayhaqī’s earlier list. The absence of al-Ḥikma al-qudsiyya from the so-called Longer Bibliography suggests that its author did not simply copy earlier work-lists uncritically. The fact that this author describes al-Hikma 37. MS Istanbul, Üniversitesi 4755, foll. 316a–317b. This Longer Bibliography is also in IBN ABĪ UṢAYBI῾A, ῾Uyūn al-anbā᾿ fī ṭabaqāt al-aṭibbā᾿, ed. A. MÜLLER, 2 vols., Cairo 1299/1882, vol. 2, pp. 18–20, and is edited by GOHLMAN in IBN SĪNĀ, The Life of Avicenna, pp. 90–113. It is analyzed there on pp. 147–152 and in GUTAS, Avicenna and the Aristotelian Tradition (2nd ed.), pp. 401–410. 38. al-Ḥikma al-῾arshiyya wa-huwa kalām murtafi῾ fī l-ilāhiyyāt; IBN SĪNĀ, The Life of Avicenna, p. 104; IBN ABĪ UṢAYBI῾A, ῾Uyūn al-anbā᾿, vol. 2, p. 19. 29. 39. MS Istanbul, Murad Molla 1431, foll. 99a–100a; on the manuscript’s dating see F. GRIFFEL, “On the Character, Content, and Authorship of the Itmām Tatimmat Ṣiwān al-ḥikma and the Identity of the Author of the Muntakhab Ṣiwān al-ḥikma,” in: Journal of the American Oriental Society 133 (2013), pp. 1–20, at p. 3, n. 11. The Extended Bibliography is edited by M. Shāfi῾ in AL-BAYHAQĪ, Tatimmat Ṣiwān, pp. 187–190. The list is analyzed in GUTAS, Avicenna and the Aristotelian Tradition (2nd ed.), pp. 401–410.

ON THE AUTHENTICITY OF THE THRONE EPISTLE

207

al-῾arshiyya as an “elevated text” on metaphysics and/or divine matters (ilāhiyyāt) suggests that he truly knew this text or received this information from another reader who knew it. No earlier bibliographer had described the book as such. For al-Bayhaqī it was simply “a single volume book,” a vague description he might have come up with based on Avicenna’s own characterization in his letter to Bahmanyār. There, al-Ḥikma al-῾arshiyya is a short but substantial and comprehensive work. Al-Bayhaqī likely never saw a copy of al-Ḥikma al-῾arshiyya. The description in the Longer Bibliography, however, can hardly be based on Avicenna’s words or on that of al-Bayhaqī. Of all we know, the author of the Longer Bibliography, who compiled it before 588/1192, was the first scholar who, after a possible destruction of the autograph in a fire in Ghazna in 546/1151, confirmed the existence of al-Ḥikma al-῾arshiyya, and describes it. In fact, among the many bibliographers of Avicenna’s œuvre, the author of the Longer Bibliography was the first who saw the book or received reliable information on its existence. But was this al-Ḥikma al-῾arshiyya the same text that Avicenna had lost in 421/1030? Today we have printed versions of a text most often titled al-Risāla al-῾arshiyya that is in its manuscripts unanimously ascribed to Avicenna. There are more than fifty manuscripts of this work, the great majority of them in libraries in Iran and in Turkey.40 Although the great majority of them have the title al-Risāla al-῾arshiyya, some are titled al-Ḥikma al-῾arshiyya.41 In fact, given that this is the only extant text by Avicenna with the word “῾arshiyya” in its title, it has always been equated with al-Ḥikma al-῾arshiyya, mentioned in al-Mubāḥathāt and in al-Bayhaqī’s biography of Avicenna. Al-Risāla al-῾arshiyya was first 40. Y. MAHDAVĪ, Fihrist-i nuskhathā-yi muṣannafāt-i Ibn Sīnā, Tehran 1333/1954, pp. 75–76 (no. 61). See also the list of manuscripts in GUTAS, Avicenna and the Aristotelian Tradition (2nd ed.), p. 485. These two lists do not include the majority of manuscripts in libraries in Iran. The union catalogue of manuscripts in Iran lists more than thirty. See M. DIRĀYATĪ, Fihristvārah-yi dastneveshthā-yi Īrān, 12 vols., Tehran 1389/2010, vol. 4, pp. 724–725. See also the list of fourteen MSS in libraries in Turkey in A. R. KARABULUT, İstanbul ve Anadolu Kütüphanelerinde Mevcut El Yazmasi Eserler Ansiklopedisi, 3 vols., Kayseri, s.d. [2005], vol. 1, p. 451. Cf. also C. BROCKELMANN, Geschichte der arabischen Litteratur. Supplementbände, 3 vols., Leiden 1937–42, vol. 1, pp. 814 (no. 9c) and 820 (no. 68c). 41. See e.g. MS Tehran, Kitābkhāneh-yi Majlis-i Shūrā-yi Islāmī, no. 14473; cf. IBN YŪSUF SHĪRĀZĪ e.a., Fihrist-i Kitābkhāneh-yi Majlis-i Shūrā-yi Millī (later: Fihrist-i asnād-i Kitābkhāneh-yi Majlis-i Shūrā-yi Islāmī), Tehran 1353– / 1974–, vol. 38, p. 626.

208

FRANK GRIFFEL

edited in 1934–35 in Hyderabad, India, within a collection of works by Avicenna and a second time around 1981 in Cairo.42 The text of these publications is rather short and does not exceed two dozens pages in print.43 Highly knowledgeable modern bibliographers of Avicenna, such as Yaḥyā Mahdavī and George C. Anawati, understood al-Risāla al-῾arshiyya to be authentic and to be the same as al-Ḥikma al-῾arshiyya.44 Only twenty years ago, in 2002, David C. Reisman first raised doubts about this equation and wrote that he is “not convinced that [al-Risāla al-῾arshiyya] belongs to the authentic Ibn Sīnā corpus for both style and content suggest a post-Ibn al-῾Arabī environment.”45 Robert Wisnovsky reports Reisman’s reservations in an article that appeared a year later (and that was edited by Reisman). In private conversation, Wisnovsky writes, Reisman alerted him “to pieces of terminological, conceptual, and grammatical evidence which cause him to suspect that the Risāla al-῾aršiyya, which we have before us today (…) is not Avicenna’s al-Ḥikma al-῾aršiyya but instead a later work, probably of the Ishrāqī tradition.” Wisnovsky found those pieces of evidence 42. IBN SĪNĀ, al-Risāla al-῾arshiyya fī tawḥīdihi wa-ṣifātihi, text no. 4 in Majmū῾ rasā᾿il al-Shaykh al-Ra᾿īs, Hyderabad 1353–54/1934–36; IBN SĪNĀ, al-Risāla al-῾arshiyya fī ḥaqā᾿iq al-tawḥīd wa-iṭhbāt al-nubuwwa, ed. I. HILĀL, s.l. (Cairo), s.d. (~1981). In the khatima to the Hyderabad print (p. 19) Zayn al-῾Ābidīn al-Mūsāwī informs his readers that he established the text together with ῾Abdallāh ibn Aḥmad al-῾Alawī based on two manuscripts, MS Rampur, Ḥikma 82 and one from the Āṣafiya Library in Hyderabad. The second edition by Hilāl is based on the first together with what the editor describes as MS Tehran, Kitābkhāneh-yi Majlis-i Shūrā-yi Islāmī, 7851. I was unable to verify that there is indeed a manuscript of al-Risāla al-῾arshiyya under this number. 43. In 1951 and 2009 Arthur J. Arberry and Olga Lizzini translated parts of this text into English and Italian and in 1980 Egbert Meyer produced an annotated German paraphrase with analytical commentary that is still one of the most lucid introductions into Avicenna’s theology. See A. L. ARBERRY, Avicenna on Theology, London 1951, pp. 25–37; O. LIZZINI, “IV. L’Epistola del Trono (al-Risāla al-῾arshiyya),” in: G. AGAMBEN – E. COCCIA (eds.), Angeli. Ebraismo, Cristianesimo, Islam, Vicenza 2009, pp. 1889–1893; E. MEYER, “Philosophischer Gottesglaube: Ibn Sīnā’s Thronschrift,” in: Zeitschrift der deutschen morgenländischen Gesellschaft 130 (1980), pp. 226–278. See also the full English translation in A. K. TUFT, “The Attributes of God. A Translation of al-Risāla al-῾Arshīyah of Ibn Sīnā,” MA thesis, University of Washington, Seattle, 1972, pp. 11-37. See references to further part-translations and translations in other languages in J. JANSSENS, An Annotated Bibliography on Ibn Sīnā. Second Supplement (1995–2009), Tempe, Ariz. 2017, pp. 109– 110. 44. MAHDAVĪ, Fihrist-i nuskhathā-yi muṣannafāt, pp. 75–76 (no. 61); G. C. ANAWATI, Mu᾿allafāt Ibn Sīnā, Cairo 1950, pp. 239, 241–242 (nos. 179, 183). 45. REISMAN, Making of the Avicennan Tradition, p. 21, n. 2.

ON THE AUTHENTICITY OF THE THRONE EPISTLE

209

compelling and himself began to doubt the authenticity of the text in question. Yet he cautiously maintained its ascription to Avicenna based on the fact that on the issue that he was inquiring in his particular article – Avicenna’s distinction between immanent and transcendent causes – the text does not diverge from others by the Eminent Master.46 Finally, in 2014, Dimitri Gutas in his inventory of Avicenna’s works lists this text among those that are, in his opinion, possible pseudo-epigraphies, which would need to be authenticated. He describes the question about this text as threefold: “[W]hether the ῾Aršiyya that Avicenna mentioned was completed or rewritten, whether the extant treatise (…) is by Avicenna, and whether the two are identical.”47 3. Is the Extant ῾Arshiyya By Avicenna? Here I would like to address the last two of Gutas’ three questions. I believe there is sufficient evidence to conclude that the ῾Arshiyya in the two printed editions is not by Avicenna and that it is therefore not identical with the work mentioned in the earliest sources on his life and his works. The argument that I make is twofold: (1) the work uses a technical term that refers to Ash῾arite theology and that Avicenna was most likely not familiar with and (2) the structure of the book suggests that it was written in response to Ash῾arite theology, yet no other work by Avicenna shows any signs that he was concerned with or even informed about this school of thought. Mahdavī’s and Anawati’s bibliographies as well as the Union Catalogue of manuscripts in Iran report two prominent parallel titles of al-Risāla al-῾arshiyya. These are (1) Ḥaqā᾿iq ῾ilm al-tawḥīd (“The True Meanings of Knowing [or: Understanding] God’s Unity”) and (2) Ma῾rifat Allāh wa-ṣifātihi wa-af῾ālihi “ (“On Knowing God, His Attributes, and His Actions”).48 Both titles are much more precise descriptions of the printed text’s subject matter. In fact, both of 46. R. WISNOVSKY, “Towards a History of Avicenna’s Distinction between Immanent and Transcendent Causes,” in: D C. REISMAN with A. AL-RAHIM (eds.), Before and After Avicenna. Proceedings of the First Conference of the Avicenna Study Group, Leiden 2003, pp. 49–68, at pp. 65, n. 32 and 67, n. 35. 47. GUTAS, Avicenna and the Aristotelian Tradition (2nd ed.), p. 485. 48. See the bibliography details in note 37.

210

FRANK GRIFFEL

these titles appear right at the beginning of the text, almost as its incipit: One of those with whom I am affiliated asked me to compose for him an epistle that includes the true meanings of knowing (or: understanding) God’s unity (ḥaqā᾿iq ῾ilm al-tawḥīd ) in the way that one has to believe in God, His attributes, and His actions, [and by doing so that I] avoid [slipping to] the side of emulating others but rather be inclined [to produce] pure verification on the length of an ikhtiṣār. I respond to him and his request asking our Lord, God – exalted is He – for help. This is the short epistle (al-risāla al-wajīza) [he asked for] and it includes three roots: the first root is on the proof of the Being Necessary (By Virtue of Itself), the second root is on His simplicity (waḥdāniyya), the third root on the denial of (any) cause for Him.49

Pre-modern Muslim authors often described the length of a work in three categories, shorter ones (wajīz, mukhtaṣar), mid-size ones (wasīṭ, talkhīṣ, mulakhkhaṣ), or long and comprehensive ones (basīṭ, shāmil). The author of al-Risāla al-῾arshiyya clarifies that this epistle falls into the first category, the shortest one. The three roots or foundations (sing. aṣl) described in the second paragraph cover roughly a third of the epistle’s text. Once the author has completed his presentation of the third root, the text continues with “teachings on the attributes according to how we obtain them based on these well-established roots” (al-qawl fī l-ṣifāt ῾alā l-wajh alladhī talaqqaynāhu min hādhihi l-uṣūl al-mumahhaḍa).50 This midpart of the epistle discusses seven different attributes. Finally, the epistle ends with “teachings on [how] actions proceed (or: emanate) from Him” (al-qawl fī ṣudūr al-af῾āl ῾anhu),51 and a final (sub-)chapter titled “teachings on His decree and determination” (al-qawl fī qaḍā᾿ihi wa-qadarihi).52 All the teachings presented in this short epistle are thoroughly Avicennan. If they were not and differences between the teachings in this text and others by the Eminent Master became visible, the text would have hardly enjoyed its status of an authentic work by Avicenna for so long. The proof for the existence of a Being Necessary By Virtue 49. al-Risāla al-῾arshiyya, p. 2. 3–7 / p. 15. 4–11. Pages referenced to this text are first to the Hyderabad edition and then to that of Hilāl, divided by a slash. 50. Ibid., p. 7. 11–12 / p. 23. 14. 51. Ibid., p. 15. 5 / p. 37. 11. 52. Ibid., p. 16. 3 / p. 38. 1.

ON THE AUTHENTICITY OF THE THRONE EPISTLE

211

of Itself, for instance, uses the same strategy and the same language as Avicenna uses, for instance, in his al-Najāt.53 For all teachings in this epistle one can find passages in authentic works by Avicenna that teach the same. My argument against the work’s authenticity is not based on any doctrinal differences or divergences of this text with other, authentic texts by Avicenna. I believe that the author was a thoroughly convinced Avicennan who was intimately familiar with the teachings of Avicenna. The text we are looking at presents to us — in a succinct way — Avicenna’s teaching on God, hence his theology. Avicenna followed Plotinus in his argument that the First Cause of all being must be completely simple. Plotinus’ argument for the doctrine of divine simplicity goes like this: Given that an infinite regress of causes is impossible, there must be a first principle (= starting point), which is the first (or: ultimate) cause of all things. If the first principle of all things were composed of parts, then there must be a reason, meaning a cause, for the composition itself and that reason (or cause) would be ontologically prior to it. In that case, however, the assumed first principle would not be the first principle. Therefore, the first principle cannot be composed of parts, but must be absolutely simple.54 Or, in a different way of saying: if God were divided into different parts just like any other being, then there would need to be a reason for this division and that reason would be the First Principle. Hence God would not be the first cause and not the First Principle, meaning He would not be God. The Avicennan author of al-Risāla al-῾arshiyya expresses this argument in the second root or foundation of the epistle’s first part, where he says that the Being Necessary By Virtue of Itself cannot be “two in any way” (ithnān bi-wajh min al-wujūh). This follows apodictically from Avicenna’s prove for the existence of a Being Necessary By Virtue of Itself. If there were two beings of this kind, or if that being were divided into something that can be understood as twofold, then such a division is either on the level of accidents or on the level of essences 53. IBN SĪNĀ, al-Najāt min al-gharq fī baḥr al-ḍalālāt, ed. M. T. DĀNISHPAZHŪH, Tehran 1364/1985, pp. 546–553. 54. PLOTINUS, Enneads II. 9 .1, VI. 9. 3. The argument is already hinted at in PLATO, Parmenides 137c. IBN SĪNĀ, al-Shifā᾿, al-Ilāhiyyāt, pp. 271–273; ID., al-Najāt, pp. 227– 228 / pp. 551–553; GRIFFEL, Formation of Post-Classical Philosophy, p. 393.

212

FRANK GRIFFEL

(sing. dhāt). In the first case it would mean that two accidents exist and no accident is capable of existing by itself but it depends on a substrate and an essence. Hence none of them can be necessary by virtue of itself and both must be caused (ma῾lūl). If the division is on the level of essences that means that both of them are composed (murakkab). The author neglects to explain this step of the argument which we may add here: any two things of the same kind must have something in common and something that distinguishes them, otherwise they would not be two different things. Hence they are composed of common elements and different ones. “What is composed is caused,” the argument in this text continues, “hence none of the two is a being necessary [by virtue of itself].”55 Therefore, the true Being Necessary by Virtue of Itself cannot have accidents or be divided on the level of essence. This section concludes: If through [all] this it has been established that the Being Necessary [by Virtue of Itself] cannot be two, but is rather the All-Truth (kull ḥaqq), then by virtue of its core essence (al-ḥaqīqa al-dhātiyya), through which He is a truth, He is one united [being] (mutaffiq wāḥid), and nothing else shares with Him in that unity. How does the All-Truth obtain its existence other than through Himself?56

The argument that existence is at the core of this being leads to the conclusion that God must be only existence. If He is radically simple, then His “core essence” (al-ḥaqīqa al-dhātiyya) and existence (wujūd) must be one and the same. That is expressed in the next root or foundation on the denial of causes from God. This is by far the longest of the three roots and of all sub-divisions in this text – stretching over a fifth of it overall – and it includes a thorough rejection of the view that God could have an essence or a quiddity (māhiyya) that is distinct from His existence.57 The author of the epistle is careful not to leave any misunderstandings here. He explains the four Aristotelian causes and which of them are immanent and transcendent.58 The text 55. al-Risāla al-῾arshiyya, p. 3. 13 / p. 17. 8. 56. Ibid., p. 3. 16–18 / p. 17. 14–16; cf. ARBERRY, Avicenna on Theology, p. 26; TUFTS, “Attributes of God,” pp. 12–13; MEYER, “Philosophischer Gottesglaube,” p. 236. 57. al-Risāla al-῾arshiyya, pp. 3. 19 – 7. 10 / pp. 17. 7 – 23. 13. 58. MEYER, “Philosophischer Gottesglaube,” p. 237, remarks that usually discussions of the four causes are part of physics and that this kind of treatment in metaphysics is unique for Avicenna.

ON THE AUTHENTICITY OF THE THRONE EPISTLE

213

zooms in on the efficient cause (῾illa fā῾iliyya) and after establishing that the Being Necessary By Virtue of Itself cannot have such a cause it continues: Since it is established that He has no efficient cause, it follows from this line of reasoning (i῾tibār) that His quiddity cannot be something other than His anniyya, meaning His existence, and that He is neither a substance nor an accident, nor can He be two (ithnān) so that each one of them receives its existence from the another (…).59

This last point leads straight into an “explanation” (bayān) why God’s essence or quiddity must be the same as His existence. The author argues particularly against the view that the essence or quiddity could be the cause of its own existence.60 Two other “explanations” follow, clarifying why God cannot be an accident and not a substance.61 The text, however, continues to be concerned with the question why something cannot be the cause of its own existence and concludes: “Therefore the Being Necessary [By Virtue of Itself] is simple and does not receive existence from anybody; It is the necessary of existence in all aspects and does not receive existence from something else.”62 Further explanations about why God does not have a material cause or why there is nothing contingent connected to it follow until the author at one point picks up the style of a kalām argument (in qīla… fa…) and focuses on one particular position, namely the view that God must have attributes that are constituent to His essence: If it is said that [any one of] His attribute[s] is not additional to the essence but rather inside it (dākhil) as a constituent of the essence and that the essence in its existence cannot be conceived without those attributes, then the essence is composed and through that [Its] simplicity is destroyed.63 59. al-Risāla al-῾arshiyya, p. 4. 8–10 / p. 18. 16–18; cf. ARBERRY, Avicenna on Theology, p. 27; TUFTS, “Attributes of God,” p. 14. 60. al-Risāla al-῾arshiyya, p. 4. 16–22 / p. 19. 7–14. 61. On why God cannot be a substance see M. LEGENHAUSEN, “Ibn Sina’s Argument Against God’s Being a Substance,” in: C. KANZIAN – M. LEGENHAUSEN (eds), Substance and Attribute in Islamic Philosophy: Western and Islamic Traditions in Dialogue, Heusenstamm 2007, pp. 117–142. 62. al-Risāla al-῾arshiyya, p. 5. 9–10 / p. 20. 11–12. 63. al-Risāla al-῾arshiyya, p. 6. 4–5 / p. 21. 18–20. Cf. ARBERRY, Avicenna On Theology, p. 30; TUFTS, “Attributes of God,” p. 17. Reading tankharamu with the Hyderabad

214

FRANK GRIFFEL

This sentence, however, comes without much introduction and with no follow-up. Rather, the text continues with familiar Avicennan teachings (God cannot change and has no opposite, cannot not exist) before it moves to the proofs why God does not have formal and final causes. God cannot pursue a goal, as that would imply imperfection on His side and contingency. Rather God is pure good (jawād maḥḍ) and He is true perfection (kamāl ḥaqq). God does not approve of something or prefer something over something else, nor does He find something repugnant. This long sub-chapter on why God has no causes ends in what is in my opinion one of the most interesting passages in the epistle that betrays its inauthenticity. Like in the passage we just translated, the author argues against a group of opponents, without however, fully clarifying who that is. Here, the author argues against what he calls “this proposition” (hādhihi l-qaḍiya), without identifying it. The proposition is the view of Muslim theologians that God approves of some things and finds others repugnant. This leads the author of our text into discussing the Mu῾tazilite aṣlaḥ-theory: Because would [God] approve of something or find something repugnant, then whatever He approves of would exist and do so forever; whatever He finds repugnant would not exist and be void. This proposition is invalidated by [the fact that] things that exist change.64 This is so because whatever is simple in all its aspects does not approve of something and of its opposite. “Tending toward what is the best or what is better” (ri῾āyat al-aṣlaḥ wa-l-ṣalāḥ), as a group from the ṣifātiyya raves, is impossible for Him. Because, if it were obligatory to Him to do what is better, then gratitude would not be demanded toward Him, and neither praise. This, because He would simply be fulfilling what is His obligation to perform. He would be like someone in this immanent world (al-shāhid) whose religion is determined and who deserves nothing, rather his actions are from God, and this one gets what we will explain below [in the chapter on al-qaḍā᾿ wa-l-qadar].65

edition. Hilāl adds a footnote that this sentence in incomprehensible to him and serves no purpose. 64. I follow Meyer’s suggestion (“Philosophischer Gottesglaube,” p. 244) and understand ikhtilāf al-mawjūdāt as changes of contrarian attributes. 65. al-Risāla al-῾arshiyya, p. 7. 4–10 / p. 23. 5–13. Cf. ARBERRY, Avicenna on Theology, pp. 31–32; TUFTS, “Attributes of God,” pp. 18–19.

ON THE AUTHENTICITY OF THE THRONE EPISTLE

215

4. Why this Text is Not By Avicenna The last quoted paragraph is a polemical outburst against the Mu῾tazila’s aṣlaḥ-theory in the debate about theodicy. The Mu῾tazila teach that God has to act in a way that it will increase the benefit of His creation. In fact, He cannot do anything that would deliberately harm His creation. If, however, God has no preferences for what to create, as Avicenna teaches, then this position must be wrong and God’s actions may well include some that harm His creation. Such a polemical outburst against teachings of the Mu῾tazila is not out of character for Avicenna and would not be an indicator for the text’s inauthenticity. The use of the word “ṣifātiyya” within this paragraph, however, is. This word was not yet in use among the speakers of Arabic that Avicenna was in contact with, and if it existed there, Avicenna would not have used it in this way. The word “ṣifātiyya” can be translated as “attributionalist” and it is a term of art developed in Muslim heresiographic literature.66 It describes the Ash῾arite school, or to be more precise, any Kullābite theological group in Islam which teaches that the divine attributes are entities that subsist within the divine. In the 8th/14th century, when this term is firmly established, Ibn Taymiyya uses “attibutionalist mutakallimūn” (al-mutakallima al-ṣifātiyya) as an umbrella term for “Ibn Kullāb, al-Ash῾ārī, and Ibn Karrām, as well as their followers.”67 The term ṣifātiyya developed in the early 5th/11th century among Ash῾arites in Khorasan as a positive description of their own theology and in opposition to Mu῾tazilite teachings on the divine attributes. The latter taught that God has divine attributes, or rather can be characterized by divine names, such as “knowledgeable,” “living,” or “willing,” but all these names refer to God’s essence and not to entities (ma῾ānī) that can be distinguished from the essence. In contrast Kullābite theologians taught there are real attributes (ṣifāt) in God, whereas many other theological schools in Islam, most importantly Mu῾tazilites, denied this. 66. ARBERRY, Avicenna on Theology, pp. 31–32, translates “Qualitarians.” Tufts adopts this translation and adds in a footnote: “probably a general reference to all those groups (Ḥanbalites, Ash῾arites, anthropomorphists, etc.) who affirmed the attributes of God.” 67. IBN TAYMIYYA, al-Radd ῾alā l-manṭiqiyyīn, ed. ῾A. AL-KUTUBĪ, Bombay 1368/1949, p. 65. 16–17.

216

FRANK GRIFFEL

In heresiographic literature “ṣifātiyya” is first used by the Nishapurian Ash῾arite ῾Abd al-Qāhir al-Baghdādī (d. 429/1038) in his al-Farq bayna l-firaq.68 The name “ṣifātiyya” creates a theological identity for the school of al-Ash῾arī, which goes through its formative period at the turn of the 5th/11th century and the decades after. In the second half of that century the school will benefit greatly from official sponsorship by the Seljuq state. Before this sponsorship began in 455/1063 with the appointment of Niẓām al-Mulk (d. 458/1092) to the Seljuq vizierate, however, the Ash῾arite school was a small and overall insignificant local theological group in Khorasan. Avicenna takes no notice of it. For him, kalām and Mu῾tazilism are synonymous and there is no sign that he ever reacted in his works to an Ash῾arite position or arguments. Other than here in al-Risāla al-῾arshiyya, the term “ṣifātiyya” does not appear in works attributed to him, as far as I can see.69 The real Avicenna likely did not know this word. If he knew it, he would not have used it the way it is used here. In this text the Mu῾tazilites are called “a group from the ṣifātiyya” (jamā῾a min al-ṣifātiyya). How did this come about? In his comprehensive study of heresiographical and doxographical literature in Islam, Josef van Ess notes that while ῾Abd al-Qāhir al-Baghdādī first uses ṣifātiyya in the meaning of “attributionalists” and “Ash῾arites” in the mid-5th/11th century, the term has a rather slow career.70 It does not appear in al-Ghazālī, for instance, or any other major Ash῾arite authors until Tāj al-Dīn al-Shahrastānī (d. 548/1153) employs it prominently in the first half of the 6th/12th century, again in a heresiographical context. In his Book of Religions and Sects (Kitāb al-Milal wa-l-niḥal), whose first version was finished in 521/1127–28, the term ṣifātiyya takes an important role. For al-Shahrastānī it is an umbrella term for all those theological groups in Islam “that do not – like the Mu῾tazilites – deny the independence of the divine 68. AL-BAGHDĀDĪ, al-Farq bayna l-firaq, ed. M. M. ῾ABD AL-ḤAMĪD, Cairo s. d., p. 337. 7. See J. VAN ESS, Der Eine und das Andere. Beobachtungen an islamischen häresiographischen Texten, 2 vols., Berlin 2011, vol. 1, p. 710. 69. This result has been established by searching the digitalized texts in the database Ḥikmat-i islāmī of NoorSoft.org, Tehran. 70. The term is used earlier in other meanings. Al-Ḥakīm al-Samarqandī (d. 342/953), for instance, uses ṣifātiyya in the meaning of a group who argued that the Qur’ān is one of the attributes of God. See VAN ESS, Der Eine und das Andere, vol. 1, pp. 450–452.

ON THE AUTHENTICITY OF THE THRONE EPISTLE

217

attributes.”71 The opposite of the term ṣifātiyya is mu῾aṭṭila,72 which is best translated as “those who take (attributes) away from God.” These, in al-Shahrastānī’s view, are the Mu῾tazilites. The ṣifātiyya in al-Sahrastānī’s book include the four groups of (1) the salaf (“the earliest forefathers”), (2) the mushabbiha (“anthropomorphists”), (3) the Ash῾arites, and (4) the Karrāmites.73 The ṣifātiyya is one of the four principal divisions in Islam, together with Shi‘ites, Khārijites, and Mu῾tazilites. The Sunnīs are made up of several sub-groups of the ṣifātiyya. Not all of the ṣifātiyya are Sunnis, however, because the Karrāmites were not regarded as Sunnīs.74 All these groups are called ṣifātiyya because they subscribe to a theology that affirms (thabata) the divine attributes. Not all do this with kalām arguments and some go too far and liken God to His creatures. The most competent among them were, according to al-Shahrastānī, Ibn Kullāb (d. c. 241/855), al-Qalānisī (d. after 309/922), and al-Ḥārith al-Muḥāsibī (d. 243/ 857).75 This prominent use of ṣifātiyya in al-Shahrastānī’s al-Milal wa-lniḥal, establishes the term in Islamic theological and philosophical literature. After writing this way about Islamic theology, al-Shahrastānī’s book also includes an important second part on falsafa that led to its proliferation among members of the philosophical movement in Islam. Fakhr al-Dīn al-Rāzī (d. 606/1210), who refers to al-Shahrastānī’s al-Milal wa-l-niḥal in his œuvre,76 uses “ṣifātiyya” both as a term of classification in his own heresiography and his Persian catalogue of the sciences, (Jāmi῾ al-῾ulūm), as well as in his comprehensive textbook of philosophy al-Mabāḥith al-mashriqiyya.77 71. Ibid., vol. 2, p. 878. 72. AL-SHAHRASTĀNĪ, al-Milal wa-l-niḥal, ed. M. F. BADRĀN, 2 vols., Cairo 1370– 75 / 1951–55, vol. 1, p. 145; French translation D. GIMARET – G. MONNOT – J. JOLIVET, Livre des Religions et des Sectes, 2 vols., Louvain / Paris 1986, vol. 1, p. 309. 73. Ibid., vol. 1, pp. 145–193; French trans., vol. 1, pp. 308–361. 74. The Karrāmites only pretend to be Sunnīs, see ibid., vol. 1, p. 180; French trans., vol. 1, p. 347. 75. Ibid., vol. 1, p. 33; French trans., vol. 1, pp. 146–147. 76. AL-RĀZĪ, al-Munāẓarāt fī bilād mā warā᾿ al-nahr, in: F. KHOLEIF, A Study on Fakhr al-Dīn al-Rāzī and His Controversies in Transoxiana, Beirut 1966, pp. 39–40; ID., al-Riyāḍ al-mu᾿niqa [or: mūniqa] fī arā᾿ ahl al-῾ilm, ed. A. JUM῾A, Tunis 2004, pp. 43, 183. 77. AL-RĀZĪ, al-Riyāḍ al-mu᾿niqa, pp. 29. 8, 103–105; ID., Jāmi῾ al-῾ulūm, ed. ‘A. ĀL DAVŪD, Tehran 1382 / 2003, p. 455 .peanult. On the latter usage see VAN ESS, Der Eine und das Andere, vol. 2, p. 1062.

218

FRANK GRIFFEL

In the latter text, he contrasts Avicenna’s teachings on the divine knowledge with that of the ṣifātiyya – here in the unspecific meaning of Muslim theologians or mutakallimūn – and argues that the two are not very different.78 His younger contemporary Shihāb al-Dīn Yaḥyā al-Suhrawardī (d. c. 587/1192) uses “ṣifātiyya” in the unspecified meaning of “Ash῾arites” or “Muslim theologians who assert attributes of the essence” in his Ḥikmat al-ishrāq.79 Given that the word ṣifātiyya is not used much before al-Shahrastānī’s al-Milal wa-l-niḥal,80 a text that uses this word was likely written after this book became available around 521/1127. But why is the meaning in al-Risāla al-῾arshiyya different from that in al-Shahrastānī’s book? Here it is an umbrella term that includes the Mu῾tazila, yet elsewhere it is one that is consciously created against the Mu῾tazila and excludes them.81 The answer leads us back to al-Risāla al-῾arshiyya and to Avicenna’s theology. The latter teaches that God is complete unity and that He is existence. God’s existence is, together with His necessity, the only thing that can be said positively about Him. Avicenna denied that God had any positive attributes. The only thing one can say positively about Him is that He exists, that He is necessary, and that He has concomitants that follow from these two, such as being “the One,” and “Allāh.” All other predications are, so Avicenna, either negative or they describe relations.82 “Immaterial” is a negative predication (not material), for instance, and so is “completely simple” (= not multiple). God is not an accident and He is not depending on or connected to anything, He does not change, does not have a body, and so on. Like negative attributes, God also has relational ones. Most of the divine names that appear in the Qur’ān and in a famous ḥadīṭh about the ninety-nine names of God (“the All-Powerful,” the 78. AL-RĀZĪ, al-Mabāḥith al-mashriqiyya, vol. 2, p. 484. 10–13; see GRIFFEL, Formation of Post-Classical Philosophy, p. 407. 79. AL-SUHRAWARDĪ, The Philosophy of Illumination. A New Critical Edition of the Text of Hikmat al-Ishraq With English Translation, ed. and trans. J. WALBRIDGE – H. ZIAI, Provo 1999, p. 115. 80. VAN ESS, Der Eine und das Andere, vol. 2, p. 877 is explicit: “Der Begriff wird (…) vor seiner (scil al-Shahrastānī’s) Zeit kaum gebraucht.” 81. MEYER, “Philosophischer Gottesglaube,” p. 244, n. 63, already stumbled on this problem. 82. See M SEBTI – D. DE SMET, “Avicenna’s Philosophical Approach to the Qur’an in the Light of His Tafsīr Sūrat al-Ikhlāṣ,” in: Journal of Qur’anic Studies 11 (2009), pp. 134– 148.

ON THE AUTHENTICITY OF THE THRONE EPISTLE

219

“All-Seeing” etc.) describe, according to Avicenna, relations that God has with His creation. For Avicenna, God has no positive attributes at all. The Eminent Master indeed teaches a moderate form of what is called negative or apophatic theology. Muslim theologians took issue with such a negative theology. They asked how God can possibly be the originator of this highly complex world if He is only pure existence and pure necessity? Necessity and existence can only be understood as the necessity and existence of something. The fact that God is, first of all, “the giver of existence” (mūjid) and “the one that makes necessity” (mūjib) immediately leads to the question of what does He give existence to and what does He make necessary? Muslim theologians argue that “existence” is only a predicate and cannot be understood as an entity by itself. In that respect it is similar to “unity” which makes no sense to invoke unless it applies to something.83 In his Tahāfut al-falāsifa of 488/1095, al-Ghazālī makes these points within the eighth discussion, which is devoted to the refutation (ibṭāl) of Avicenna’s position that God is pure existence. For al-Ghazālī this is a negative theology that makes no positive statements about what God truly is and what He does. Denying that God has a quiddity – or saying that God’s quiddity is the same as His existence – means the denial of any core meaning (ḥaqīqa) that we can associate to the word “God.” What is left is the word “existence” as a meaningless combination of sounds (lafẓ) “that has no referent at all once it does not relate to a quiddity (māhiyya).”84 The Mu῾tazilites may have had disputes with Ash‘arites on the independence of the divine attributes and whether they subsist as entities (ma῾ānī) in God or not, but they did not deny positive attributes per se the way Avicenna did. They taught that the divine names are different qualifications that all apply to the divine essence, which itself is undivided.85 Mu῾taziltes believed that God is the perfect mode of attributes like “knowing” (῾alīm) or “just” (῾adl) that He ascribes 83. See the argument in AL-RĀZĪ, al-Mabāḥith al-mashriqiyya, vol. 1, p. 125. 13–15. 84. AL-GHAZĀLĪ, Tahāfut al-falāsifa, p. 118. 11–14. 85. On Mu῾tazilite attribute theories see J. VAN ESS, Theologie und Gesellschaft im 2. und 3. Jahrhundert Hidschra. Eine Geschichte des religiösen Denkens im frühen Islam, 6 vols., Berlin 1991–97, vol. 4, pp. 425–459; M. ALLARD, Le problème des attributs divins dans la doctrine d’al-Aš῾arī et de ses premiers grands disciples, Beirut 1965, pp. 113– 122, and R. M. FRANK, “The Divine Attributes According to the Teachings of Abū l-Hudhayl al-῾Allāf,” in: Le Muséon 82 (1969), pp. 451–506.

220

FRANK GRIFFEL

to Himself in the Qur’ān. The early Mu῾tazilite al-Naẓẓām (d. c. 225/840) taught, for instance, that “God is by Himself continuously knowing, living, powerful, hearing, seeing, and eternal, not through a [separate] knowledge, power, life, hearing, seeing, and eternity [that He has].”86 God is perfect unity and none of His attributes can be distinct from Him. His attributes are not entities (ma῾ānī) that inhere in Him like in the case of humans. Mu῾tazilites affirmed numerous attributes and hence in principle they subscribed to al-Ghazālī’s argument from the eighth discussion of his Tahāfut. Ibn al-Malāḥimī (d. 536/1141), who wrote a Mu῾tazilite refutation of Avicenna’s philosophical system that can be dated between the years 532/1137 and 536/1141, focuses in the sixth of twenty chapters on the divine attributes. First he reports Avicenna’s distinction between positive, negative, and relational attributes and then he zooms in on the falāsifa, “who reduce all His attributes – exalted is He – to His necessary existence.”87 The falāsifa understand omnipotent, knowledgeable, alive, and having a will either as expressions referring to God’s necessary existence or as relational attributes. They deny multiplicity in God and hence cannot allow any other positive predicate than existence and necessity – which in His case are one and the same. Ibn al-Malāḥimī counters these teachings with a five-page explanation “on what the Muslims teach on the proof of His attributes – exalted is He.”88 “Muslims” here stands for Mu῾tazilites, who teach that God has two types of attributes, those that result from His essence (dhāt) and those that result from His actions. The former apply always to Him, the latter only after the beginning of creation. Without going into the details of this question, we may simply conclude that Mu῾tazilites and Ash῾arites were in agreement that God has numerous positive attributes, even if they disagreed how they are conceptualized. Hence, from the perspective of a follower of Avicenna, both groups could be understood as ṣifātiyya, even if only Ash῾arites claimed that label for themselves. From the point of view of falsafa, the word 86. AL-ASH῾ARĪ, Maqālāt al-Islāmiyyīn wa-ikhtilāf al-muṣallīn, ed. H. RITTER, Die dogmatischen Lehren der Anhänger des Islams, 3rd ed., Wiesbaden 1980, p. 486. 11–12; cf. VAN ESS, Theologie und Gesellschaft, vol. 6, p. 134. 87. IBN AL-MALĀḤIMĪ, Tuḥfat al-mutakallimīn fī l-radd ῾alā l-falāsifa, ed. H. ANSARI – W. MADELUNG, Tehran 1387/2008, p. 43. 18–19. 88. Ibid., pp. 44–48.

ON THE AUTHENTICITY OF THE THRONE EPISTLE

221

ṣifātiyya can refer to all groups who –unlike Avicenna and his followers – teach that God has positive attributes other than pure existence and necessity. Hence, the Mu῾tazilites become in this Avicennan text “a group from among the ṣifātiyya.” The usage of this term also illustrates that the theological conflict on the divine attributes is very much on the mind of whoever authored al-Risāla al-῾arshiyya. Earlier in the text, he had already included a polemic on this subject. We highlighted an isolated passage in the middle of the third aṣl, where the text, in the style of a kalām argument, argues against the view that a divine attribute exists inside (dākhil) the divine essence “as a constituent of the essence” and that God’s existence cannot be conceived without attributes like these. This position is dismissed with the argument that such a concept of divine attributes violates divine simplicity.89 This passage seems to respond to arguments made by philosophical critics of Avicenna, such as Fakhr al-Dīn al-Rāzī (d. 606/1210), who argues that the Being Necessary By Virtue of Itself must have a quiddity that is distinct from pure existence. In his philosophical books The Eastern Investigations (al-Mabāḥith al-mashriqiyya) and The Compendium on Philosophy and Logic (al-Mulakhkhaṣ fī l-ḥikma wa-l-manṭiq) al-Rāzī develops a philosophy where God’s māhiyya is different from His existence.90 This may well have been understood that according to al-Rāzī – and similar critics like him – God must have something “inside” of Him that is not just existence and necessity. Such a “something inside” would indeed be an attribute. The author of al-Risāla al-῾arshiyya seems to be aware of this kind of philosophical criticism of Avicenna and he responds to it in this epistle. 5. Al-Risāla al-῾arshiyya As a Response to Ash῾arite kalām The usage of the term ṣifātiyya is the smoking gun in my attempt to show that al-Risāla al-῾arshiyya was not authored by Avicenna. If the usage of that word generated slowly among Khorasanian Ash῾arites of the 5th/11th century and becomes widespread only after al-Shahrastānī uses it around 521/1127 in his al-Milal wa-l-niḥal, then Avicenna unlikely ever used it. There is still the possibility that this word, 89. al-Risāla al-῾arshiyya, p. 6. 4–5 / p. 21. 18–20. See above pp. 213–14. 90. GRIFFEL, Formation of Post-Classical Philosophy, pp. 387–416.

222

FRANK GRIFFEL

together with the passage, where it is used, is an interpolation in a genuine text by Avicenna. To show that this is not the case I will supplement this argument with an additional one, namely the text’s awareness of and response to Ash῾arite teachings. The focus on the divine attributes in al-Risāla al-῾arshiyya comes as a result of Ash‘arite criticism and is a development of the 6th/12th century. After al-Ghazālī criticized Avicenna’s theology of a radically simple God in his Tahāfut, the point appears in other kalām texts and it becomes the main point of a philosophical criticism directed against Avicenna. Existence, so the objection goes, is a predicate empty of all content. We can say that something exists, but to say that there is pure existence without such a something is meaningless and empty. If Avicenna teaches that God is pure existence without positive attributes, meaning without such a something, he constructs a meaningless concept of God. Sharaf al-Dīn al-Mas῾ūdī, who was active during the middle of the 6th/12th century and who died around 590/1194, makes this point in his critical commentary on Avicenna’s al-Ishārāt wa-l-tanbīhāt and he refers to al-Ghazālī’s Tahāfut. Avicenna’s proof for God’s existence achieves to show that the Necessary Being has no cause, but it does not prove that this being has no content, i.e. no quiddity that is distinct from existence (māhiyya warā᾿a al-wujūd).91 Earlier, during the first half of the 6th/12th century, Abū l-Barakāt al-Baghdādī in his Carefully Considered Book (al-Kitāb al-Mu῾tabar) had already produced a philosophical metaphysic where God has positive attributes, such as “knowledge,” (῾ilm) “omnipotence,” (qudra), and “wisdom” (ḥikma), which Abū l-Barakāt conceived as concomitants of an undivided and unified divine essence.92 In fact, the argumentative force of this objection was so strong that it even spilled over into the Avicennan movement. Both Bahmanyār ibn 91. A. SHIHADEH, Doubts on Avicenna: A Study and Edition of Sharaf al-Dīn al-Mas῾ūdī’s Commentary on the Ishārāt, Leiden 2016, p. 71; Arab. text, pp. 259–261. See also the 6th discussion in AL-GHAZĀLĪ’s Tahāfut al-falāsifa, pp. 96–109. 92. Cf. e.g. ABŪ L-BARAKĀT AL-BAGHDĀDĪ, al-Kitāb al-Mu῾tabar fī l-ḥikma, 3 vols., Hyderabad 1357–1358 /1938–1939, vol. 3, p. 123. 7–11, where the idea of concomitance is expressed by the fact that in a triangle the sum of all angles are two right angles. Abū l-Barakāt al-Baghdādī’s philosophical concept of God is understudied, see only S. PINES, “Studies in Abu l-Barakât al-Baghdâdî’s Poetics and Metaphysics,” in: Scripta Hierosolymitana 6 (1960), pp. 120–198, at pp. 166–176, 194–195, reprinted in ID., Collected Works of Shlomo Pines. I. Studies in Abu’l-Barakāt al-Baghdādī. Physics and Metaphysics, Jerusalem 1979, text 5.

ON THE AUTHENTICITY OF THE THRONE EPISTLE

223

al-Marzubān and Ibn Khayyām cautiously express the idea that God must have a “content” or a “core” (ḥaqīqa) that existence applies to.93 It has already been explained that later in the 6th/12th century, this issue will become a major point of criticism in Fakhr al-Dīn al-Rāzī’s books of philosophy (ḥikma). It seems that the question of whether or not God can be simply existence and necessity grew to a major philosophical question in the course of the 6th/12th century, much more than it was in the days of Avicenna. The position of al-Risāla al-῾arshiyya is faithfully Avicennan in the sense that it defends his view that God’s “content” or “core” can only be simple existence. This faithful attitude was probably one of the reasons why the authenticity of this work was never seriously called into question. In my presentation of the epistle’s content above, I stopped at the end of the third aṣl, where the word sifātiyya appears. The next chapter deals with the divine attributes and is divided into seven sub-chapters, each dealing with a single attribute. These seven attributes are (1) God’s knowledge, (2) God’s life, (3) God’s will, (4) God’s omnipotence, (5) God’s hearing, (6) God’s seeing, and (7) God’s speech. Listing the attributes in this way, however, is not fully accurate. Both our text and Avicenna do not teach that there are attributes such as “knowledge”, “life,” or “will” that God has. Rather, they explain what it means when humans describe God with attributes (= active participles) such as “knowledgeable” (῾ālim), “living” (ḥayy), or “willing” (murīd). Avicenna did not teach that God or the “First Principle” has any positive attributes, such as knowledge or life. At most, Avicenna was willing to comment on certain divine names that appear in the Qur’ān and in the ḥadīth corpus as active participles. Accordingly, the seven sub-chapters of the part on attributes explain how phrases such as God “is knowledgeable” (kawnihi ῾āliman), “is alive” (kawnihi hayyan) or “willing” (murīdan) are to be understood. What we have here is an Avicennan explanation how certain divine attributes or names can be understood. None of the attributes can be positive, which is why the text explains certain divine names, such as “knowledgeable,” (῾ālim), “omnipotent” (qādir), or “living” (ḥayy) as either relational or negative attributes. Here the choices our author 93. GRIFFEL, Formation of Post-Classical Philosophy, pp. 412–413.

224

FRANK GRIFFEL

makes are important. The seven divine names commented on are the equivalents to the seven attributes of the essence in Ash῾arite kalām: (1) omnipotence (qudra), (2) knowledge (῾ilm), (3) will (irāda), (4) life (ḥayāt), (5+6) hearing and seeing (sam῾ wa-baṣr), and (7) speech (kalām).94 For the Ash῾arites, these attributes of the essence not only represent qualifications but they are the corresponding substances of those qualifications. These ṣifāt are real entitative existents, by means of which God is worthy to be described by the corresponding qualifications (omnipotent, knowledgeable, living, etc.) and names (“the omnipotent,” “the knowledgeable,” etc.).95 The al-Risāla al-῾arshiyya does not refer to this particular Ash῾arite understanding of the divine attributes of essence. Yet by choosing to explain these seven qualifications and by structuring them this way, the author of al-Risāla al-῾arshiyya reveals background knowledge of Ash῾arite theology that Avicenna probably never had. As we have said earlier, during Avicenna’s days, the Ash῾ariyya was a rather insignificant local school in Khorasan. Even if Avicenna knew this group and knew some of their core teachings he would have very unlikely structured one of his compositions in a way that it implicitly responds to their claims. This is, however, precisely what al-Risāla al-῾arshiyya does. In these seven sections, its author explains what it truly means when we say that God is knowledgeable, omnipotent, or willing. The real Avicenna dealt with this subject rarely, and only in one of his works, his Persian Dānishnāmeh-yi ῾Alā᾿ī, does he make a systematic attempt to explain divine attributes. That book’s part on metaphysics includes a relatively long and explicit presentation of Avicenna’s theology.96 Here, Avicenna explains in eighteen individual chapters what he means when he says that God has a “will” (khwāst) as well as “generosity” (jūd), and that God is “powerful” (qādir) and “wise” (ḥakīm). Jules Janssens could show that much of the textual material in Dānishnāmeh-yi 94. For this list see e. g. D. GIMARET, La doctrine d’al-Ash’ari, Paris 1990, pp. 259–267, al-Milal wa-l-niḥal, vol. 1, p. 152, French trans., vol. 1, p. 318, or AL-GHAZĀLĪ, al-Risāla al-Qudsiyya, in: A. TIBAWI, “Al-Ghazālī’s Sojourn in Damascus and Jerusalem,” in: Islamic Quarterly 9 (1965), pp. 65–122, at pp. 84–85 and pp. 104–106. Al-Ash῾arī originally had an eighth attribute (baqā᾿), which later Ash῾arites dropped. 95. D. GIMARET, art. “Ṣifa” in: H. A. R. GIBB e.a. (eds.), Encyclopaedia of Islam. New Edition, 13 vols., Leiden / London 1954–2009, vol. 9, pp. 551–552. 96. IBN SĪNĀ, Dānishnāmeh-yi ῾Alā᾿-i. Ilāhiyyāt, ed. M. MU῾ĪN, Tehran 1331/1952, pp. 65–101. AL-SHAHRASTĀNĪ,

ON THE AUTHENTICITY OF THE THRONE EPISTLE

225

῾Alā᾿ī appears in a similar form among the Arabic text-fragments that make up Avicenna’s Notes (al-Ta῾līqāt), again a book that was put together from Avicenna’s Nachlass.97 There might have even been an Arabic version of Dānishnāmeh-yi ῾Alā᾿ī.98 Other than in Dānishnāmeh-yi ῾Alā᾿ī, Avicenna hardly discusses the divine attributes anywhere else in his extant authentic œuvre. It has already been said that divine attributes are, according to Avicenna, only negative descriptions of God or descriptions of His relations. Al-Risāla al-῾arshiyya presents teachings very similar to Dānishnāmeh-yi ῾Alā᾿ī — again, there is nothing un-Avicennan here. Its author may have wished to offer a presentation of Avicenna’s understanding of the seven Ash῾arite attributes of the essence that any student of Ash῾arite theology, or accomplished Ash῾arite theologian for that matter, may directly benefit from. Hence, al-Risāla al-῾arshiyya stems from a period when Avicennism and Ash῾arism competed with one another for followers. Such direct competition happened, for instance, during the first half of the 6th/12th century, when we see other works, such as al-Lawkarī’s Bayān al-ḥaqq wa-ḍamān al-ṣidq engaging in strategies of writing that put the teachings of falsafa side-by-side to that of kalām.99 The two final sections of al-Risāla al-῾arshiyya of merely four to five pages deal with God’s actions.100 Again, this choice reveals background knowledge of Ash῾arite kalām, where God’s actions are dealt with after an explanation of the divine attributes. The text first explains the concept of emanation (ṣudūr, fayḍ) and clarifies that “from the one only one emanates” (lā yaṣduru min [dhātihi al-wāḥida] illā wāḥid). It fails to 97. IBN SĪNĀ, al-Ta῾līqāt, ed. S. Ḥ. MŪSAVIYĀN, Tehran 1391/2013, pp. 3–28, 524– 584 (# 1–6, 944–1010). See J. L. JANSSENS, “Le Dânesh-Nâmeh d’Ibn Sînâ: Un text à revoir?” in: Bulletin de philosophie médiévale 28 (1986), pp. 163–177 (Reprint in ID., Ibn Sīnā and His Influence on the Arabic and Latin World, Aldershot 2006, text vii). In the introduction to his edition of IBN SĪNĀ, al-Ta῾līqāt, p. 12, S. Ḥ. Mūsaviyān suggests that these passages are translations from Dānishnāmeh-yi Alā᾿i. Some of this material (i.e. al-Ta῾līqāt, pp. 3–28) also appears in an appendix (al-Risāla al-mulḥiqa bi-l-῾ilm al-rubūbiyya) to al-Lawkarī’s philosophical summa Bayān al-ḥaqq, which is extent in MS Tehran, Dānishgāh 250, foll. 212a–218a; on that see J. L. JANSSENS, “Al-Lawkarī’s Reception of Ibn Sīnā’s Ilāhiyyāt,” in: D. N. HASSE – A. BERTOLACCI (eds.), The Arabic, Hebrew and Latin Reception of Avicenna’s Metaphysics, Berlin 2012, pp. 7–26, at p. 17. 98. See the discussion in GRIFFEL, Formation of Post-Classical Philosophy, pp. 423–424. 99. GRIFFEL, Formation of Post-Classical Philosophy, p. 187. See also JANSSENS, “Al-Lawkarī’s Reception of Ibn Sīnā’s Ilāhiyyāt.” 100. al-Risāla al-῾arshiyya, pp. 15. 5–18. 15 / pp. 36. 1–42. 2.

226

FRANK GRIFFEL

address how multiplicity comes about in God’s creation, a problem that is highlighted in Abū l-Barakāt al-Baghdādī’s and Fakhr al-Dīn al-Rāzī’s philosophical responses to Avicenna.101 This might be too complex and difficult a question to address in such a short text. The last section of the text picks up the language of Islamic theology and deals with “God’s decree and determination.” Here we find teachings that Avicenna usually deals with under the heading of divine providence (῾ināya). The epistle explains what the two phrases “His decree” (qaḍā᾿uhu) and “His determination” (qadaruhu), which are part of every Sunni profession of faith (῾aqīda), refer to. The decree is “His allencompassing knowledge of the objects of knowledge as His creations and His generations” and “the firm plan that continues according to one sunna.”102 “His determination is [the fact that] the (secondary) causes are made necessary for [their] effects and that He has no final cause that burdens Him.”103 From the latter results “the divine wisdom” (al-ḥikma al-ilāhiyya) that manifests itself in the way all things exist. At the very end of the epistle its author clarifies Avicenna’s explanation of theodicy, i.e. how evil (sharr) comes about in a fully determined universe. Evil is the privation of existence and comes about as necessary concomitants (luzūm) of all creations, just as the dark shadow is a necessary concomitant of sunlight. God, however is “the Existent per se” (al-mawjūd al-muṭlaq) and hence “the Pure Good” (al-khayr al-maḥḍ), “and no evil ever spoils Him.”104 6. Conclusions The short Throne Epistle (al-Risāla al-῾arshiyya), which has until now almost unanimously been ascribed to Avicenna, shows many features of a work that was written in the second or third quarter of 101. AL-BAGHDĀDĪ, al-Mu῾tabar, vol. 3, pp. 148–152, 156–158; AL-RĀZĪ, al-Mabāḥith al-mashriqiyya, vol. 2, pp. 529–535. 102. al-ḥukm al-thābit al-mustamirru ῾alā sunan wāḥid; Risāla al-῾arshiyya, pp. 16. 5, 18. 3 / pp. 37. 13, 41. 5. Cf. these teachings on divine decree and determination with those of al-Ghazālī where the qaḍā᾿ is also the making of a plan of creation and qadar its execution through secondary causality, see F. GRIFFEL, Al-Ghazālī’s Philosophical Theology, New York 2009, pp. 237–241. 103. Risāla al-῾Arshiyya, p. 16. 5–7 / p. 37. 14–15. On the teachings on qaḍā᾿ wa-qadar see the analysis in MEYER, “Philosophischer Gottesglaube,” pp. 272–274. 104. Risāla al-῾arshiyya, pp. 16–18 / pp. 40–41.

ON THE AUTHENTICITY OF THE THRONE EPISTLE

227

the 6th/12th century. Two pieces of evidence are decisive. First, the text uses the word “ṣifātiyya” in the meaning of “Muslim theologians who ascribe positive attributes to God.” During the times of Avicenna, however, the term ṣifātiyya, was not used outside of the Ash῾arite school. Of all we know, its use became widespread only after al-Shahrastānī’s al-Milal wa-l-niḥal, which started circulating in 521/1127–28. Second, the text reveals that its author had significant background knowledge of Ash῾arite theology. Its division into seven subchapters mirrors the seven divine attributes of the essence in Ash῾arite kalām. This implicit concern with Ash῾arism is untypical for Avicenna, who in his authentic works is not known to have reacted to Ash῾arism and likely had no or only a very superficial knowledge of it. This together with the fact that the first description of al-Risāla al-῾arshiyya appears in the Longer Bibliography of Avicenna’s works, which was written sometime before 588/1192 and after al-Bayhaqī’s Tatimmat (published shortly after 553/1158–69),105 make me conclude that the current text of al-Risāla al-῾arshiyya was composed roughly in the half century between 525/1130 and 575/1180 in a milieu where members of the Avicennan movement (falsafa) and students of Ash῾arite theology interacted with one another and – implicitly at least – competed for followers. This milieu was most likely characterized by the culture of madrasa education in the Islamic east, i.e. in Iraq, Iran, and Central Asia. The author of al-Risāla al-῾arshiyya was a faithful follower of Avicenna whose forgery may have been prompted by the widespread conviction that the unique copy of Avicenna’s al-Ḥikma al-῾arshiyya got lost. Expecting there to be a market for an Avicennan throne-text, he produced this short epistle and may have sold it to a community of copyists and book-dealers, who distributed the work widely. This, if we assume that the forgery was deliberate. It is equally possible that a faithful follower of Avicenna from the community of falāsifa in the 6th/12th century wrote this epistle and that it was later mistaken for one of Avicenna’s genuine works.106 In both cases, knowledge of a genuine work by Avicenna under the title al-Ḥikma al-῾arshiyya, which 105. See above p. XX. 106. On scholars of the first half of the 6th/12th century who were faithful followers of Avicenna see GRIFFEL, Formation of Post-Classical Philosophy, pp. 176–190.

228

FRANK GRIFFEL

got lost during a raid in Isfahan, contributed to this text being accepted as authentic. For many centuries, al-Risāla al-῾arshiyya was understood to be the Throne-Text that Avicenna either lost in Isfahan or that he later re-wrote. In fact, the closeness of the work’s teachings and its style to a work by Avicenna made the detection of the forgery unlikely. This is not a case where a doctrinal enemy tried to foist a text on a highly authoritative author in order to change his public perception and the range of teachings the wider public would associate with him. Rather, a close follower of Avicenna simply reproduced genuine teachings of the Eminent Master. Forgeries like these are rarely detected as there is not much at stake here. The teachings in al-Risāla al-῾arshiyya are Avicenna’s even if the words and the style aren’t always. Once we can accept that Avicenna was not the author of al-Risāla al-῾arshiyya we might further inquire into these differences in word and style. Certain elements of Avicenna’s philosophy are stressed over others, which may provide insights into the reception history of this philosophy. The epistle stresses, for instance, the idea that God’s knowledge (or rather: “Him being knowledgeable”) is single and that “He knows all things by virtue of a single knowledge.”107 While this is part and parcel of Avicenna’s teachings on divine knowledge, its prominent statement right at the beginning of the relevant section might be in response to a critique that Avicenna’s teachings on God’s knowledge of universals diversifies it and deviates from earlier Greek teachings. Such criticism, for instance, already appears in al-Ghazālī, becomes stronger in Abū l-Barakāt al-Baghdādī, and also was picked up by Fakhr al-Dīn al-Rāzī.108 In fact, the language of this epistle with its frequent use of words such as “preponderating factor” (murajjiḥ) and attributes, for instance, that exist “in addition to the essence” (zā᾿id ῾alā l-dhāt) reminds more of authors of the 6th/12th century, such as Fakhr al-Dīn al-Rāzī particularly, than Avicenna. Avicenna does use those phrases, yet probably not as frequently as they appear in this epistle. The talk of attributes that are “additional” to the divine essence has its origin in kalām and its usage in this text is another indicator that its author was intimately 107. ῾ālimiyyatuhu shay᾿un wāḥidun, Risāla al-῾arshiyya, pp. 8. 4, 9. 8, 9. 15 / pp. 24. 14–15, 26. 10. 108. GRIFFEL, Formation of Post-Classical Philosophy, pp. 403–407.

ON THE AUTHENTICITY OF THE THRONE EPISTLE

229

familiar with the language of Ash῾arite mutakallimūn in ways that Avicenna never was. Similar, the text’s quotations from the Qur’ān and the ḥadīth corpus might prove a fruitful avenue of further research. When the text quotes the inauthentic ḥadīth that God’s first creation was “the intellect” (awwalu mā khalaqa Allāhu… al-῾aql) and adds the authentic ḥadīth according to which the first creation was indeed the pen (al-qalam) it mirrors al-Ghazālī’s practice in works like his Fayṣal al-tafriqa.109 Like al-Ghazālī, the text puts much weight on Qur’ānic passages that say that God does not change His habit (sunna). The interpretation of these verses, however, is different from that of the great Ash῾arite. Whereas for al-Ghazālī these verses simply mean that God will not break His own laws of creation (or: laws of nature), here the verses are interpreted as Qur’ānic evidence for a creation without beginning and a pre-eternal world. That idea is utterly un-Ghazalian and the re-purposing of this Qur’ānic verse may well be part of the underlying anti-Ash῾arite polemic in this text.110 The subtle anti-Ash῾arite stand of the text is next to the use of the word ṣifātiyya the clearest indication that it simply does not fit into the lifetime of Avicenna in the early 5th/11th century when Ash῾arism was not yet a theological force to be reckoned with. Reading it as a faithful Avicennan text of the mid-6th/12th century does not diminish its fascination. Here, an expert scholar of Avicenna’s philosophy has written for us a most helpful introduction into the Eminent Master’s theology. Frank GRIFFEL Department of Religious Studies – Yale University 320 York Street New Haven, CT 06511, USA [email protected]

109. al-Risāla al-῾arshiyya, p. 15. 14–16 / p. 36. 11–15. Cf. AL-GHAZĀLĪ, Fayṣal altafriqa bayna l-Islām wa-l-zandaqa, ed. M. BĪJŪ, Damascus 1413/1993, pp. 36–37. 110. wa-lan tajida li-sunnati Lllāhi tabdīlan, Q 33: 62, 48: 23 (where it says: lā tajidu…) and similar in Q 35: 43 and 30: 30; al-Risāla al-῾arshiyya, p. 15. 16 / p. 36. 16. On al-Ghazālī’s use of these verses see GRIFFEL, Al-Ghazālī’s Philosophical Theology, pp. 198–199.

APPROACHING A CRISIS: AL-GHAZĀLĪ IN 488/1095 Taneli KUKKONEN Abstract Al-Ghazālī’s 1095 crisis and departure from Baghdad forms the crux of his autobiography. It is an affecting narrative: at the same time, al-Ghazālī’s remarks on his difficulties in adopting the Sufi path plainly are meant to align with his teachings regarding the relationship between knowledge and action. In this essay I provide remarks on what I term the “innocent” and “critical” schools of Ghazālī interpretation, after which I propose a new reading that combines aspects of both. According to this interpretation Ghazālī is both quite sincere and entirely deliberate in his authorship; he describes a real crisis, even as he interprets it in light of his theoretical convictions.

Every interpreter of Abū Ḥāmid Muḥammad al-Ghazālī (d. 505/ 1111) sooner or later must contend with what was involved in the celebrated scholar’s 488/1095 departure from Baghdad, his relinquishing of a prestigious teaching appointment, and the subsequent turn taken by his career and authorship. Al-Ghazālī’s own intellectual autobiography, The Deliverer from Error (al-Munqidh min al-ḍalāl) paints a picture of a spiritual crisis consequent upon al-Ghazālī’s examination of the different classes of seekers (aṣnāf al-ṭālibīn). The story al-Ghazālī tells has him first affirming intellectually the superiority of the Sufi path and determined to adopt it for himself – after which, however, a period of vacillation follows. Al-Ghazālī flails this way and that, unable to give up his worldly attachments. It is only after a dramatic breakdown, complete with aphasia and a total loss of appetite, that al-Ghazālī finally turns to God and trusts Him to clear the path for him. Al-Ghazālī’s entire late career in life, he intimates, becomes explicable in light of the divine dispensation that followed, a trajectory that transported al-Ghazālī through an extended programme of individual and personal transformation to a mission whose ambition grew to be the renewal of the Islamic faith as a whole.1 1. AL-GHAZĀLĪ, al-Munqidh min al-ḍalāl, ed. K. AYYĀD – J. SALĪBA, Beirut 1969, pp. 100–109.

232

TANELI KUKKONEN

In the great majority of cases, scholars as well as lay readers have treated al-Ghazālī’s story as a straightforward recounting of historical events. This of course has not stopped readers from reading and projecting into that narrative whatever they find appealing. At the same time, a small body of critical scholarship has formed that seeks to challenge this narrative and to unearth covert motives for al-Ghazālī’s actions and words. Such alternative explanations, which often are incompatible and mutually exclusive, range from the medical to the social to the political. In this essay (1) I first provide some brief remarks on what I term the “innocent” and “critical” approaches to al-Ghazālī’s autobiography. (2) I then introduce one particular critical reading – Kenneth Garden’s – as a way of pivoting to (3) a reconsideration of how al-Ghazālī’s writings from 1095, but also after, relate to what we can know about his crisis.2 What is more, I treat this evidence from a systematic and philosophical standpoint. The latter aspect, I take it, is more controversial than the former. I offer this small study, tentative in both findings and tone, to Jules Janssens, who has done so much to enhance our understanding of al-Ghazālī’s thought in its context. 1. Innocent and critical readings The prevailing approach to al-Ghazālī’s account of the events of 1095 continues to be “innocent,” in the sense that al-Ghazālī’s words themselves are taken to be innocent – more or less self-explanatory and manifest in their meaning. Decade upon decade, and in the 21st century no less than in the 20th, retellings continue to proliferate that more or less paraphrase al-Ghazālī. A gloss may be put on his words that emphasizes this or that aspect, but even this is presented as faithful amplification, at most, of what is fundamentally a faithful and dependable narrative. The underlying assumption often tends to be that al-Ghazālī’s professed travails tell us something about an experience we all can recognize and with which we can identify. A projection 2. I refer to “1095” throughout this essay, primarily because it provides a convenient shorthand, but also because the Common Era designation allows us a slightly wider catchment of materials issuing from al-Ghazālī’s pen.

APPROACHING A CRISIS: AL-GHAZĀLĪ IN 488/1095

233

that cuts across centuries can stand in for some universal human constant, or at least an aspirational ideal that al-Ghazālī represents. None of which, of course, is all that innocent. Ill-defined gestures towards a “turn to mysticism” in al-Ghazālī, for instance, do little to deepen our understanding of al-Ghazālī. Rather than elucidate, they tend to obscure what was involved in one individual legal and theological scholar’s (al-Ghazālī) grappling with the vaunted transformative potential of one particular ascetic and contemplative tradition (Sufism, in its post-consolidation but pre-brotherhood phase) at a particular moment in time (500/1106, the publication date for al-Ghazālī’s autobiography).3 Beyond this elementary observation, and getting more into the details of al-Ghazālī’s presentation, one must also question what gets projected on to al-Ghazālī’s story in each instance, and out of what motives. This goes hand in hand with asking what gets lost when al-Ghazālī’s story is pressed to serve a later author’s agenda, whatever that agenda may be. It is easy enough to extract out of al-Ghazālī’s words an outlook that promotes irrationalism, otherworldliness, selfabnegation, a rebuke of worldly religious authority, the glorification of Sufism, or any combination of these. Yet none of the above are quite the same thing, and the differences do matter. For example, one can acknowledge the failure of reasoned argument to bring about a life change without thereby promoting irrationalism, as Aristotle’s words in the Nicomachean Ethics already make clear. Similarly, one can extol the virtues of an ascetic-contemplative path as a general proposition without thereby signing off on every aspect of Sufism as it existed at a specific point in time.4 A careful reading will respect these differences and honestly try to account for the nuances in intellectual positioning that result from reading al-Ghazālī this way or that. In the realm of popular writing, innocent readings of al-Ghazālī are to be taken more or less in stride. How to account for the prevailing lack of critical readings of al-Ghazālī’s autobiography in scholarly 3. To contextualize al-Ghazālī within the broader sweep of Sufism see A. KNYSH, Islamic Mysticism. A Short History, Leiden 2000, pp. 140–149. 4. For the first point see ARISTOTLE, Nicomachean Ethics, X, 9; for the second, K. GARDEN, The First Islamic Reviver, Oxford 2014, pp. 69, 79.

234

TANELI KUKKONEN

circles is another matter, and a question best left for another day. Impressionistically speaking, part of the enthusiasm for al-Ghazālī on the part of Western scholars surely has to do with the fact that his autobiography ostensibly presents them with an exemplary Muslim intellectual of critical disposition – an imaginary fellow traveller and ready-made identification figure that can be made to stand in for “Muslim existentialism,” “proto-modern selfhood,” or whatever else is deemed desirable.5 One may additionally point to the way in which aspects of al-Ghazālī’s self-presentation have appealed to ecumenically minded truth-seekers, often of a basically Christian orientation, within Western letters and academia.6 And one may observe how aspects of al-Ghazālī’s story have been superimposed on (equally idealized) readings of one European philosopher or another – Augustine or Descartes, Rousseau, Hume, or Kierkegaard. Against these “innocent” readings we may set a small but growing body of critical historiography. Some accounts take their start from the fact that al-Ghazālī’s four years in Baghdad were spent working intensively, and across a wide spectrum of venues, in service of the Saljuq caliphate and the intellectual establishment whose job it was to bolster its legitimacy. It might accordingly be postulated that the unprecedented political upheaval that followed the 1092 assassination of the powerful vizier Niẓām al-Mulk had something to do with al-Ghazālī putting some distance – both literal and metaphorical – between himself and the prevailing political climate. Thus, we find already in Duncan Black MacDonald’s early study the suggestion that it was the shifting political terrain within Baghdad that left al-Ghazālī exposed and heading for the exits. And Farid Jabre voices 5. For variations on a theme spanning a century of scholarship see, e.g., D. B. MACDONALD, The Life of al-Ghazzali, Piscataway 2007 (reprint of a lengthy 1899 article in the Journal of the American Oriental Society); W. M. WATT, Muslim Intellectual. A Study of al-Ghazali, Edinburgh 1963; E. ORMSBY, Al-Ghazālī. The Revival of Islam, Oxford 2008. Monograph overviews of classical Arabic/Islamic philosophy, as well as those dedicated to one or another aspect of the history of mysticism, routinely read into al-Ghazālī’s narrative what they wish to see in it. 6. It is instructive to consider from this standpoint the very valuable collection of materials collated by Richard J. McCarthy in his introduction to AL-GHAZĀLĪ, Freedom and Fulfilment, tr. R. J. McCarthy, Boston 1980, pp. 9–52. One of McCarthy’s favored informants, Vincenzo Poggi, was himself both a Jesuit priest and keenly aware of the Christian profile of many of al-Ghazālī’s greatest proponents in Western letters: op. cit., p. 42.

APPROACHING A CRISIS: AL-GHAZĀLĪ IN 488/1095

235

the suspicion that al-Ghazālī might have left Baghdad in fear of Ismā῾īlī assassins, who were incensed by al-Ghazālī’s politically occasioned invective against the Bāṭinites (as al-Ghazālī preferred to call the Ismā῾īlī Shī῾īs then engaged in missionary activities in the Caliphate) and emboldened by the assassination of al-Ghazālī’s sponsor, Niẓām al-Mulk.7 A different line of critical inquiry takes its departure from the fact that in reconstructing the arc of al-Ghazālī’s life and career we have other biographical and autobiographical sources to draw on besides the Deliverer. Although these added resources throw but little direct light on the events of 1095 they do provide some scintillating tidbits – most spectacularly, the claim that al-Ghazālī would have accepted the teachings of the Sufis already some years prior.8 Also of note is the prominence in al-Ghazālī’s preserved correspondence of his vow, taken upon Abraham’s tomb at Hebron in the year 489/1096, never again to accept the patronage of rulers; similarly with his denouncement of disputation and partisanship.9 These nuggets of information provide some clues as to al-Ghazālī’s state of mind close to the events of 1095. They also provide a helpful point of comparison with the stylized account contained in the Deliverer and provide us with an occasion to ask why certain facets feature more prominently in the Deliverer, others less.10 Finally, the more one pays attention to the structure of the Deliverer, the plainer it becomes that al-Ghazālī’s presentation is deliberate – in places downright schematic – so much so that it is manifestly 7. F. JABRE, “La biographie et l’oeuvre de Ghazālī reconsidérées à la lumière des Ṭabaqāt de Sobki,” in: MIDEO 1 (1954), pp. 73–102. 8. We owe the claim to Abū Bakr b. al-‘Arabī’s (d. 543/1148) account of his conversations and correspondence with al-Ghazālī, exchanges that occurred midway through 490/1097. For context see F. GRIFFEL, Al-Ghazālī’s Philosophical Theology, Oxford 2009, pp. 62–71. 9. An English translation of al-Ghazālī’s letter, originally penned in Persian not long after the Deliverer was completed, can be found in K. GARDEN, “Coming Down from the Mountaintop: Al-Ghazālī’s Autobiographical Writings in Context,” in: The Muslim World 101 (2011), p. 590 and in GARDEN, Revival, p. 164. The three vows are also brought up in a later piece of correspondence in which al-Ghazālī declines a different invitation from a ruler: GRIFFEL, Philosophical Theology, p. 51. 10. For a collection of biographical materials see M. HOZIEN, “Ghazālī and His Early Biographers,” in: Islam and Science 9 (2011), pp. 95–122; for al-Ghazālī’s letters translated into German, D. KRAWULSKY, Briefe und Reden des Abū Ḥāmid Muḥammad al-Ghazālī, Freiburg im Breisgau 1971.

236

TANELI KUKKONEN

problematic to regard his prose as autobiographical in any straightforward sense. Josef van Ess may have been the first to point out that the framing device of taking a tour of four schools of thought has a proximate model in ‘Umar Khayyām; while the tropes of Islamic spiritual writing inform precisely those passages that describe al-Ghazālī’s crisis and its resolution in a trust in God.11 The Deliverer’s portrayal of a naturally inquisitive and sharp mind sifting through the virtues of the different schools and their methodologies, with a criterion of truth that has to be generated independently of any of them, harkens back to Galen; and the notion of spiritual medicine informs the whole of the Deliverer in ways whose importance we are only now beginning to appreciate once more.12 Even the Deliverer’s famous opening remarks about a thoroughgoing skepticism are structured so as to teach the reader about possible sources of knowledge, much the same way any theologian or philosopher would do – another sign that we are dealing with a pedagogical narrative, rather than one that is meant to recount actual events in any straightforward fashion.13 The problem with any reading of al-Ghazālī’s Deliverer that is overly innocent is that it fails to pay sufficient attention to context – in particular, the overall shape of al-Ghazālī’s authorship – and that even the best close reading all too easily falls prey to tendentious emphases on the part of the interpreter, thereby obscuring those parts of al-Ghazālī’s account that fail to conform with the reader’s expectations. 11. J. VAN ESS, “Quelques remarques sur le Munqidh min aḍ-ḍalāl,” in: A.-M. TURKI (ed.), Ghazālī, la raison et le miracle, Paris 1987, pp. 57–68. 12. The prominence of spiritual medicine in al-Ghazālī will hardly have escaped his contemporaries. For Galen see S. MENN, “The Discourse on the Method and the Tradition of Intellectual Autobiography,” in: J. MILLER – B. INWOOD (eds.), Hellenistic and Early Modern Philosophy, Cambridge 2003, pp. 141–191; for spiritual medicine, T. KUKKONEN, “Al-Ghazālī on the Origins of Ethics,” in: Numen 63 (2016), pp. 271–298 and S. VASALOU, “Ethics as Medicine: Moral Theory, Expertise, and Practical Reasoning in al-Ghazālī’s Ethics,” in: Archiv für Geschichte der Philosophie, forthcoming. Regarding the tradition of prophetic medicine being interpreted along these lines see F. GRIFFEL – K. HACHMEIER, “Prophets as Physicians of the Soul: A Dispute About the Relationship Between Reason and Revelation Reported by al-Tawḥīdī in his Book of Delightful and Intimate Conversations (Kitāb al-Imtā‘ wa-l-mu’ānasa),” in: Mélanges de l’Université Saint-Joseph 53 (2010– 2011), pp. 223–257. 13. See T. KUKKONEN, “Al-Ghazālī’s Skepticism Revisited,” in: H. LAGERLUND (ed.), Rethinking the History of Skepticism. The Missing Medieval Background, Leiden 2010, pp. 29–59.

APPROACHING A CRISIS: AL-GHAZĀLĪ IN 488/1095

237

The problem with most critical readings is that a hermeneutics of suspicion only carries one so far when dealing with an intellectual trajectory as idiosyncratic as al-Ghazālī’s. A mere cataloguing of influences will not suffice to elucidate al-Ghazālī’s purposes; nor will an alternative account of al-Ghazālī’s motivations succeed if it ends up raising more questions and objections than the (allegedly) naive and innocent reading it is meant to replace. Navigating some sort of middle course would be desirable; yet doing so in some principled fashion, as opposed to producing a mere hodgepodge of conflicting impressions, can prove challenging. 2. From Action to Crisis All this comes into focus once we examine one ambitious reading from the “critical” camp, namely, that put forward by Kenneth Garden in a recent series of studies.14 Garden shows a healthy desire to integrate our understanding of what went on in 1095 with our understanding of al-Ghazālī’s subsequent career. Garden’s scholarship also highlights al-Ghazālī’s acuity in navigating the choppy political waters of his age and his seriousness of intent when carving out a space for a “revived” religiosity among his followers, at least, if not society at large. However, as I hope to demonstrate, a lack of faith on Garden’s part when it comes to al-Ghazālī’s testimony of a debilitating personal crisis – or else a lack of interest in it – leads Garden to miss some obvious clues in the very materials he adduces, ones that could further help him in the act of integration at which he aims. As a result, I believe Garden discounts the testimonial value of al-Ghazālī’s autobiography needlessly, while also losing out on the possibility of connecting al-Ghazālī’s self-critique with the kind of work he ended up doing. Garden begins from a determination – salutary to my mind – to interrogate al-Ghazālī’s state of mind ca. 1095 primarily on the basis of writings stemming from that period. Of especial significance to 14. GARDEN, “Mountaintop,” pp. 581–596; GARDEN, Reviver; K. GARDEN, “Revisiting al-Ghazālī’s Crisis through His Scale for Action (Mīzān al-‘Amal),” in: G. TAMER (ed.), Islam and Rationality. The Impact of al-Ghazālī. Papers Collected on His 900th Anniversary. Volume I, Leiden 2015, pp. 207–228.

238

TANELI KUKKONEN

Garden is The Scale of Action (Mīzān al-‘amal), an exercise in moral psychology that Jules Janssens has shown to rely partly on Avicenna, but even more on al-Rāghib al-Iṣfahānī’s Islamic adaptation of Miskawayh’s Reformation of Morals.15 In Garden’s view, what is most remarkable about the Scale is how much of it previews the conceptual framework underlying the post-crisis Revival of the Religious Sciences and al-Ghazālī’s subsequent authorship. In the Scale, al-Ghazālī already sets his sights on happiness or felicity (sa‘āda) as the ultimate telos or aim of humanity as created by God. Al-Ghazālī also puts forward contemplative knowledge as the fulfilment of that telos, and claims that this is what is involved in the blissful vision of God, His attributes, and His works promised by some of the Sufi saints. The Scale of Action thus represents an early stage in al-Ghazālī’s project of reading the Islamic ascetic and contemplative tradition against the theoretical edifice of the Arabic Aristotelians and vice versa – a fundamentally ecumenical and eirenic outlook that carries through the remainder of al-Ghazālī’s career, as established already in numerous studies.16 In addition, Garden points out the Scale’s placement as the final link in a trilogy of treatises all dating back to 1095. The Scale of Action is presented as an appendix, for all intents and purposes, to The Yardstick of Knowledge (Mi‘yār al-‘ilm), a concise exposition of Aristotelian logic in the mould of al-Fārābī and Avicenna;17 the Yardstick in turn is announced in al-Ghazālī’s celebrated Incoherence of the Philosophers (Tahāfut al-falāsifa), the best known of all the works published during al-Ghazālī’s tenure at the Niẓāmiyya college in Baghdad. If the Incoherence represents an attempt to ringfence those of the philosophers’ theoretical doctrines that warrant censure, therefore, and those that the philosophers (contrary to their lofty claims) are 15. J. JANSSENS, “Al-Ghazālī’s Mīzān al-‘Amal: an Ethical Summa Based on Ibn Sīnā and al-Rāghib al-Iṣfahānī,” in: A. AKASOY – W. RAVEN (eds.), Islamic Thought in the Middle Ages, Leiden 2008, pp. 123–137. See further Y. MOHAMED, The Path of Virtue. The Ethical Philosophy of al-Rāghib al-Iṣfahānī, Kuala Lumpur 2006. 16. Fittingly, one of the best studies we have issues from the pen of Jules Janssens: see J. JANSSENS, “Al-Ghazālī between Philosophy (Falsafa) and Sufism (Taṣawwuf): His Complex Attitude in the Marvels of the Heart (Ajā’ib al-Qalb) of the Iḥyā’ ‘Ulūm al-Dīn,” in: The Muslim World 101 (2011), pp. 614–632. 17. J. JANSSENS, “Al-Ghazzālī’s Mi‘yār al-‘ilm fī fann al-manṭiq: sources avicenniennes et farabiennes,” in: AHDLMA 69 (2002), pp. 39–66.

APPROACHING A CRISIS: AL-GHAZĀLĪ IN 488/1095

239

unable to demonstrate; and if the Yardstick represents an attempt on al-Ghazālī’s part to demonstrate to any interested party what he had already stated in the introduction to the Incoherence, namely, that logic is part of humanity’s universal endowment as granted by God, not a discipline foreign and inimical to the Muslim public; then the Scale serves as an equally expansive affirmation of what al-Ghazālī had posited at the very tail end of the Incoherence, namely, that much of the moral psychology and soteriology of the philosophers can be accepted and indeed embraced by the devout Muslim.18 In other words, we may consider al-Ghazālī to have canvased all three dimensions of the philosophical curriculum during his final year in Baghdad: logic, theoretical philosophy, and practical philosophy.19 From all this Garden concludes that much of what al-Ghazālī taught post-1095, he already held to be true prior to his departure from Baghdad. In fact, Garden, similarly to Frank Griffel, maintains that there is little to separate the theoretical positions held by al-Ghazālī throughout his entire active career.20 From an early age al-Ghazālī was engaged in a creative appropriation and adaptation of Arabic Aristotelianism, especially the vibrant mutation issuing from the pen of Ibn Sīnā: if correct, this would put him al-Ghazālī squarely at the centre of the “Avicennian pandemic” of the 5th/11th and 6th/12th centuries. Garden holds this perspective in common with a sizable contingent of the current generation of al-Ghazālī scholars. Indeed, while the balance of Avicennian vs. Ash‘arite impulses in al-Ghazālī remains a contested matter when it comes to any number of individual issues, it is hardly controversial anymore to maintain that the philosophical mode of analysis and explanation exerted a powerful influence on al-Ghazālī, from the beginning of his career to the end. 18. See The Incoherence of the Philosophers, 2nd ed., tr. M. E. MARMURA, Provo 2000, pp. 208–214. 19. Notice how the Deliverer from Error, some eleven years later, mirrors this effort. In the Deliverer each of the three domains is broken into two subfields (logic and mathematics, physics and metaphysics, ethics and politics), with al-Ghazālī conveying much the same outlook when it comes to each: logic and mathematics are to be accepted by the reflecting Muslim; physics and metaphysics are on shakier ground than what the philosophers let on; religious and philosophical ethics and politics can be harmonized, though questions remain about the relative primacy of the different modes of instruction. See Munqidh, pp. 79–90. 20. GRIFFEL, Philosophical Theology, p. 43; GARDEN, Reviver, p. 59.

240

TANELI KUKKONEN

Where this intersects with our interpretation of the events of 1095 is in coming up with a narrative that would motivate al-Ghazālī to make a radical change in his career. Garden begins from a portrayal of al-Ghazālī as someone who at the onset of 1095 was supremely confident both in his own abilities and in his command of the facts. Garden accordingly opts for a version of the story in which all the moves made by al-Ghazālī are deliberate acts of self-recreation. On Garden’s telling, al-Ghazālī, realizing that the stance of the knowit-all theoretician and philosopher was unlikely to win over many followers, chose instead to cultivate an image as an iconoclast religious revivalist and a putative Sufi master. Al-Ghazālī’s commitment to the Sufi path accordingly was at least as much a pragmatic decision about how to become credentialed as an instructor to the (literate, well-todo) general public as it was about any felt need for authentic selfimprovement. And his “conversion” was carefully calibrated to follow the expected script of someone turning their attentions from the world to a single-minded pursuit of God. The impression one gets is one of a stage-managed set of maneuvers, orchestrated to prepare the grounds for al-Ghazālī’s subsequent teaching of a newly minted discipline, the “science of the path of the afterlife,” alongside an attendant “revival of the religious sciences.” The picture is seductive in its way, if not quite as novel as might be thought: much the same view was put forward already by ‘Abd al-Dā’im al-Baqarī in the early twentieth century.21 However, perhaps paradoxically, perhaps not – such is the way of things sometimes – Garden’s preferred interpretation becomes the more brittle, the more one tries to get a precise fix on the assumptions on which it rests. To pick a few examples: Garden makes reference to the fact that al-Ghazālī’s account of his 1095 crisis reproduces some of the tropes of tawba, or repentance, an archetypal return or “turning to” God that held a powerful sway in the Muslim imaginary.22 This surely has the ring of truth to it. Yet if an adherence to a literary trope was all that motivated al-Ghazālī, could he not have come up with a better and more flattering story? What is the point of recording the fluctuations of al-Ghazālī’s resolve this way and that – of him 21. For al-Baqarī see Freedom and Fulfilment, pp. 23–25. 22. For tawba see A. KHALIL, Repentance and the Return to God, Albany 2018.

APPROACHING A CRISIS: AL-GHAZĀLĪ IN 488/1095

241

experiencing again and again the pull of fame and fortune, familiar concern and familiar comforts? If the purpose of al-Ghazālī’s story was simply to consolidate his positioning as God’s chosen reviver of religion for his day and age, could this not have been achieved with a narrative that was less personally embarrassing to him – one that at the very least had him see the error of his ways all at once and thereafter put him on the correct path right away? The problem is compounded once we consider al-Ghazālī’s description of his own actions immediately following his release from the malaise that had afflicted him. Al-Ghazālī, as per the Deliverer, freely concedes that he floated a fiction about performing the pilgrimage to Mecca and Medina as a pretext for his leaving Baghdad, with this neither being his true intention nor what actually happened (although he did perform the pilgrimage later, this time with pure intent, as he takes care to point out).23 Why call attention to such duplicity? This facet of the Deliverer becomes all the more puzzling once we consider its composition in 500/1106, just as al-Ghazālī was being called back to service by Sanjar, the ruler of Khorasan. While appeals to independent-mindedness might be expected in an environment where al-Ghazālī wished to proclaim his effective autonomy from the demands placed by sponsors and rulers, deliberate deceit is another matter, and not likely to have been received warmly. The same goes for the way in which al-Ghazālī brings up a suspicion floated by some in Baghdad in 1095 that he would have left because of some kind of falling-out with the Saljuq chain of command.24 Although al-Ghazālī repudiates any such insinuation, the very fact that he feels the need to bring it up in the Deliverer indicates that there were uncomfortable and intransigent known facts about 1095 that were still floating around twelve years later, ones he could not very easily suppress or elide. Stepping back for a second, and putting two and two together: If al-Ghazālī in 1095 had been fully in control of orchestrating his own narrative and the way subsequent history portrayed him then he could, e.g., easily have embarked on his pilgrimage in an ostentatious manner and then executed some performative version of a conversion 23. Munqidh, pp. 104–106. 24. Munqidh, p. 105.

242

TANELI KUKKONEN

en route to Mecca. A single Damascene moment (to misuse a familiar phrase) would surely have done the job Garden envisions for a stage-managed, so-called “crisis,” and done so much more efficiently and tidily than the accounting of events we find in the Deliverer. We are thus forced to contend with the possibility that we have in the text of al-Ghazālī’s autobiography some version of a lectio difficilior, an echo or residue (if not quite a full and frank accounting) of real-life events that al-Ghazālī could not have erased out of the public record even if he had wanted to. And this comports with prima facie considerations that register as common sense anyway. The Deliverer was published in 500/1106, a mere eleven years removed from the events that it claimed to portray, and it took as its subject one of the caliphate’s most visible public intellectuals. It is unlikely that even an author as audacious as al-Ghazālī would have dared to float a story about his final months in Baghdad that had no basis in reality. It is equally implausible that he should have playacted the entire sequence of events, seeing as more attractive versions of the story could have been staged by al-Ghazālī if that was what he was after. We may therefore regard the unvarnished outwardly visible facets of al-Ghazālī’s 1095 crisis – his general malaise, his inability to teach, his retreat into an uncommunicative crouch – as something that was both genuinely felt and in the public eye. 3. From Crisis to Revival This leaves us with the question of what could have brought about a breakdown of this magnitude in a person as confident in his powers as al-Ghazālī gives every appearance of being. Medicalizing labels – burnout, stress, anxiety – are descriptions in search of an explanation; at any rate, if that is what we seek, then we can do no better than to turn to al-Ghazālī himself, who says that the doctors examining him concluded that his was an illness that had spread from the heart (qalb) to the temperaments (mizāj).25 Whether these are physicians of al-Ghazālī’s own imagining or some real-life medical practitioners of the age, the conclusion should be read in light of the psychophysiological picture which al-Ghazālī had adopted by that point in his life. There 25. Munqidh, p. 104.

APPROACHING A CRISIS: AL-GHAZĀLĪ IN 488/1095

243

was an intellectual infirmity that had set its roots in al-Ghazālī midway through 1095, and this now threatened his ability to do the intellectual work he needed to do.26 But how, if – to repeat the point – al-Ghazālī regarded himself as the sovereign master of every discourse he had tackled? Two known features about al-Ghazālī’s works and authorship intertwine here. First, consider the range of intellectual work in which al-Ghazālī was engaged during late 1094 and the early part of 1095. His feats in learned polemic and disputation, which were both credentializing in scholarly circles and required of him as a public officiary, had resulted in the two masterworks The Infamies of the Bāṭinites and The Incoherence of the Philosophers. Al-Ghazālī’s remaining dialectical work from the period, The Mean in Belief (al-Iqtiṣād fī al-i‘tiqād), while competently executed, starkly exposed the limits to al-Ghazālī’s interest in pursuing the endless dialectical pathways of kalām theology as it was practiced at the time, as well as the limitations of the kalām treatise as a vessel for al-Ghazālī’s own ideas as they took new shape following his absorption of philosophical cosmology and psychology; while The Jerusalem Letter, a compact creedal statement that was released later in 1095 following al-Ghazālī’s departure from Baghdad, tells us that he had nothing of interest to add to that genre.27 The same in a sense goes for The Yardstick of Knowledge and The Touchstone of Speculation (Miḥakk al-naẓar), al-Ghazālī’s two exercises in recasting the logic of the falāsifa in a mould more inviting to nonphilosophers. While these works show that al-Ghazālī had capably 26. The heart is a central technical term in al-Ghazālī’s mature work, from the Revival of the Religious Sciences onwards. It is al-Ghazālī’s preferred term for what the Sufis regarded as the true, higher self and what the philosophers termed the intellect (῾aql). It is the heart that is receptive to reality and to the true nature of things: by God’s grace, the heart may come to enjoy unobstructed insight (baṣīra) and thereby reach the true happiness and ultimate bliss that God has intended for humankind to experience. When al-Ghazālī says that it was his heart that had fallen sick, therefore, what he means to say is that his ailment was in the first place a sickness of the soul – an intellectual one, no less – and only in the second place a physical affliction. On the heart see further JANSSENS, “Marvels”; T. KUKKONEN, “Al-Ghazālī on the Structure of the Soul,” in: The Muslim World 102 (2012), pp. 541–561; A. TREIGER, Inspired Knowledge in Islamic Thought, New York – Abingdon 2012. 27. This assessment of the Mean in Belief and the Jerusalem Letter relies to a great extent on the analysis of R. M. FRANK, Al-Ghazālī and the Ash‘arite School, Durham / London 1994, pp. 28–75.

244

TANELI KUKKONEN

internalized both the procedures of Aristotelian syllogistic and the demonstrative ideal it was designed to serve, they show no promise towards al-Ghazālī making advances in the field of logic, taken as such. Nor again do al-Ghazālī’s logical writings from the period connect up in any meaningful way with those aspects of his post-crisis project that have to do with making the psychology and cosmology of Avicennian Aristotelianism safe for mainstream Muslim consumption. (It would be another fifteen years before al-Ghazālī’s preface to his late legal work al-Mustaṣfā would provide a bold opening for integrating philosophical logic into the theorization of Islamic law.) I can do no better on this point than to cite Tony Street’s verdict: “Ġazālī was a promoter of logic, but not a practitioner.”28 This leaves The Scale of Action as the major document from which to glean al-Ghazālī’s true preoccupations at the time. Garden is absolutely right to insist that the pursuit of contemplative happiness, as well as an effort at understanding what might be involved in that pursuit, forms as clear a throughline for al-Ghazālī’s constructive project (as opposed to his dialectical work) as one can hope to identify. This is hardly a new observation, however, Garden provides an additional piece of evidence, and a very valuable one, that arrives at us from the direction of chapters 7 and 8 of the Scale. In these chapters, which issue from al-Ghazālī’s own pen rather than being adaptations of either Avicenna or al-Rāghib al-Iṣfahānī, al-Ghazālī presents a hierarchy between the various classes of seekers. Those who pursue the purification of the soul through a tempering of its lower parts, appetite and spirit – ostensibly the subject of the Scale – are the majority, while those few who are capable of theoretical reflection and set on its path from an early enough age form an elite minority. To this picture al-Ghazālī next adds a curious flourish: he says that after completing their course of theoretical studies, the rationally oriented minority may return to a life of practice and spiritual striving,29 at 28. T. STREET, “Arabic Logic,” in: D. M. GABBAY – J. WOODS (eds.), Handbook of the History of Logic. Volume One. Greek, Indian and Arabic Logic, Amsterdam 2004, p. 559. See further U. RUDOLPH, “Die Neubewertung der Logik durch al-Ġazālī,” in: D. PERLER – U. RUDOLPH (eds.), Logik und Theologie. Das Organon im arabischen und lateinischen Mittelalter, Leiden 2005, pp. 73–97. 29. I say “return” because al-Ghazālī subscribes to the view, common in philosophical circles since late Antiquity, that a pre-philosophical tempering of the lower soul and

APPROACHING A CRISIS: AL-GHAZĀLĪ IN 488/1095

245

which point they may perhaps become privy to further divine disclosures of a type hitherto unseen. There is no mistaking that al-Ghazālī fancies himself as belonging to this group, the ‘elect among the elect.’30 Combine this with the plain words of the Deliverer when it comes to his encounter with Sufism. After characterizing the Sufis as the “lords of states” (arbāb al-aḥwāl) whose path comprises both knowledge and action (‘ilm wa-‘amal),31 al-Ghazālī admits that he “began from a survey of their knowledge by reading their books,” and that this was due to personal predilection: “knowledge came easier to me than action.”32 This was followed by an attempt at “wayfaring” (sulūk), that is, putting into practice their programme of relinquishing worldly concerns and severing attachments. Now, this cannot be construed either as the pre-philosophical character training prescribed by the philosophical tradition or the long programme of reformation advocated by the Sufis for the majority of seekers. Rather, al-Ghazālī must have aimed at (as indeed his words make clear) a swift refresher tour of duty, at the end of which he hoped – even expected – to experience the kind of transfixing and transformative visions that both the Sufis (as he reads them) and the philosophers (as he understands them) describe as an intimate acquaintance (ma‘rifa) of the object perceived, rather than just a set of external beliefs held about it.33 What a difference there is between knowing the definitions and causes and conditions of health and satiety and [actually] being healthy and formation of good character is needed for a reflective disposition to gain a foothold in the first place. For brief remarks see T. KUKKONEN, “Al-Ghazālī on the Emotions,” in: TAMER (ed.), Islam and Rationality, pp. 138–164. 30. See AL-GHAZĀLĪ, Mīzān al-‘amal, ed. S. DUNYĀ, Cairo 1964, pp. 221–228. 31. Munqidh, pp. 101-102. 32. Munqidh, p. 100. 33. This interpretation of al-Ghazālī’s deep dive into Sufism dissolves the tension between al-Ghazālī adopting the Sufi path in 1095 (as per the Deliverer) and the report we have from Abū Bakr b. al-‘Arabī that puts his acceptance of the veracity of Sufi teachings several years earlier. If the al-Ghazālī of the Baghdad years believed in what the Scale of Action teaches, then he may well have thought that a theoretical appreciation of the Sufis would suffice (at that point) and that any practical application could wait until after he had worked out the essential theoretical harmony between the perspectives of the theoreticians – i.e., the philosophers – and the more contemplative Sufi masters. This would place the latter effort in 1095, with the catastrophic findings described in the Deliverer becoming apparent only then, compounded by the fact that al-Ghazālī had no Sufi master to guide him.

246

TANELI KUKKONEN

sated! And how great a difference there is between your knowing the definition of drunkenness – that it is a term denoting a state resulting from the predominance of vapours rising from the stomach to the seat of reason – and actually being drunk! Indeed, the drunkard, while drunk, does not know the definition and concept of drunkenness; he has no knowledge of it. The one who is sober, by contrast, knows the definition of drunkenness and what goes into it, though he does not experience the inebriation for himself.34

Al-Ghazālī’s famous comparison between having full scientific knowledge of intoxication and becoming intoxicated oneself is buttressed by another, less often cited one between possessing medical science and being healthy, and a third about knowing what asceticism entails and actually denouncing the present world. All three examples point in the same direction: al-Ghazālī believed he had already learned all he could by way of listening to others and studying the theory of things (samā‘ wa-‘ilm), and now it was time to put all this knowledge into practice. The expected fruit of such practice was a qualitatively different access to the divine truths, such as was promised both by the Muslim contemplative tradition and by Avicenna’s philosophy.35 Now suppose that al-Ghazālī was denied precisely such a taste (dhawq) of the Truth, or the “realities of things” as al-Ghazālī would have put it. (This is in fact confirmed indirectly by the Deliverer when al-Ghazālī refers to the profound truths he learned regarding the fruits of the Sufi path as something that came to him in bits and pieces over the following ten years.) Might al-Ghazālī not have concluded that – consistent with a standing theme throughout his later authorship – what he thought were settled truths in his heart were in fact all but empty names for him (notwithstanding any firm belief al-Ghazālī may have professed in the major articles of faith), or at best externally defined concepts without an internalized understanding such as would facilitate an “exchange of qualities” (tabaddul al-ṣifāt)? Was true intellection of the demonstrative sort yet missing, after all, or was something more – and different – from mere argument required in the first place? When al-Ghazālī in the Deliverer says that in 1095 his grasp of the fundamental religious truths – God, prophecy, the Day 34. Munqidh, pp. 101–102. 35. Altogether too complex to address here is the crucial question of what kinds of divine truths are supposed to be accessed by the perfect contemplator.

APPROACHING A CRISIS: AL-GHAZĀLĪ IN 488/1095

247

of Judgement – and of the need to denounce the present world had not, in fact, been reached by way of any determinate and precise proof (lā bi-dalīl mu‘ayyan muharrar), but instead through different avenues and causes that escape accounting, this smacks of post facto editorializing.36 One way to read the statement is that it is meant to put al-Ghazālī’s cognitive state – as sized up by al-Ghazālī in 1106, of course – regarding religious matters on a level no higher than that of the ordinary believer.37 Worse yet, imagine al-Ghazālī coming to grips with the fact that he could not even approach the contemplative bliss he thought was promised to him, because what he thought would be easy spiritual exercises – a setting aside of worldly concerns, a scrutiny of the workings of his own lower soul – instead proved insurmountable obstacles to him. If the supposed clarity he had reached by this point regarding the need for renouncement and the soul’s abstention from passion (taqwā wa-kaff al-nafs ‘an al-hawā) was somehow not enough to animate him, then this could only mean that the words themselves had failed to come alive for him.38 Might this not have provided a genuine jolt to al-Ghazālī’s conscience, a shock to the system? And would al-Ghazālī not have sought an explanation to his predicament in the cognitive and moral psychology he had by now wholeheartedly adopted? If so, al-Ghazālī will have discovered that the closest explanation for why his intellectual capacity was not operational in the desired way was that his lower soul was interfering with its connection to the supernal world. The mirror of al-Ghazālī’s soul, far from being polished, was in fact tarnished with all manner of worldly distractions and delusions. Indeed, the likeliest model was one of a vicious 36. Munqidh, p. 102. 37. For a careful analysis of the formation of ordinary belief in al-Ghazālī see R. M. FRANK, “Al-Ghazālī on taqlīd: Scholars, Theologians, and Philosophers,” in: Zeitschrift für Geschichte der Arabisch-Islamischen Wissenschaften 7 (1991–1992), pp. 207–252. 38. Munqidh, p. 103, and compare here al-Ghazali’s lengthy disquisition in the Revival’s Book of Knowledge on the two types of certitude (yaqīn), one propositional and speculative, the other of a type that motivates a person towards action: Iḥyā’ ‘ulūm al-dīn I, Jeddah 2011, vol. 1, pp. 270–278. I have commented on the significance of this passage to the interpretation of al-Ghazālī in T. KUKKONEN, “Meditating on the Meditations: Al-Ghazālī, Teresa of Ávila, Descartes,” in: Oxford Studies in Medieval Philosophy 8 (2020), pp. 111– 143.

248

TANELI KUKKONEN

feedback loop, with false beliefs feeding bad habits and vice versa.39 But this would have meant that something in al-Ghazālī’s present lifestyle – indeed, a combination of factors – was feeding these tendencies: the reward of worldly goods, the flattery of rulers, power over students and their adoration, and (last but not least) the satisfaction of vanquishing one’s enemies in intellectual jousting.40 And yet it was precisely al-Ghazālī’s positioning at the top of Islamic society that provided the conditions for his continued exploration of these very topics. And it was those very public venues (courtly meetings, teaching situations, familiar genres of literary production) in which that exploration would be expected to take place. The conundrum would be enough to stymie anyone. Garden in fact acknowledges some of this. He allows that in the course of 1095, al-Ghazālī “perhaps… was concerned that as long as the theoretical prospect of a more luminous insight through the subsequent practice of Sufism remained only a theory, he could not be certain of his own attainment of felicity […] Perhaps he was also tormented by doubts about the sufficiency of his philosophically guided ethical practice.”41 Yet he chooses not to elaborate any further; in general, Garden’s work makes scarce allowances for the possibility that any torments al-Ghazālī may have felt would have substantially altered the way he approached the attainment of salvation and felicity. I believe this to be a missed opportunity both in terms of how we account for the changes attributed to al-Ghazālī by the various sources and when it comes to correlating these personal changes to his shifting authorial interests post-1095. As regards the first point, the clues are all there, in the biographical record established by Garden, Griffel, and others. Consider for instance the famous vow al-Ghazālī made at Abraham’s tomb. Its key 39. This model is sketched out, with extensive documentation, in: T. KUKKONEN, “Al-Ghazālī on Error,” in: F. GRIFFEL (ed.), Islam and Rationality. The Impact of al-Ghazālī. Papers Collected on His 900th Anniversary. Volume II, Leiden 2016, pp. 3–31. 40. Notice again the exact wording of al-Ghazālī’s testimony as relates to his state of mind in 1095. Al-Ghazālī says in the Deliverer that his self-assessment brought about the startling discovery that even his noblest activities, instruction and teaching (tadrīs wata‘allum), were not only geared towards sciences – law and theology – that ultimately had little worth when weighed against the pursuit of the hereafter; they were additionally pursued for the sake of “high position and widespread fame” (jāh wa-intishār al-ṣīt) rather than for the sake of God. Munqidh, p. 103. 41. GARDEN, Reviver, pp. 58–59.

APPROACHING A CRISIS: AL-GHAZĀLĪ IN 488/1095

249

components are al-Ghazālī’s refusal to appear before rulers, to accept their sponsorship, or to engage in partisan disputation. While these terms can be given a purely political interpretation, they can just as easily be seen as al-Ghazālī setting limits and safeguards on his own exposure to the temptations of appetite and spirit, closely in line with what the Revival recommends and in accordance of what al-Ghazālī’s analysis is in that work regarding the peculiar pitfalls that beset scholars.42 There is a well-known passage in The Jewels of the Qur’ān that indicates that al-Ghazālī was very far from considering himself exempt from such temptations.43 A further data point is provided by ‘Abd al-Ghāfir al-Fārisī (d. 529/1134) in his partially preserved biography of al-Ghazālī, written on the basis of encounters and conversations al-Fārisī had with al-Ghazālī first in Baghdad and then in Ṭūs towards the end of the latter’s life. In his account, al-Fārisī expresses amazement at the change in demeanour undergone by al-Ghazālī: he specifically notes the absence of arrogance, impatience, and combativeness that had previously characterized the famous scholar’s conduct. Al-Fārisī acknowledges that at first he was skeptical, suspecting al-Ghazālī’s reformed habitus to be a mere affectation, until he became convinced of his sincerity. Al-Fārisī notes that even the fresh controversies to which al-Ghazālī was drawn post-1106 failed to rouse in him the feistiness for which he was previously known.44 All this speaks to a concerted effort on al-Ghazālī’s part to tame and counter the dominance of the spirited part of his soul in particular, or “the ghūl of the intellect,” so named for the ability of ghaḍab to twist the reasoning part of the soul to its own limited ends of seeking recognition, victory, and the subjugation of others even at the expense of an earnest dedication to the cause of truth. But it also speaks to those very tendencies being pronounced in al-Ghazālī’s previous life, a fact to which we find independent testimony elsewhere. 42. The ties to high rank and money (jāh wa-māl) mentioned in the Deliverer (Munqidh, p. 103) map on the objects of the spirit and appetite, as per the analysis given in the Revival of the Religious Sciences: Iḥyā’, XXVII, XXVIII. 43. Al-Jawāhir al-Qur’ān, ed. M. R. RIḌĀ AL-QABBĀNĪ, Beirut 1990, pp. 45–46. A full study of al-Ghazālī’s analysis of the peculiar psychopathology of the conceited scholar has yet to be written but should prove illuminating. 44. An English translation of the relevant passage is supplied by MCCARTHY, Freedom and Fulfilment, pp. 14–18.

250

TANELI KUKKONEN

To all this we may add the curious fact that exactly around the time al-Ghazālī’s crisis is said to have occurred, he also reached a personal milestone of sorts. If we follow Frank Griffel’s reasoning in shifting the presumed year of al-Ghazālī’s birth to ca. 448/1056, as I believe we have reason to do, then in 488/1095 he will have been close to 40 years of age. A few isolated remarks made in al-Ghazālī’s works then begin to acquire a fresh importance. There is an intriguing fragment in Oh Youth, a hortatory later work that takes the form of a letter to a disciple. In this epistolary tract, al-Ghazālī cites a Prophetic saying according to which whosoever reaches the age of forty without having their virtue (khayr) overpower their vice (sharr) should begin preparing for hellfire. In the Revival, a parallel, more colourful tradition has it that Satan will rub his palm in the face of one who has not repented by the age of forty.45 If we accept Griffel’s revised date of birth for al-Ghazālī, the crisis of 1095 occurred just as he was approaching that pivotal age. This seems to me sufficiently close to count as more than mere coincidence.46 The ḥadīth will have reinforced in al-Ghazālī’s mind that there was no time to lose. Al-Ghazālī himself puts it that he perceived himself as standing on the edge of a crumbling bluff.47 As regards the second point, having to do with the shift in al-Ghazālī’s preoccupations as an author and his style: Griffel concludes that al-Ghazālī’s theological and philosophical doctrines do not shift preand post-1095, while Garden maintains that “the central concepts of The Revival of the Religious Sciences are already present in The Scale of Action.” Both are correct; yet I think that Garden goes too far when 45. Ayyu-hā l-walad, ed. and tr. T. MAYER, Cambridge 2005, p. 5; Iḥyā᾿, XXI, vol. 5, p. 103. 46. GRIFFEL, Philosophical Psychology, p. 42 notes the significance of the age of 40 as a literary motif, but does not make the connection with the two Prophetic sayings. 47. Munqidh, p. 103. Al-Ghazālī may have considered the ḥadīth to have sound psychological underpinnings: in the Book of Repentance, in the Revival of the Religious Sciences, he expresses the conviction that the intellect is perfected around the age of 40, and in the same context points out that if the passions are allowed to run rampant before the maturation of the intellect, they will prove extremely hard if not impossible to bring under control. See Iḥyā’, XXXI, vol. 7, pp. 33–34; also Iḥyā’, XXII, vol. 5, pp. 201–202; for comments, KUKKONEN, “Emotions,” pp. 148–151. The Book of Repentance in the Revival, unsurprisingly, underlines the importance of repenting at haste: “each hour, nay, each breath is a precious jewel for which there is no replacement” (Iḥyā’, XXXI, vol. 7, pp. 41–42).

APPROACHING A CRISIS: AL-GHAZĀLĪ IN 488/1095

251

he says regarding al-Ghazālī that “488/1095 brought no radical reassessment of his intellectual convictions.”48 It all depends on what one means by “radical,” “intellectual,” and “convictions,” I suppose. It appears to me that something very significant – even profound – happens with the realignment of al-Ghazālī’s authorship post-1095. Al-Ghazālī pivots from a predominantly dialectical literary production to a direct and hortatory mode of address, while his objects of interest expand from a care only for the concerns of the intellectual and spiritual elites to the overall shape of every Muslim’s religious life and its hidden spiritual dimension across an incredible range of happenings both great and small. Al-Ghazālī’s analysis of the conditions for divine disclosures undergoes a subtle shift, one that favours the ordinary religious aspirant possessed of sound heart and pure intention; his stress on the dangers of antinomianism and insistence on the hidden mysteries of the ritual ordinances and their purificatory power signals that not only ethical striving, but lifelong Islamic (or prophetically guided) ethical striving is necessary for the soul’s purification to reach a sufficient level for the conditions of felicity to be met. While it would be too hasty to attribute all this to al-Ghazālī’s crisis of 1095, it can scarcely be denied that the trendlines all point in the same direction.49 5. Concluding remarks The approach I have advocated here, and the working hypotheses that undergird it, will strike some as curiously retrograde in character. 48. GRIFFEL, Philosophical Theology, p. 43; GARDEN, Reviver, p. 59. 49. It can reasonably be queried whether the Deliverer serves as a counterexample to my characterization of al-Ghazālī’s output post-1095 as being low on dialectic. However, once one digs deeper into the way the Deliverer’s polemics work, it seems questionable to characterize the work as being dialectical in nature. The theologians and philosophers are put in their place, it is true; yet the useful aspects of their work are made to serve al-Ghazālī’s newly established hierarchy of the sciences, as indeed are the Sufis (albeit that the latter are brought in line by way of praise rather than criticism). In none of the three cases does al-Ghazālī engage in extensive back-and-forth argumentation on individual points of doctrine, which is the true hallmark of dialectical disputation. This is only done with regard to the Bāṭinites, who in effect get to act as the foil against whom al-Ghazālī establishes the methodological point about the ineluctability of reason. In terms of content, there is nothing in the Deliverer that would add to the picture drawn in the pre-crisis Infamies: see F. MITHA, Al-Ghazālī and the Ismailis, London / New York 2001; KUKKONEN, “Skepticism Revisited,” pp. 34–38.

252

TANELI KUKKONEN

Surely, I do not mean to say that a combination of theoretical considerations and personal soul-searching would have motivated a sea change not only in al-Ghazālī’s authorship, but in his life? That seems even more old-fashioned, somehow, than the notion that al-Ghazālī would have fled Baghdad because he was in mortal fear of condemning his soul to perdition. Yet that is exactly what I want to posit. I do so, moreover, essentially as an update to the “innocent” reading I have just cited – which, to be clear, I also endorse, at least up to a point. On the reading that I advance here, al-Ghazālī, as he entered into 1095, had hit an impasse when it came to integrating the work he had done across the various disciplines that purported to impart religious learning (theology, law, philosophy, Sufism). He had produced an assortment of polemical and dialectical treatises, and his appointment probably required of him more work in the same vein. Yet following the masterful Incoherence of the Philosophers and The Infamies of the Bāṭinites he will certainly have felt the onset of diminishing returns, while both the successes and the limitations of The Mean in Belief showed that the investigatory procedures and literary strictures of kalām would serve only to steer him further astray from his true purpose, both when it comes to the doctrines he wished to explore and the methods he wished to employ. The Yardstick of Knowledge and The Touchstone of Speculation were competent efforts at recasting Aristotelian logic but did not provide a platform for further advancement. And whatever al-Ghazālī believed himself to have achieved with putting together The Scale of Action, its greatest successes simultaneously pointed to the ways in which al-Ghazālī was about to hit a wall not only professionally, but also in an existential sense. The troubles that al-Ghazālī faced on a theoretical level were mirrored in his inability to give himself over to spiritual practice, or at least to achieve thereby the hoped-for results. It is at this juncture that al-Ghazālī’s disavowal of his entire previous life begins to make sense, as does his subsequent cleaving to a “revival of the religious sciences.” Where that story takes us is a story for a different day. For now, and by way of conclusion, let me reiterate that none of what I have said is meant to downplay in any shape or form the role played by deliberate craft in al-Ghazālī’s self-presentation. In the Deliverer, and very much including its framing of the events of 1095, it is plain to see

APPROACHING A CRISIS: AL-GHAZĀLĪ IN 488/1095

253

that al-Ghazālī wants to convey to his readership aspects of what is a carefully considered and tightly argued view of human nature – essentially, the central project of his entire mature authorship – through the vehicle of his own life’s story. Al-Ghazālī’s own intellectual and moral trajectory, as retold by the man in the Deliverer from Error, is meant to act as evidence for the persuasiveness of that worldview, the same time that the theory is meant to explain some of what took place in 1095. There can hardly be any doubt that al-Ghazālī’s recounting of his crisis is coloured by these theoretical considerations and that a large amount of after-the-fact rationalization is going on in the pages of the text. However, acknowledging all this by no means necessitates abandoning the notion that al-Ghazālī is telling us about a real crisis that befell him in Baghdad in 1095. Nor do we need to think that al-Ghazālī would not be trying to relate the contours of that crisis to his readership as earnestly as he can (albeit from a perspective that he had developed on it over the subsequent decade). What I have tried to sketch out here – and a sketch is really all it can claim to be – is a way to interpret al-Ghazālī that (1) ties his evolving perspectives on the relation of knowledge to action to an evolving understanding both of (2) what he ought to aim for as a scholar and (3) how he should orient himself as a human being in order to accomplish those goals. Were such an accounting executed in a fuller manner, our understanding of al-Ghazālī’s authorship would be correspondingly enriched. Taneli KUKKONEN Division of Arts and Humanities – New York University Abu Dhabi [email protected]

AVICENNA’S SHIFĀ᾿ AND A COMMENTARY ON IT IN TWENTY-FIVE VOLUMES: THE BODLEIAN POCOCKE COLLECTION AND FAKHR AL-DĪN AL-RĀZĪ* Amos BERTOLACCI Abstract The present paper overviews the Pococke collection of manuscripts of Avicenna’s magnum opus Kitāb al-Shifā᾿ (Book of the Cure/Healing) preserved in the Bodleian Library of Oxford, which originally consisted, in all likelihood, of twenty-five items. This codicological evidence is then confronted with the historical report according to which Fakhr al-Dīn al-Rāzī commented on the Shifā᾿ by means of an exegesis made of twenty-five volumes. The contribution explores whether these two apparently unrelated pieces of evidence – relevant, respectively, for the transmission of Avicenna’s masterpiece in philosophy and the history of Avicennism – may be interconnected, trying to determine which relation, if any, links the twenty-five codices of the Shifā᾿ actually preserved in the Bodelian Pococke collection with the twenty-five volumes of the alleged commentary on the Shifā᾿ by al-Rāzī.

One of the fields of scholarly research related to the philosophy of Avicenna which Jules Janssens has pioneered, with his unique talent for detecting new frontiers of investigation and providing methodological orientation in unexplored areas, is the influence that the Shaykh al-ra᾿īs exerted on the philosophical and theological thought of Fakhr al-Dīn al-Rāzī (d. 606/1210). Janssens has devoted seminal articles to this key juncture in the history of the reception of Avicenna’s thought,1 and the * A draft of this paper was read at the conference “After Avicenna. Online conference on post-Avicennian Islamic philosophy and theology”, University of Jyväskylä, Finland, 2-3 June 2021. I wish to thank the organizer Jari Kaukua and all the partecipants for the precious feedback, as well as Gaia Celli and Silvia Di Vincenzo for the useful previous discussions on this topic. My gratitude also goes to an anonymous reader of the article, from whose comments I have greatly profited. 1. J. JANSSENS, “Ibn Sīnā’s Impact on Faḫr al-Dīn al-Rāzī’s Mabāḥiṯ al-Mašriqiyya, with Particular Regard to the Section Entitled al-Ilāhiyyāt al-maḥḍa: An Essay of Critical Evaluation,” in: Documenti e Studi sulla Tradizione Filosofica Medievale 21 (2010),

256

AMOS BERTOLACCI

following considerations are just a complement to the studies that he has authored or promoted in this field, as a sign of esteem and gratitude for his masterly contribution to Avicennian studies. The present paper consists of three parts. The first is an overview of the Pococke collection of manuscripts of Avicenna’s magnum opus Kitāb al-Shifā’ (Book of the Cure/Healing) preserved in the Bodleian Library of Oxford, which originally consisted, in all likelihood, of twenty-five items. In the second part, I analyze the historical evidence according to which Fakhr al-Dīn al-Rāzī commented on the Shifā’ by means of an exegesis made of twenty-five volumes. In the third and final part, I check whether these two apparently unrelated pieces of evidence – relevant, respectively, for the transmission of Avicenna’s masterpiece in philosophy and the history of Avicennism – may be interconnected, trying to determine which relation, if any, links the twenty-five codices of the Shifā᾿ actually preserved in the Bodelian Pococke collection with the twenty-five volumes of the alleged commentary on the Shifā᾿ by al-Rāzī. 1. The Pococke Collection of Manuscripts of the Shifā᾿ What is called here “Pococke collection” is a series of sixteen extant manuscripts of Avicenna’s Shifā᾿ housed in the Bodleian Library in Oxford, collected by the English orientalist Edward Pococke (16041691) and purchased by the Bodleian Library in 1692 (Oxford, Bodleian Library, Pococke 109-124).2 Among other codices of the Shifā᾿ preserved in the same library, these sixteen items share distinctive codicological features (similar format and an almost identical hand), and follow one another in a clearly discernible sequence. In fact, all the manuscripts of the collection were copied between 601 and 604/12041207 by the copyist mentioned in a few of them (Muḥammad b. ῾Alī pp. 259-285; ID., “Avicennian Elements in Faḫr al-Dīn al-Rāzī’s Discussion of Place, Void and Directions in the al-Mabāḥiṯ al-mašriqiyya,” in: D. N. HASSE – A. BERTOLACCI (eds.), The Arabic, Hebrew and Latin Reception of Avicenna’s Physics and Cosmology, Berlin 2018, pp. 43-63. 2. C. WAKEFIELD, “Arabic Manuscripts in the Bodleian Library: the Seventeenth-Century Collections,” in: G. A. RUSSELL (ed.), The Arabick Interest of the Natural Philosophers in Seventeenth-Century England, Leiden 1994, pp. 128-146 (summary description of the over four hundred manuscripts owned by Pococke, pp. 134-135; the Shifā᾿ among them, p. 135, n. 124).

AVICENNA’S SHIFĀ᾿ AND A COMMENTARY ON IT

257

b. al-Ḥasan), possibly in Aleppo, the place of copying of one of these manuscripts. Ownership statements in twelve of them attest three different owners: Muḥammad b. al-Ḥusayn, Yaḥyā b. Muḥammad b. al-Labūdī, and Ḥasan b. ῾Alī al-Mutaṭabbib. To this sample of codices in Oxford, the manuscript İstanbul, Süleymaniye Kütüphanesi, Fatih 3211, should in all likelihood be added, because of the similar content and format.3 The overall seventeen manuscripts extant in Oxford and Istanbul – together with other manuscripts whose fate is presently unknown4 – formed a larger collection of apparently twentyfive codices, according to the most plausible reconstruction of their orginal extent, meant to cover cumulatively, one after the other, the entire Shifā᾿.5 The relevant data regarding the Pococke collection are summarized in the following table:6 3. On ms. Fatih 3211, see K. HIRSCHLER, Medieval Damascus. Plurality and Diversity in an Arabic Library. The Ashrafiya Library Catalogue, Edinburgh 2016, p. 27 and n. 35. 4. As in the case of the manuscript Istanbul Fatih 3211, the possibility of retrieving the still missing copies of this collection in other libraries worldwide cannot be excluded. 5. The Pococke collection was first analyzed in 1954 by Y. Mahdavī, who pointed out its importance as first known complete copy of the Shifā᾿ (Y. MAHDAVĪ, Fihrist-i nuskhahā-yi muṣannafāt-i Ibn-i Sīnā, Tehran 1333Hš/1954, p. 170). Mahdavi was pioneer also in suggesting that the manuscript Süleymaniye Kütüphanesi, Fatih 3211, was part of this collection (Mahdavi presents this manuscript as containing the Mathematics, without any further specification, but places it correctly at the beginning of the mathematical part as item no. 19). Mahdavi’s account of the issue, of exceptional level for the time in which it was written, remains fundamental in many respects: see the developments in IBN SĪNĀ, Al-Shifā᾿, al-Ṭabī῾iyyāt, I: al-Samā῾ al-ṭabī῾ī, ed. D. AL-YĀSĪN, Beirut 1966, p. 24 (in which however the name of the copyist, Muḥammad b. ῾Alī b. al-Ḥasan, is confounded with that of one of the owners, Muḥammad b. al-Ḥusayn), and D. C. REISMAN, “Mahdavi Supplement. Šifā᾿ Manuscripts,” Handout distributed at the International Conference “The Manuscript Tradition of Avicenna’s Kitāb al-Shifā’: The Current State of Research and Future Prospects,” Scuola Normale Superiore di Pisa, Pisa, 22-24 September 2010. 6. Legenda: sh = Shifā᾿ ; m = Manṭiq (= logic) of the Shifā᾿ ; ṭ = Ṭabī῾iyyāt (= natural philosophy) of the Shifā᾿ ; r = Riyāḍiyyāt (= mathematics) of the Shifā᾿; i = Ilāhiyyāt (= metaphysics) of the Shifā᾿. Within Manṭiq, Ṭabī῾iyyāt, and Riyāḍiyyāt, the sections (funūn) are designed by lower-case Roman numerals (i, ii, iii, etc.) written as deponents (ex. gr.: mv = section five of Manṭiq/logic; ṭii-viii = sections ii-viii of Ṭabī῾iyyāt/natural philosophy; ri-ii, iv = sections i, ii, and iv of Riyāḍiyyāt/mathematics). Within Ilāhiyyāt, Manṭiq, Ṭabī῾iyyāt, and Riyāḍiyyāt, the treatises (maqālāt) and chapters (fuṣūl) are designed respectively by upper-case Roman numerals (I, II, III, etc.) and Arabic numerals (1, 2, 3, etc.) (ex. gr.: Ilāhiyyāt I.1-3 = chapters 1-3 of treatise I of Ilāhiyyāt; ṭi.I.1-2 = chapters 1-2 of treatise I of section i of Ṭabī῾iyyāt/natural philosophy). Further information is available at https://www.avicennaproject.eu/ (“Other Resources” – “All Šifā᾿ Manuscripts” – “Pococke Collection”).

?

Pococke 124 (200 foll.)

Pococke 123 (243 foll.)

Pococke 122 (249 foll.)

?

Pococke 121 (294 foll.)

Pococke 120 (259 foll.)

?

?

Pococke 119 (216 foll.)

?

Pococke 118 (201 foll.)

Pococke 115 (255 foll.)

1

2

3

4

5

6

7

8

9

10

11

12

13

Ms and folia

Nr

ṭi.IV.8-iii (acephalous)

ṭi.II.10-IV.7 + fragments of ṭvi.II.3 and ṭviii.XII.15

ṭi.I-II.9

mviii.II.4-ix

 ???-mviii.II.3

mvi.I.7-???

mv.II.10-vi.I.6

miv.IX.4-v.II.9

miv.V-IX.3

miv.I-IV

mii.VI.2-iii

mii.II.2-VI.1

mi-ii.II.1

Content

Owner(s)

Muḥammad b. [no frontispiece, usual place ῾Alī b. al-Ḥasan of ownership notes in the other mss]

Muḥammad b. al-Ḥusayn

Muḥammad b. al-Ḥusayn

Muḥammad b. al-Ḥusayn

[no frontispiece, usual place of ownership notes in the other mss]

Muḥammad b. Muḥammad b. al-Ḥusayn ῾Alī b. al-Ḥasan

Muḥammad b. ῾Alī b. al-Ḥasan

Muḥammad b. Muḥammad b. al-Ḥusayn; ῾Alī b. al-Ḥasan Yaḥyā b. Muḥammad b. al-Labūdī

Copyist 11-20 Ramaḍān 601H

Dating

22 Sha῾bān 603H

Rajab 603H

11-20 Rabī῾ II 603H

23 Dhū l-ḥijja 602H

28 Rajab 602H

14 Dhū l-ḥijja 601H

Ḥalab 21-30 Shawwāl (Aleppo) 601H

Place

14th part (13th corr. marg. alia manus) of sh

12th part of sh; 2nd part of ṭ; 13th part follows

end of m

7th part of sh; 8th part follows

6th part of m; 7th part follows

4th part of m

3rd part of sh

2nd part (juz᾿) of m; 3rd part follows

Serial Nr in Ms

258 AMOS BERTOLACCI

Pococke 116 (257 foll.)

?

Pococke 114 (128 foll.)

Pococke 113 (209 foll.)

?

İstanbul, Fatih 3211 (253 foll.)

Pococke 109 (219 foll.)

Pococke 112 (182 foll.)

Pococke 111 (253 foll.)

Oxford, Pococke 117 (188 foll.)

Pococke 110 (189 foll.)

?

14

15

16

17

18

19

20

21

22

23

24

25

iVIII.7-X

iV-VIII.6

iI-IV

riv.V-XII

riv.I-IV

rii-iii

ri

ṭviii.XIII-XIX

ṭviii.VII-XII (mutilous at the end)

ṭviii.I-VI

ṭvi.II.3-vii

ṭiv-vi.II.2

21-30 Rabī῾ I [= 21-29 Rabī῾ II ?] 604H 21-30 Jumādā I 604H

Muḥammad b. al-Ḥusayn

1-10 Rabī῾ II [= Rabī῾ I ?] 604H

Muḥammad b. al-Ḥusayn

Muḥammad b. al-Ḥusayn

1-10 Ṣafar 604H 21st part of sh

Muḥammad b. al-Ḥusayn; Yaḥyā b. Muḥammad b. al-Labūdī; Ḥasan b. ῾Alī al-Mutaṭabbib

24th volume of sh

23rd part of sh

22nd volume of sh

20th part of sh

11-20 Muḥarram 604H

[no frontispiece, usual place of ownership notes in the other mss]

19th part of sh

[no colophon]

16th volume (mujallada) of sh

4th part of ṭ; 14th part of sh; 15th part follows

3 Muḥarram 604H

9 Dhū l-Qa῾da 603H

11-20 Ramaḍān 603H

Muḥammad b. al-Ḥusayn; Yaḥyā b. Muḥammad b. al-Labūdī

Muḥammad b. al-Ḥusayn

Muḥammad b. Muḥammad b. al-Ḥusayn ῾Alī b. al-Ḥasan

AVICENNA’S SHIFĀ᾿ AND A COMMENTARY ON IT

259

260

AMOS BERTOLACCI

The peculiar catalogue numbers of the items of the collection – sequential, in a reverse order, for the first seven extant manuscripts, disordered for the remaining ones – are not at stake here. What interests most is the content of the manuscripts: the twenty-five elements of the Pococke collection in its original status consisted, according to the most plausible reconstruction, of ten manuscripts of logic, eight of natural philosophy, four of mathematics, and three of metaphysics. In this sequence, one after the other, they covered the entire scope of the Shifā᾿. Among the various types of copies of the Shifā᾿ presently known, the Pococke collection can be classified as a unique instance of a copy which is both “cumulative” (i.e. it contains all the Shifā᾿, in sequential portions) and “segmented” (that is, the portions in question do not correspond to the main articulations that Avicenna assigns to the work). In this kind of “cumulative segmented” copy, as I have called it elsewhere,7 the entire Shifā᾿ is copied into a comprehensive series of interconnected manuscripts, by means, however, of a partition that mirrors imprecisely the structure of the work: no single manuscript contains any of the four main “parts” of the Shifā᾿ (logic, natural philosophy, mathematics, and metaphysics) in its entirety, and not always do manuscripts squarely correspond to entire “sections” of these parts (one or more of the nine sections of logic, the eight of natural philosophy, the four of mathematics, and the single one of metaphysics).8 More often, termini of the process of copying are rather “treatises” of the single sections, and even “chapters” of the single treatises.9 In this way, the copying of the manuscripts often stops in medias res, with a certain treatise of a certain section of the work, or with a certain chapter of a certain treatise, and continues in the following manuscript 7. A. BERTOLACCI, “Avicenna’s Kitāb al-Šifā᾿ (Book of the Cure/Healing): The Manuscripts Preserved in Turkey and Their Significance,” in: J. JABBOUR (ed.), The Reception of the Classical Arabic Philosophy in the Ottoman Empire. Proceedings of the Workshop of the International Associated Laboratory “Philosophie dans l’aire ottomane”, Istanbul, 2-4 November 2015, Mélanges de l’Université Saint-Joseph 67 (2017-18), pp. 265-304 (esp. pp. 281-282). 8. Only items 19 and 20 of the collection contain precisely, respectively, the first section of mathematics, and the second and third sections of it. 9. For instance, the first extant manuscript of the collection (Pococke 124) starts abruptly with the second chapter of the second treatise of the second section of logic (ii. II.2) and ends, no less abruptly, with the first chapter of the sixth treatise of it (ii.VI.1).

AVICENNA’S SHIFĀ᾿ AND A COMMENTARY ON IT

261

from the subsequent treatise or chapter. Thus, basic units of the work like sections or treatises, which, as a rule, are integrally copied in other manuscripts of the Shifā᾿, in the Pococke collection are often split into distinct codices. Because of this fragmentation, the individual manuscripts of the Pococke collection contain in most cases only continuous “segments” of the work. Whereas “cumulativeness” is not a specificity of the Pococke collection (also other series of manuscripts elsewhere in the world contain cumulatively the entire Shifa᾿), “segmentation” no doubt is.10 The former feature can be explained by the massive extent of the Shifā᾿, which made it hard to copy the entire work into a single codex. The latter feature, on the other hand, rather than a deliberate choice by the copyist, was probably a necessity imposed by the dimensions of the material support that he used for the copy, in other words by the length of the single bundles of paper that he had at his disposal for the cumulative copy. It is difficult to assess whether the exemplar of the Shifā᾿ from which the Pococke collection stems was a single whole manuscript of Avicenna’s encyclopedia,11 or consisted itself of more than one manuscript.12 One-volume copies of the Shifā᾿ are attested coevally (VIIth/XIIIth c.) in the manuscript tradition of Avicenna’s work.13 Since both “part” (juz᾿ ) and “volume” (mujallad) mean “quire” in Arabic codicology, the possibility of a partition of the ancestor of the Pococke collection into twenty-five quires of large dimensions cannot be excluded, although the variable length of the 10. In BERTOLACCI, “Avicenna’s Kitāb al-Šifā᾿ (Book of the Cure/Healing): The Manuscripts Preserved in Turkey and Their Significance,” pp. 280-281, I have distinguished “cumulative sorted” copies of the Shifā᾿, instantiated by various manuscripts worldwide, which cumulate, one after the other, precise portions of the work, and “cumulative segmented” copies, whose only instance known so far is the Pococke collection. 11. A. NICOLL, Bibliotecæ Bodleianæ Codicum Manuscriptorum Orientalium Catalogi Partis Secundæ, Volume: II, Oxford 1821, p. 582 (“Id vero praeterea notandum est, ad unum idemque exemplar pertinere codd. Poc. 109-124”). 12. In the manuscript of the Shifā᾿ Kabul, Afghan National Archive 2295, the copyist of the part on logic marks the end of the “volumes” (mujalladāt) of the antigraph (aṣl), mentioning in one case also one of its “fascicles” (dafātīr). However, contrary to what happens in the Pococke collection, this partition is precise (i.e. the “volumes” correspond to the main sections of logic) and remains within the framework of a single manuscript. 13. For ms. Cairo, Maktabat al-Azhar al-Sharīf, Bekhīt 331 falsafa (khuṣūṣiyya), 44988 (῾umūmiyya) (VIIth/XIIIth c.) as first known “compact” copy of the Shifā᾿, see BERTOLACCI, “Avicenna’s Kitāb al-Šifā᾿ (Book of the Cure/Healing): The Manuscripts Preserved in Turkey and Their Significance,” p. 280.

262

AMOS BERTOLACCI

single items of the collection (from a mininum of 128 folia in ms. Pococke 114, to a maximum of 259 folia in ms. Pococke 120) militates against this hypothesis.14 For sure, no other known copy of the Shifā᾿ consists of so many individual manuscripts as the Pococke collection does. That the collection was originally made of twenty-five volumes can be inferred from its last extant item. The explicit of ms. Pococke 110 states that the manuscript contains the twenty-fourth “volume” (mujallada) of the Shifā᾿,15 a numerical indication confirmed by the medallion at the beginning of the manuscript, which embeds the phrase “Twentyfourth part (juz᾿) of the Shifā᾿”.16 Since ms. Pococke 110 contains almost three treatises of the Ilāhiyyāt (V-VIII, 6) and since a bit more than two treatises are needed in order to reach the end of the work (VIII, 7, IX-X), it is reasonable to suppose that the twenty-fifth volume of the series was long enough to reach the end of the Ilāhiyyāt and, with it, of the entire Shifā᾿. The first and last volumes of the overall series might be missing because of their position, as sometimes the first and last folia in a manuscript get lost. 14. Also the decorations shaped as medallions (shamsa) in all the items of the Pococke collection that are integrally preserved (i.e. that are not mutilous at the beginning), with indication of the number of the section of the Shifā᾿ at hand and the occasional description of this section as “part” (juz᾿), are problematic in this regard, since they provide independent status to material units which are bare segments of the Shifā᾿ and, if meant as quires, should be bundled together rather than remaining apart. The date of these decorations is uncertain: medaillons of similar shape indicating the material partition of the copied work can be found in coeval manuscripts (see, for instance, ms. Istanbul, Damād 916, dated 597H). 15. “ the end of the 24th volume of the Kitāb al-Shifā᾿. To God belong the praise and the grace. May His prayers and His peace be upon our lord Muḥammad, his family, and his companions” (ākhiru l-mujalladati l-rābi῾ati wa-l-῾ishrīna min Kitāb al-Shifā᾿ wa-li-llāhi l-ḥamdu wa-l-minnatu wa-ṣalawātuhu ῾alā sayyidinā Muḥammad wa-ālihi wa-ṣaḥbihi wa-salāmuhu), fol. 188v. 16. In the titlepiece, this phrase ranges over three lines (al-juz᾿u / al-rābi῾u wa-l῾ishrūna / min al-Shifā᾿). In a guard page at the beginning of the manuscript (fol. 2r), a hand different from the scribe’s has written the note “[The chapter so-and-so] follows it in [volume/part] 25” (yatlūhu fī 25 [faṣl/al-faṣl ...]). A similar kind of notation recurs in the guard pages of mss. Pococke 123, 122, 120, 118, 116, and 111 (items 3, 4, 7, 12, 14, 22 in the table above), where it gives the number of the following section of the collection (yatlūhu fī ...), accompanied by the title of the chapter with which the following section starts (faṣl/al-faṣl fī ...). The same verb yatlūhu in the explicits of mss. Pococke 124, 121, 120, 118, and 116 (items 2, 6, 7, 12, 14 in the table above) introduces the number of the following section of the collection.

AVICENNA’S SHIFĀ᾿ AND A COMMENTARY ON IT

263

The collection provides evidence of a single copyist, Muḥammad b. ῾Alī b. al-Ḥasan, whose name lends itself to different identifications.17 Difficult to identify is also the owner of the collection most frequently referred to (Muḥammad b. al-Ḥusayn, mentioned in ten manuscripts), whereas the two further owners Yaḥyā b. Muḥammad b. al-Labūdī (607-670/1210-1271, mentioned in three manuscripts), and Ḥasan b. ῾Alī al-Mutaṭabbib (mentioned in one manuscript), are less shadowy figures. Ibn al-Labūdī is the owner of the collection we are best informed about, and his biography sheds further light on our issue, as we are going to see in section 3.2. Likewise, according to a possible indentification with authors known otherwise, the third owner, Ḥasan b. ῾Alī al-Mutaṭabbib, may be situated in Mamluk territories a century after Ibn al-Labūdī. This differentiated ownership, together with the migration of one of the manuscripts of the collection to Istanbul, points at dynamics of dispersion. Edward Pococke, however, was fortunate enough to find the main bulk of the collection, possibly in the same place, during one of his stays in the Near East. The Syrian city of Aleppo is specified in one of the manuscripts of the collection as the place of copying.18 Edward Pococke was in Aleppo between 1630 and 1635, where he started purchasing Arabic and other oriental manuscripts under the impulse of William Laud, chancellor of Oxford University.19 Aleppo is therefore a very promising candidate to be the city were the collection resided until Pococke’s times. A shortly later stay of him in the Near East (1637-1639), similarly devoted to collecting manuscripts, was in Istanbul, but, though at distance, also then Pococke maintained contacts with scholars in Aleppo for the acquisition of codices. This being the case, although it remains to be 17. The ḥadīth transmitter Abū al-῾Alā᾿ Muḥammad b. ῾Alī b. al-Ḥasan b. al-Ra᾿s al-Ṣūfī al-Yamanī al-Muwallad al-Baghdādī, with ṣūfī affiliations, is recorded as dying in 609H (no place of burial specified) in al-Mukhtaṣar al-muḥtāj ilayhi min Tarīkh Ibn al-Dubaythī of al-Dhahabī (d. 1348 CE). See http://hadithtransmitters.hawramani.com/?p=136574#28098f. 18. Aleppo and its hinterland in the Ottoman period / Alep et sa province à l’époque ottomane, ed. S. WINTER – M. ADE, Leiden / Boston 2019. 19. P. M. HOLT, “An Oxford Arabist: Edward Pococke (1604-91),” in: ID., Studies in the History of the Near East, London 1973, pp. 3-26 (esp. pp. 5-7); ID. “Edward Pococke (1604-1691), the First Laudian Professor of Arabic at Oxford,” in: Oxoniensia 56 (1991), pp. 119-130 (esp. pp. 121-122); WAKEFIELD, “Arabic Manuscripts in the Bodleian Library,” p. 134.

264

AMOS BERTOLACCI

assessed whether the collection was acquired by Pococke during his first or his second stay in the Levant, Aleppo remains the most likely city where this acquisition might have taken place. Finally, another important aspect of the collection is the names according to which the extant manuscripts are named: many of them are called alternatively “parts” (ajzā᾿, sg. juz᾿ ) or “volumes” (mujalladāt, sg. mujallad), in the sense of “material segments” specified above. 2. Al-Ṣafadī About al-Rāzī’s Commentary on the Shifā᾿ The fundamental piece of evidence for the second side of the present story is a short report transmitted by the Mamluk historian Ṣalāḥ al-Dīn Khalīl b. Aybak al-Ṣafadī (d. 764/1363) in part IV of his biographical dictionary al-Wāfī bi-l-wafayāt (The Completion of Obituaries), in the entry devoted to Fakhr al-Dīn al-Rāzī.20 This report figures in the second part of the entry on al-Rāzī, in the context of an enumeration of this latter’s works. This report is reproduced, emended, and critically evaluated in the standard bibliography of al-Rāzī’s works by Muḥammad Ṣ. al-Zarkān of 1963,21 from which I quote: Text 1: AL-ṢAFADĪ, al-Wāfī bi-l-wafayāt, IV, in AL-ZARKĀN, Fakhr al-Dīn al-Rāzī, p. 124:22 [a] The shaykh Rukn al-Dīn b. al-Qarī῾ used to say that he [i.e al-Rāzī]23 commented (sharaḥa) on the Shifā᾿ of Ibn Sīnā: [b] he used to claim that among the books of his father in the Maghreb24 there was the volume of [al-Rāzī’s] commentary (mujallad sharḥ) on the Ilāhiyyāt of the Shifā᾿. [c] If this is correct [i.e. if al-Rāzī indeed commented on the Shifā᾿], then it [i.e al-Rāzī’s commentary on the whole Shifā᾿] is at least in twentyfive volumes (fa-aqalla mā yakūnu fī khamsin wa-῾ishrūna mujalladatan). 20. AL-ṢAFADĪ, al-Wāfī bi-l-wafayāt, ed. A. AL-ARNĀ᾿ŪṬ – T. MUṢṬAFĀ, Beirut 2000, vol. IV, pp. 175-186. 21. M. Ṣ. AL-ZARKĀN “Mu᾿allafāt al-Rāzī,” in: Id., Fakhr al-Dīn al-Rāzī wa-ārā᾿uhu al-kalāmiyya wa-l-falsafiyya, Beirut 1963, pp. 56-164 (work no. 114, p. 124). 22. This passage can be found also in the aforementioned edition of al-Wāfī bi-lwafayāt published at Beirut in 2000, in vol. IV, p. 180. 14-16. In the previous edition of part IV of the work by S. DEDERING, Damascus 1959, p. 256. 4-5, on the other hand, the passage is flawed by the omission of a sentence by homoioteleuton. 23. ay al-Rāzī in the text reported by al-Zarkān is absent in the consulted editions of al-Wāfī bi-l-wafayāt and it looks an addition. 24. bi-l-ḍarb ed. Beirut 2000; om. ed. Damascus 1959; bi-l-maghrib al-Zarkān.

AVICENNA’S SHIFĀ᾿ AND A COMMENTARY ON IT

265

We don’t know much about the protagonist of sections [a] and [b] of this text: the various forms according to which the crucial element of his name is transmitted in historical sources (“Ibn al-Qarī῾” in Text 1; “Ibn al-Qūba῾” in the consulted editions of al-Wāfī bi-l-wafayāt)25 are indicative of a more general lack of information. What Ibn al-Qarī῾ (or: Ibn al-Qūba῾) contends in section [a], namely that al-Rāzī commented on Avicenna’s Shifā᾿, is argued by him on the basis of section [b], i.e. on account of the presence of al-Rāzī’s commentary on the Ilāhiyyāt (Science of Divine Things) of the Shifā᾿ in his father’s library. In other words, the alleged commentary by al-Rāzī on the Shifā᾿ that Ibn al-Qarī῾ found in his father’s library, consisted primarily, if not exclusively, of a commentary on the metaphysical part of Avicenna’s summa. In section [c] al-Ṣafadī himself takes the stage and, assuming that the information by al-Qarī῾ in sections [a] and [b] is trustworthy, he speculates about the length of al-Rāzī’s overall commentary on the Shifā᾿, supposing that it was made of at least twenty-five volumes.26 The information about al-Qarī῾ that we can obtain from other historical sources fits with the picture conveyed by Text 1, clarifying it in important respects. Another Mamluk historian (but also commentator of the Qur᾿ān and collector of ḥadīths), Abū l-Fidā᾿ Ismā῾īl b. ῾Umar b. Kathīr (d. 774/1373), in his universal history of the world from the creation to the end of time, al-Bidāya wa-l-nihāya (The Beginning and the End), provides relevant information on our man, called “Ibn al-Qūba῾” in the edition from which I am quoting: Text 2: IBN KATHĪR, al-Bidāya wa-l-nihāya, ed. ῾A. AL-TURKĪ, Cairo 1998, vol. XVIII, pp. 407. 9 - 408. 2 25. The spelling al-Qūba῾ is made explicit in the consulted edition of IBN AL-KATHĪR, Bidāya, vol. XVIII, p. 407 (see below, Text 2). 26. Also in other cases, the catalogue of al-Rāzī’s works in the dedicated entry of al-Wāfī bi-l-wafayāt provides information on the number of volumes (mujalladāt) the single works are made of (occasionally it even clarifies the extent of these volumes, i.e. whether they are big or small). It is therefore arguable that the number of twenty-five volumes in [c] refers to the estimated length of al-Rāzī’s commentary on the Shifā᾿ and not of his commentary on the Ilāhiyyāt: in this latter case, the number of twenty-five volumes would be exorbitant in itself, contrary to what Ibn al-Qarī῾ reports, and in contrast with al-Ṣafadī’s usual practice of providing information on the volumes of the works of al-Rāzī in their entirety, not of parts of them. Also al-Zarkān classifies Text 1 under the rubric “commentary on the Shifā᾿”.

266

AMOS BERTOLACCI

[Among the notables who died in this year, i.e. 738H, there was] the shaykh imām ῾allāma Rukn al-Dīn b. al-Qūba῾ Abū ῾Abd Allāh Muḥammad b. Muḥammad b. ῾Abd al-Raḥmān b. Yūsuf b. ῾Abd al-Raḥmān b. ῾Abd al-Jalīl al-Qurashī al-Ḥāshimī al-Ja῾farī al-Tūnisī al-Mālikī, known as Ibn al-Qūba῾. He was one of the excellent notables and intelligent lords, and one of those who embraced the many arts and the copious sciences in religion, poetry, and medicine. He was teacher at the Mankūtamuriyya [school] and held a position at the Manṣūrī hospital. He died in that year, in the early morning of the 17th Dhū l-Ḥijja [14th July 1338], at the age of seventy-four. He left abundant money and furniture, which the public treasury inherited.

Without entering into details, Ibn Kathīr’s report is important in three respects. First, the nisba al-Tūnisī (“the Tunisian”) in the full name of Ibn al-Qarī῾ (or Ibn al-Qūba῾) corroborates the setting of his father’s library in the Maghreb that we learn from Text 1. Second, Text 2 attests that Ibn al-Qarī῾ was active in two prime cultural institutions in Cairo, the Mankūtamuriyya school and the Manṣūrī hospital, being therefore a prime exponent of the Mamluk capital’s intelligentsia.27 Third, Text 2 insists on the medical interests of Ibn al-Qarī῾, 27. A Muḥammad Rukn al-Dīn b. al-Qarī῾, in all likelihood the same as al-Ṣafadī’s informant, is mentioned in historical sources (AL-MAQRĪZĪ, al-Sulūk li-ma῾rifat duwal al-mulūk, ed. M. M. ZIYĀDA, Cairo 1939-1958, vol. II/2, p. 449; IBN ḤAJAR AL-῾ASQALĀNĪ, al-Durar al-kāmina fī a῾yān al-mī῾a al-thāmina, éd. A. F. AL-MAZĪDĪ, Beyrouth 1996, vol. I, no. 692, pp. 159-161) as taking part in a scholarly dispute in Cairo in 738/1337, the same year of his death (which occurred in the last month of that year) according to Text 1. See M. EYCHENNE, Liens personnels, clientélisme et réseaux de pouvoir dans le sultanat mamelouk (milieu XIIIe - fin XIVe siècle), Beyrouth / Damas 2013, p. 124 (consulted on May 21 2020 at https://books.openedition.org/ifpo/3908?lang=en: “En 738/1337, l’émir Sanǧar al-Ǧāwlī, en qualité d’administrateur des biens waqf du défunt sultan al-Manṣūr Qalāwūn (nāẓir awqāf al-Manṣūriyya), voulut nommer Šihāb al-Dīn Aḥmad al-῾Asǧadī comme professeur de ḥadīṯ de la Qubba al-Manṣūriyya au Caire après la mort du titulaire de la charge. Mais, les cadis et l’ensemble des ῾ulamā’ s’opposèrent à sa nomination, arguant de son incompétence et rédigèrent une plainte en ce sens à l’intention du sultan. Lorsque le placet fut lu au sultan à la Dār al-῾Adl, celui-ci questionna les cadis au sujet du professeur. Le qāḍī l-quḍāt šāfi῾ī Ibn Ǧamā῾a le critiqua tandis que l’émir Sanǧar al-Ǧāwlī prit sa défense et fit son éloge. Le sultan al-Nāṣir Muḥammad ordonna alors de réunir un maǧlis afin de statuer sur son cas. Les cadis et de nombreux fuqahā’ se réunirent à la Madrasa al-Manṣūriyya. Ils firent tous un mauvais accueil à Sanǧar al-Ǧāwlī pour amoindrir les qualités de Aḥmad al-῾Asǧadī. Rukn al-Dīn Muḥammad Ibn al-Qarī῾ accusa Aḥmad al-῾Asǧadī d’avoir fait trois erreurs de prononciation lors de sa lecture de la Fātiḥa, la sourate d’ouverture du Coran. Mais pour sa part, le qāḍī l-quḍāt ḥanafī Ḥasan al-Ġūrī apporta son soutien à Aḥmad al-῾Asǧadī et fit son éloge en déclarant attester de ses compétences pour remplir la charge à laquelle il avait été nommé. Les insultes entre lui et Ibn Ǧamā῾a se mirent à fuser et ils se séparèrent sur ces entrefaites. Ḥasan al-Ġūrī informa de

AVICENNA’S SHIFĀ᾿ AND A COMMENTARY ON IT

267

both theoretical and practical. From Text 1 and Text 2 Ibn al-Qarī῾ turns out to instantiate the interesting case of a scholar who is, at the same time, an amateur of the philosophy of Avicenna and a professional physician, as an attestation of the convergence of philosophy and medicine, on the footsteps of the Shaykh al-ra᾿īs, in the Avicennian tradition.28 It would be important, of course, to assess when Ibn al-Qarī῾ moved from Tunisia to Egypt, if, as it seems, he was born in Tunisia, as well as the nature and contents of his father’s library in the Maghreb. On the basis of his age at death as stated in Text 2, Ibn al-Qarī῾ was born approximately in 665/1266-1267. If his report about his father’s library is a memory of his youth, we may guess that the volume which he takes to contain al-Rāzī’s commentary on the Ilāhiyyāt must have entered the library of his father towards the end of the VIIth/ XIIIth century at the latest, a few decades after al-Rāzī’s death in 606/ 1210. The fundamental element of Text 1 for our purposes is section [c]. Every element of it is significant. The initial sentence “If this is correct” indicates that al-Ṣafadī has no means to verify the core of al-Qarī῾’s report, namely that he had no access to al-Rāzī’s commentary on the Shifā᾿. Of crucial importance is al-Ṣafadī’s suggestion about the supposed extent of al-Rāzī’s overall commentary on the Shifā᾿ (“it is ... in twenty-five volumes”). The formula “it is in” (yakūnu/takūnu fī), followed by the indication of the number of volumes, is typical of the la situation l’émir Tāǧār al-Dawādār pour qu’il apporte son soutien à Aḥmad al-῾Asǧadī et qu’il atteste de ses compétences auprès du sultan. Al-Nāṣir Muḥammad convoqua Ibn Ǧamā῾a et Sanǧar al-Ǧāwlī à la Dār al-῾Adl et les questionna sur le déroulement du maǧlis. L’émir et le cadi se mirent à se disputer et à se contredire l’un l’autre. Le sultan trancha finalement en faveur d’Ibn Ǧamā῾a et interdit à Aḥmad al-῾Asǧadī d’occuper la charge de professeur. L’émir Sanǧar al-Ǧāwlī, furieux de la décision du sultan, voulut être relevé de sa charge d’administrateur (nāẓir) du Māristān al-Manṣūrī mais les émirs le mirent en garde contre les conséquences d’une telle décision. Finalement il resta en fonction, et, selon Ibn Ḥaǧar al-῾Asqalānī, Abū Ḥayyān fut nommé par Sanǧar al-Ǧāwlī à la place de Aḥmad al-῾Asǧadī.” 28. See, on this, 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 2006, pp. 371-422 (esp. pp. 383-392); A. BERTOLACCI, “Metaphysics, Elemental Transformation, Medicine: A Specimen of Avicenna’s System of Thought in Ms. Escorial 621,” in: Oriens 47.3-4 (2019), pp. 389-417.

268

AMOS BERTOLACCI

specifications that al-Ṣafadī adds to the titles of the other works of al-Rāzī in the same entry of al-Wāfī bi-l-wafayāt, with occasional hints at their length, i.e. of whether they are long or short. The key term mujallad or mujallada in the above text is therefore translated as “volume”, in the sense of “tome”, to designate a codicological unit of a work which results from the material process of copying and is independent from its content, being often called also “part” (juz᾿) in this sense.29 The number twenty-five of these “volumes” in our text comes, however, to a surprise, both because it is determinate and precise about a work which al-Ṣafadī ostensibly does not know firsthand, and because it proposes an articulation of al-Rāzī’s commentary incongruous with the standard partition of the Shifā᾿, which notoriously consists of twenty-two, rather than twenty-five, units. Puzzling is also the adverbial locution “at least” that qualifies the number twenty-five (“at least in twenty-five volumes”). All these indications are baffling. In general, the information that Text 1 provides is mainly negative. From this text we know, first of all, that the supposed commentary by al-Rāzī did not circulate in the first half of the VIIIth/XIVth century in Cairo, where Ibn al-Qarī῾ was active, since this latter gets news about this commentary from elsewhere, i.e. from his father’s library in Tunisia. Nor this commentary was accessible in the same century in Egypt and Syria, where al-Ṣafadī established professional and personal links with the major scholars and writers of the time, since he is not able to verify personally Ibn al-Qarī῾’s allegation, which he dubitatively reports and about which he can only advance a guess. On account of the uncertain evidence provided by Text 1 and the absence of any other attestation, al-Zarkān records al-Rāzī’s commentary on the Shifā᾿ among the spurious works of al-Rāzī, followed in this by the majority of the Razian scholars to date, who rarely take the issue into consideration.30 29. See G. HUMBERT, “Le ǧuz᾿ dans les manuscrits arabes médiévaux,” in: F. DÉROCHE – F. RICHARD (eds.), Scribes et manuscrits du Moyen-Orient, Paris 1997, pp. 77-86 (esp. p. 78). 30. Among the few scholars who spend some words on this commentary, see Y. T. LANGERMANN, “Criticism of Authority in the Writings of Moses Maimonides and Fakhr al-Dīn al-Rāzī,” in: Early Science and Medicine 7.3 (2002), pp. 255-275 (esp. p. 266, n. 33); A. DHANANI, “The Impact of Ibn Sīnā’s Critique of Atomism on Subsequent

AVICENNA’S SHIFĀ᾿ AND A COMMENTARY ON IT

269

The crucial issue of the format in twenty-five volumes of al-Rāzī’s commentary on the Shifā᾿ is relegated to insignificance by a painstaking interpreter of Text 1 like al-Zarkān himself: Text 3: AL-ZARKĀN, Fakhr al-Dīn al-Rāzī, p. 124, n. 1: What al-Ṣafadī says about the estimated extent of the commentary [by al-Rāzī] contains an evident exaggeration, since the Ilāhiyyāt of the Shifā᾿ is not a twenty-fifth (lit.: one of twenty-five parts, juz᾿ ) of the original text (aṣl) of the book (i.e. of the Shifā᾿); rather, it can be estimated as one fifth (lit.: five [parts]) [of it].

Al-Zarkān’s estimate of the length of the Ilāhiyyāt as about one fifth of the overall length of the Shifā᾿ is probably wrong; al-Ṣafadī would be equally wrong if he really had taken the Ilāhiyyāt to be one twenty-fifth of the whole.31 Mathematical speculations apart, before dismissing al-Ṣafadī’s count of the volumes of al-Rāzī’s commentary on the Shifā᾿ as an “evident exaggeration” (mubālagha ẓāhira), we should wonder whether Ṣafadī’s appraisal may depend on a consideration different from the one that al-Zarkān ascribes to him, namely the sheer proportion of the textual extents of, respectively, the Ilāhiyyāt and the Shifā᾿ expounded in Text 3. How could a scholar like al-Ṣafadī – an “all-round humanist”, as Franz Rosenthal calls him,32 and a knowledgeable informant and reader of Avicenna’s influential summa Kitāb al-Ishārāt wa-l-tanbīhāt (Book of Pointers and Reminders), as Gerard Endress has documented33 – have made such a gross mistake about the distribution of text within Avicenna’s masterpiece? Moreover, why precisely a proportion of one to twenty-five is at stake, rather Kalām Discussions of Atomism,” in: Arabic Sciences and Philosophy 25 (2015), pp. 79-104 (esp. p. 91 and n. 50); S. DI VINCENZO, “Early Exegetical Practice on Avicenna’s Šifā᾿: Faḫr al-Dīn al-Rāzī’s Marginalia to Logic,” in: Arabic Sciences and Philosophy 28.1 (2018), pp. 31-66 (esp. p. 47, n. 61). For a survey of al-Rāzī’s philosophical and theological production, see A. SHIHADEH, The Teleological Ethics of Fakhr al-Dīn al-Rāzī, Leiden 2006, and F. GRIFFEL, “On Fakhr al-Dīn al-Rāzī’s Life and the Patronage He Received,” in: Journal of Islamic Studies 18 (2007), pp. 313-344. 31. In the Cairo standard printing of Avicenna’s encyclopedia, the Ilāhiyyāt covers 453 pages, in front of the over 5240 pages of the entire work, although the typesetting of the various sections of the Shifā᾿ in the Cairo edition is variable. Only a precise word count of Avicenna’s summa and of its various parts can clarify the issue. 32. F. ROSENTHAL, “al-Ṣafadī,” in: P. BEARMAN e.a. (eds.), Encyclopaedia of Islam, Second Edition, vol. VIII, Leiden 1995, pp. 759-760. 33. ENDRESS, “Reading Avicenna in the Madrasa,” p. 406 and n. 10; p. 411 and n. 130; p. 415.

270

AMOS BERTOLACCI

than, for example, a proportion of one to twenty or of one to thirty, in al-Ṣafadī’s guesses on the issue? In fact, it is hard to believe that in Text 1 al-Ṣafadī is reasoning as al-Zarkān takes him to be doing. In al-Zarkān’s view, his reasoning would be: as the Ilāhiyyāt is a twenty-fifth of the Shifā᾿, so al-Rāzī’s commentary on the Ilāhiyyāt is a twenty-fifth of his overall commentary on the Shifā᾿, and since the former consists of one volume (according to the report in Text 1 [b]), the latter should consist of twenty-five volumes. This line of reasoning, however, would be meaningful only assuming that, according to al-Ṣafadī, al-Rāzī did provide on each part of the Shifā᾿ an amount of exegesis proportional to the extent of the text of this part, with an uniformity and regularity which are hardly conceivable. Al-Ṣafadī’s point in Text 1 may be quite different and much simpler. Contrary to what al-Zarkān believes, he may be simply maintaining that if al-Rāzī actually penned a commentary on the Shifā᾿, as someone reports, this commentary was made of twenty-five volumes. This contention in Text 1 [c] might well be absolute, in the sense of being made by al-Ṣafadī on the basis of his own knowledge of the issue, without necessarily depending on what al-Ṣafadī reports in Text 1 [b] about Ibn al-Qarī῾’s eye-witness of a specific parcel of al-Rāzī’s alleged commentary in the Maghreb. On the reliability of this testimony al-Ṣafadī remains scrupulously hesitant, as we have seen. If this is really al-Ṣafadī’s point in Text 1 [c], then his statement can be explained on the basis of two complementary assumptions: first, al-Ṣafadī took al-Rāzī’s commentary on the Shifā᾿ as consisting of marginal annotations to Avicenna’s text, rather than as a lemmatic independent exegesis; second, he was acquainted with a division of the Shifā᾿ in twenty-five volumes. These two assumptions entail that al-Rāzī’s commentary on the Shifā᾿, in so far as it belonged to the genre of marginalia, could only consist of the same number of volumes of which the Shifā᾿ was made, i.e. twenty-five. The first assumption is being ascertained by current research on al-Rāzī’,34 and corroborated by the ḥāshiya format of the earliest known exegesis of the Shifā᾿, on 34. Recent research on al-Rāzī has shown the existence of an exegesis by him on at least the first part of the Shifā᾿ on logic, with particular regard to the sections corresponding to Porphyry’s Isagoge, and Aristotle’s Categories and De interpretatione, in the form of glosses in manuscripts (DI VINCENZO, “Early Exegetical Practice on Avicenna’s Šifā᾿”).

AVICENNA’S SHIFĀ᾿ AND A COMMENTARY ON IT

271

stage before al-Ṣafadī’s times.35 The second assumption is the topic of the following pages. In the next section, I am going to explore the possibility that what Text 1 implies about the Shifā᾿ and al-Rāzī’s commentary on it as made of twenty-five parts may depend on a material division of Avicenna’s work in twenty-five volumes like the one displayed in the Pococke collection, which al-Ṣafadī possibly took as the standard layout of the work and, hence, of the commentaries thereupon, once these latter are understood as consisting of marginalia related to Avicenna’s text. The Pococke collection and the partition of the Shifā᾿ that it deploys might have been known to al-Ṣafadī in the VIIIth/XIVth century, functioning to his eyes as a sort of paradigmatic articulation of Avicenna’s summa and of the exegesis concerning it. 3. Looking for Connections Both the segmented copy of the Shifā᾿ partially surviving in the Pococke Collection and the commentary on the Shifā᾿ that al-Ṣafadī ascribes to Fakhr al-Dīn al-Rāzī consisted of twenty-five volumes. Is this more than sheer coincidence? In the present section, I wish to test a positive answer to this question, examining the reasons that may prompt us to believe that this overlapping of data was not fortuitous. Let me dispell from the outset any undue expectation: I am not arguing that a commentary by Fakhr al-Dīn al-Rāzī on the Shifā᾿ consisting of marginal glosses is contained in the marginalia of the Pococke collection; no ownership statement by al-Rāzī is presently known there, and no continuous apparatus of glosses of exegetical character can be seen in the extant manuscripts of the collection. We cannot exclude that among the many marginalia that are being found in other manuscripts of the Shifā᾿ also glosses of al-Rāzī beyond the limited scope of those already discovered at the beginning of the logic of the Shifā᾿ will be identified in the future,36 but the Pococke collection is not a 35. In A. BERTOLACCI, “God’s Existence and Essence: The Liber de Causis and School Discussions in the Metaphysics of Avicenna,” in: D. CALMA (ed.), Reading Proclus and the Book of Causes, Volume III. On Causes and the Noetic Triad, Leiden / Boston 2022, pp. 251280 (p. 272), I have called attention to examples of exegesis of the Ilāhiyyāt in the form of glosses in manuscripts and in authors of the early VIth/XIIth century. 36. See above, n. 34.

272

AMOS BERTOLACCI

promising source of this kind of evidence. More humbly, in this last section I aim at providing a better assessment of al-Ṣafadī’s knowledge of Avicenna’s Shifā᾿, and of his possible acquaintance with the Pococke collection as an authoritative model of the work’s format. 3.1. Al-Ṣafadī and the Shifā᾿ Al-Ṣafadī is presently known mainly in connection with Avicenna’s Al-Ishārāt wa-l-tanbīhāt: on the one hand, he conveyed precious information on commentators on this summa of Avicenna; on the other hand, he was a reader and student of this work himself.37 One should not overlook, however, the attestations of high esteem that he accords to the Shifā᾿ among Avicenna’s works. In his biographical lexicon of contemporary notables and scholars, A῾yān al-῾aṣr wa-a῾wān al-naṣr (The important persons of the age), he starts the entry on his revered master Ibn al-Akfānī – soon after the full mention of his name and a generic laud of the vastity of his intellectual interests – with a praise of him built on the titles of Avicenna’s main works: Text 4: KHALĪL B. AYBAK AL-ṢAFADĪ, A῾yān al-῾aṣr wa-a῾wān al-naṣr (The important persons of the age), ed. WITKAM, p. 31. Had the ra᾿īs [i.e. Avicenna] seen him [i.e. Ibn al-Akfānī], he would have directed his pointer (ishāra) towards him; by means of him his cure/healing (shifā῾) would have been sound and his salvation (najāt) complete; his canon (qānūn) would have not chanted, and his Eastern wisdom (ḥikma mashriqiyya) would have not brought forth advantages until setting down.38

In this text, the Shifā᾿ is mentioned on a pair with all other major works of Avicenna, second in order only the Ishārāt, together with the Najāt, before the Qānūn fī l-ṭibb. It thus turns out to enjoy an utmost prestige, which presupposes familiarity on al-Ṣafadī’s part. 37. ENDRESS, “Reading Avicenna in the Madrasa,” p. 411; A. SHIHADEH, Doubts on Avicenna. A Study and Edition of Sharaf al-Dīn al-Mas῾ūdī’s Commentary on the Ishārāt, Leiden / Boston 2016, pp. 13-14 and n. 34. 38. Arabic text in J. J. WITKAM, De Egyptische arts Ibn al-Akfānī (gest. 749/1348) en zijn indeling van de wetenschappen. Editie van het Kitāb Iršād al-Qāṣid ilā Asnā al-Maqāṣid met een inleiding over het leven en werk van de auteur, Leiden 1989, p. 31, l. 4-5 of the Arabic text (Dutch translation, p. 36). Cf. Khalīl b. Aybak al-Ṣafadī, A῾yān al-῾aṣr wa-a῾wān al-naṣr (The important persons of the age), facsimile edition of MS Istanbul Atif I809, ed. F. SEZGIN, Frankfurt am Main 1990; ENDRESS, “Reading Avicenna in the madrasa,” p. 415.

AVICENNA’S SHIFĀ᾿ AND A COMMENTARY ON IT

273

3.2. Al-Ṣafadī and Ibn al-Labūdī Al-Ṣafadī was deeply rooted in the cultural environment of Syria, his homeland, where the Pococke collection originated. For instance, he resided a few years (between 723/1323 and 726/1326) in Aleppo, the city where this collection was produced in the previous century. His professional activity of copyist of works (500 according to a plausible esteem)39 might also have conveyed acquaintance with the original status and the later vicissitudes of the collection. This contextual likelihood is corroborated by some historical information we can get about the owners of the Pococke collection from al-Ṣafadī himself. The second attested owner, Yaḥyā b. Muḥammad b. al-Labūdī, is a case in point. In part XXVIII of al-Wāfī bi-l-wafayāt, al-Ṣafadī gives abundant information on this person (607-670/1210-1271), mentioned three times in the Pococke collection as owner of single volumes. The entry on Ibn al-Labūdī in al-Ṣafadī’s work largely overlaps with the similar, more extensive, entry in Ibn Abī Uṣaybi῾a’s (d. 668/1270) ῾Uyūn al-anbā᾿ fī ṭabaqāt al-aṭibbā᾿ (The Best Accounts of the Classes of Physicians, abbreviated as IAU in what follows), where additional information can be found, starting from a fuller name of our personage.40 Al-Ṣafadī, however, selects and rearranges the information that he draws from Ibn Abī Uṣaybi῾a in original ways and adds further data, with modifications and integrations that are relevant to the present investigation. Text 5: Al-Ṣafadī, al-Wāfī bi-l-wafayāt, XXVIII, no. 242, ed. I. SHABBŪḤ, Beirut 2004, pp. 312. 3-313. 14 Yaḥyā b. Muḥammad b. ῾Abdān b. ῾Abd al-Wāḥid b. ῾Abda, the leader minister (wazīr) Najm al-Dīn b. al-Labūdī al-Dimashqī, physician. He rose in medicine [i.e. he was appointed as first physician] at the court of the lord of Ḥims, Ibrāhīm,41 for whom he served as minister. Then 39. ROSENTHAL, “al-Ṣafadī,” pp. 759-760 (esp. p. 759b). 40. See the online critical edition, with English translation and commentary, of the work in E. SAVAGE-SMITH – S. SWAIN – G. J. VAN GELDER (eds.), A Literary History of Medicine, Leiden 2020, section 15. 31 at https://dh.brill.com/scholarlyeditions/library/ urn:cts:arabicLit:0668IbnAbiUsaibia/ (consulted 28 May 2020). The name provided in the entry is: al-Ṣāḥib Najm al-Dīn Abū Zakariyā Yaḥyā b. Shams al-Dīn Muḥammad b.῾Abdān b.῾Abd al-Wāḥid b. al-Lubūdī. An entry on Ibn al-Labūdī can also be found in AL-῾UMARĪ (700-749/1301-1349), Masālik al-abṣār fī mamālik al-amṣār, vol. IX (aṭibbā᾿ al-shāmm), pp. 509-511, nr. 129. 41. Al-Malik al-Manṣūr Ibrāhīm b. al-Malik al-Mujāhid (reg. 581-637H/1186-1240).

274

AMOS BERTOLACCI

he joined Nāṣir, lord of Damascus,42 who made him superintendent (nāẓir) of the treasuries. Al-Ṣāliḥ Ayyūb43 appointed him superintendent of Alexandria, after which he honored him at most by granting him every month three thousand dirhams. Then he governed during the reign of al-Ẓāhir.44 He died in the year 670 [1271] and was buried in his mausoleum near the fountain/pool (birka) of the Ḥumriyān [?].45 He made of his mausoleum the house of medicine and geometry, and he granted to it [i.e. the mausoleum] a shaykh and reciters of the Qur᾿ān. His father Shams al-Dīn was one of the great physicians; we have already mentioned him before, among the persons named “Muḥammad.” Master Najm al-Dīn abridged also: (1) the book [i.e. the Elements] of Euclid [= IAU vii], (2) the book ῾Uyūn al-ḥikma (Sources of wisdom) [of Avicenna] [= IAU iv], and (3) the book al-Mulakhkhaṣ [fī l-ḥikma wa-l-manṭiq] (The Abridgment [of wisdom and logic]) of the Imām Fakhr al-Dīn [al-Rāzī] [= IAU v]. He abridged (4) the Muṣādarāt (Premises) of Euclid [= IAU viii]. [He authored] (5) the Ghāyat al-ghāyāt fī l-muḥtāj ilayhi min Uqlīdis wa-l-mutawassiṭāt (The absolute essentials in the required parts of Euclid and the intermediate treatises) [= IAU xiii], (6) the Tadqīq al-mabāḥith al-ṭibbiyya fī taḥqīq al-masā᾿il al-khilāfiyya ῾alā ṭarīq masā᾿il khilāf al-fuqahā᾿ (Detailed study of medical topics and determination of controversial questions in the way jurists deal with questions of controversy) [= IAU xiv]. He wrote (7) the Kitāb īḍāḥ al-ra᾿y al-sakhīf min kalām al-Muwaffaq ῾Abd al-Laṭīf (Illustration of the foolish misconceptions in the utterances of Muwaffaq ῾Abd al-Laṭīf [al-Baghdādī]), which he composed when he was thirteen years old [= IAU xvi], (8) the Ghāyat al-iḥkām fī ṣinā῾at al-aḥkām (The utmost precision in the art of [legal] judgments) [= IAU xvii], (9) the al-Risāla al-saniyya fī sharḥ al-Muqaddima al-Muṭarriziyya (The splendid epistle of commentary on the Muqaddima of al-Muṭarriz) [= IAU xviii], (10) the al-anwār al-sāṭi῾a fī sharḥ al-āyāt al-bayyināt (The brilliant lights of commentary on [The Book of] Clear Signs [by Fakhr al-Dīn al-Rāzī]) [= IAU xix], (11) the book Nuzhat al-nāẓir fī l-mathal al-sā᾿ir (The pleasure of the investigator of the Current proverb [by Ibn al-Athir]) [= IAU xx], (12) the al-Risāla al-kāmila fī ῾ilm al-jabr wa-lmuqābala (The perfect epistle on algebra) [= IAU xxi], (13) the al-zāhī fī ikhtiṣār al-zīj al-shāhī (The brilliant [treatise] of abridgment of the Zīj for Malikshāh [of ῾Umar Khayyām]) [= IAU xxiii], and the (14) al-zīj almuqarrab al-mabnī ῾alā l-raṣad al-mujarrab (The close astronomical table based on experiential observations) [= IAU xxiv]. He authored also other works, like (15) the abridgments of the al-Ishārāt [wa-l-tanbīhāt] 42. Al-Malik al-Nāṣir Yūsuf II (reg. 634-658H/1236-1260). 43. Al-Malik al-Ṣāliḥ Ayyūb b. al-Malik al-Kāmil (reg. 637-647/1240-1249). 44. Al-Malik al-Ẓāhir Rukn al-Dīn Baybars I (reg. 658-676H/1260-1277). 45. The editor Shabbūḥ makes explicit the vocalization of this term at the dual case. The reference of the toponym birkat al-ḥumriyayn remains obscure to me.

AVICENNA’S SHIFĀ᾿ AND A COMMENTARY ON IT

275

(Pointers and Reminders) [of Avicenna] [= IAU iii], (16) of the Kulliyyāt (Generalities) [of Avicenna’s Canon] [= IAU i], and (17) of the al-Mu῾ālimīn (Those who know) [cf. al-Mu῾āmilīn fī l-uṣūlayn (Those who deal with the two principles [of religion and jurisprudence]) [= IAU vi] and other works. Ṣāḥib Najm al-Dīn b. al-Labūdī wrote some praises of our Lord al-Khalīl [i.e. Abraham], among which is [the following]: [part of the fourth poem devoted to al-Khalīl in IAU is quoted, with the concluding remark or gloss “I said: a poem declining from the level of excellence”].

As to the biography, with respect to IAU al-Ṣafadī presents Ibn al-Labūdī more as a minister than as a physician, more as a scholar than as a poet.46 IAU specifies Aleppo as place of birth of al-Labūdī, whereas from al-Ṣafadī we apprehend that he was Damascene.47 About the elements of al-Labūdī’s biography in Text 5 that are missing in IAU, Ibn Kathīr’s Bidāya confirms that al-Labūdī died in 670/1271 and specifies that the “house of medicine and geometry” established by al-Labūdī was a school bearing his name (al-Labūdiyya, i.e. madrasa).48 46. According to the ῾Uyūn al-anbā᾿, Ibn al-Labūdī was an outstanding physician, philosopher, and literate (several poems of his are reported in the entry, one of which is recorded as recited in Jerusalem). Born in Aleppo in 607/1210, he studied medicine in Damascus, and started his activity as court physician first at Ḥimṣ and then in Egypt. In the course of time, he added administrative capacities to scientific activities: he first became head of the treasury in Alexandria and, upon his return from Egypt to Syria, head of the Ayyubid administration for the entire province. Also his activity as jurist must have had high significance if, as it seems, our Ibn al-Labūdī should be identified with the al-Labūdī who expressed authoritative legal opinions recorded in historical sources (see M. A. J. BEG, “Djazzār,” in: P. BEARMAN e.a. (eds.), Encyclopaedia of Islam, Second Edition, vol. XII, Supplement, Leiden 2004, pp. 267-268; ID., “Ḥā᾿ik,” ibid., pp. 340-342. Cf. SAVAGE-SMITH e.a. (eds.), A Literary History of Medicine, section 15. 31, English translation, n. 2). 47. Another minor point of disagreement between the two biographical reports regards the beginning of his administrative offices (in Alexandria according to IAU, in Ḥims according to al-Ṣafadī). 48. IBN KATHĪR, al-Bidāya, vol. XVII, p. 502. 7-9: “Najm al-Dīn Yaḥyā b. Muḥammad b.῾Abd al-Wāḥid b. al-Labūdī, founder of the Labūdiyya [school] which was near a bath, the star that surpasses [reading al-mubarriz as in the ms.] the physicians. He had excellence in the knowledge of medicine, and was appointed as superintendent of the treasuries of Damascus. He was buried in his mausoleum near the Labūdiyya [school].” No information on the school in question can be found in Y. ECHE, Les bibliothèques arabes publiques et semipubliques en Mésopotamie, en Syrie et en Égypte au Moyen Âge, Damascus 1967, and in D. BEHRENS-ABOUSEIF, The Book in Mamluk Egypt and Syria (1250-1517): Scribes, Libraries and Market, Leiden / Boston 2019, who notices (p. 28): “throughout this period, emirs alongside members of the elite founded a significant number of educational institutions housed in fine monuments in Syrian cities. However, because Mamluk waqf documents regarding Syria are scarce, so is our information about their libraries. This

276

AMOS BERTOLACCI

We don’t know precisely in which city this school was founded, whether in Aleppo or in Damascus, but the role of al-Labūdī as founder of a school attests his influential cultural activity. The list of twenty-four works at the end of the entry on Ibn al-Labūdī in IAU presents abridgments and commentaries of both Avicenna and al-Rāzī. More specifically: a summary (mukhtaṣar) of the Kulliyyāt (Generalities) of Avicenna’s Canon of Medicine (no. i), of Avicenna’s Ishārāt wa-l-tanbīhāt (Pointers and Reminders) (no. iii), of Avicenna’s ῾Uyūn al-ḥikma (Sources of Wisdom) (no. iv), and of al-Rāzī’s Mulakhkhaṣ fī l-ḥikma wa-l-manṭiq (The Abridgment of wisdom and logic) (no. v), as well as a commentary (sharḥ) on the Āyāt al-bayyināt (Clear Signs) of al-Rāzī (no. xix). Since also al-Rāzī commented on Avicenna’s Kulliyyāt, Ishārāt wa-l-tanbīhāt, and ῾Uyūn al-ḥikma, al-Rāzī appears to have inspired Ibn al-Labūdī in approaching Avicenna’s corpus. Although Ibn al-Labūdī’s works shrink to seventeen items in the notice by al-Ṣafadī, this latter records all the five aforementioned works of Avicenna or al-Rāzī (nos. 16, 15, 2, 3, 10, respectively). The order of these works, however, is different. As regards Avicenna, in al-Ṣafadī the abridgment of the ῾Uyūn al-ḥikma comes at the beginning (no. 2), whereas the abridgments of the Ishārāt and of the Kulliyyāt are mentioned at the end (nos. 15 and 16, respectively). Thus, in al-Ṣafadī’s arrangement Ibn al-Labūdī’s exegesis of Avicenna somehow encapsulates his entire production.49 Text 5 is congruent with what we know about Ibn al-Labūdī as owner of three manuscripts of the Pococke collection. This text underscores Ibn al-Labūdī’s interest for mathematics, mentioning his abridgments of the three writings of Euclid also recorded in IAU (nos. 1, 4, 5; vii, viii, xiii in IAU), his own writing on algebra (no. 12; xxi in IAU), and his original writing and abridgment of astronomical

scarcity of information even in waqf documents should not necessarily be interpreted as absence of books. Rather the endowment mechanisms regarding institutional libraries seem to have been complex and random.” 49. With respect to IAU, al-Ṣafadī never mentions the name of Avicenna when he speaks of his works (see instead the reference to “Ibn Sīnā” in the three places of IAU) and he shortens the titles of this latter’s works (more fully spelt in IAU), whose knowledge is possibly assumed. By contrast, the name of al-Rāzī in IAU (“Son of the Preacher of al-Rayy”, no. v) becomes the more standard “Imām Fakhr al-Dīn” in al-Ṣafadī (no. 3).

AVICENNA’S SHIFĀ᾿ AND A COMMENTARY ON IT

277

tables (nos. 13-14; xxiii-xxiv in IAU). These mathematical works are arranged with emphasis on geometry.50 In this respect, al-Ṣafadī significantly specifies that the cultural institution that al-Ṣafadī founded out of his mausoleum was devoted to medicine and geometry (handasa). This well accords with his ownership of two volumes of the mathematics of the Shifā᾿ in the Pococke collection, the first of which (ms. 19) contains Avicenna’s treatment of geometry. We cannot be sure that Ibn al-Labūdī possessed the entire Pococke collection, rather than only the three volumes in which his name appears, although one may wonder what utility he might have had in possessing only a “segment” of logic in ms. 2 and a “segment” of music in ms. 21, besides the entire geometry in ms. 19.51 Rather than attestations of al-Labūdī’s possess of only these three volumes, the three ownership statements might function as sorts of ex libris placed towards the beginning (ms. 2) and the end (mss. 19, 21) of the collection. Al-Ṣafadī lived a few decades after the death of Ibn al-Labūdī in the same geographical area: this proximity allowed him, for example, to provide topographical details about Ibn al-Labūdī’s life (like the name of the fountain or pool, birka, in proximity of which his school was founded) unattested elsewhere.52 This being the case, we are entitled to assume that al-Ṣafadī might have been acquainted with Ibn al-Labūdī’s legacy more broadly than Text 5 documents.

50. The first work of Ibn al-Labūdī recorded by al-Ṣafadī is the abridgment of Euclid’s Elements, which comes much later in IAU (vii). The same applies to the other two works of Euclid mentioned by both IAU and al-Ṣafadī, which in this latter come much earlier in the list (nos. 4, 5; viii, xiii in IAU). 51. In the hypothesis of a fragmentary ownership, the gap regarding the part of the Shifā᾿ on astronomy in ms. 20 would be especially hard to explain, since Ibn al-Labūdī’s theoretical and practical engagement with astronomy is attested by his two works on astronomical handbooks with tables (the zīj, pl. azyāj, genre) in both IAU (nos. xxiii, xxiv) and al-Ṣafadī (nos. 13, 14), and is confirmed by what we know from other sources about his purchasing of a manuscript of astronomy in 633/1235-1236. In that year, a Yaḥyā b. Muḥammad b. al-Labūdī – in all likelihood our man – entered into possession of MS Oxford, Bodleian Library, Arch. Seld. A32 (see A. I. SABRA – N. SHEHABY, Ibn al-Haytham, Al-Shukūk ‘alā Batlamyūs, Cairo 1971, p. ix; D. RAYNAUD, A Critical Edition of Ibn al-Haytham’s On the Shape of the Eclipse, Springer 2016, p. 16. 52. Ibn Kathīr, al-Bidāya wa-l-nihāya (see above, n. 48), for example, speaks generically of a bath (ḥammām) without further indications, within a passage which is not totally clear.

278

AMOS BERTOLACCI

3.3. Al-Ṣafadī and Ḥasan b. ῾Alī al-Mutaṭabbib On the basis of a plausible identification, the third and last attested owner of the Pococke collection, Ḥasan b. ῾Alī al-Mutaṭabbib, turns out to have been a physician of some importance, contemporary of al-Ṣafadī in the VIIIth/XIVth century, with a network of professional connections involving Egypt. Purchaser also of works of philosophy lato sensu, he interacted with other physicians and commentators of Avicenna’s Canon of Medicine.53 The medical interests of Ḥasan b. ῾Alī well accord with the epithet “al-Mutaṭabbib” which accompanies his name in ms. 21 of the Pococke collection, whereas the chronological setting in the VIIIth/XIVth century is consistent with his having owned ms. 21 after Ibn al-Labūdī in the previous century. The possible acquaintace of Ḥasan b. ῾Alī al-Mutaṭabbib with al-Ṣafadī remains unattested, and a special analysis of this issue would lead too far. In general, we would like to know more about the passage of items of the Pococke collection from one owner to the other, and how the entire set apparently remained united in one single place, despite its differentiate ownership, until Pococke’s times. In this regard, the hypothesis that the entire collection rather than a part of it, despite the discontinuous ownership statements, changed property in time would surely be helpful. The presence of a manuscript of the Pococke collection (no. 19) in the Ashrafiyya Library in Damascus, from which the manuscript in question possibly departed to Istanbul together 53. A Jalāl al-Dīn al-Ḥasan b. ῾Alī al-Siwāsī (called “King of the wise men”, malik al-ḥukamā᾿, and “Head of the physicians”, ra᾿īs al-aṭibbā᾿), possibly the same as our third owner, is recorded as owner of the ms. Leiden, Or. 606 (Quṭb al-Dīn al-Shīrāzī [d. 710/1311], Sharḥ Ḥikmat al-ishrāq of al-Suhrawārdī) which he purchased in 751/1350 from Zayn al-῾arab ῾Alī b. ῾Ubayd Allāh al-Miṣrī (d. in the second half of the VIIIth/ XIVth c.; J. J. WITKAM, Inventory of the Oriental Manuscripts of the Library of the University of Leiden, vol. 1, pp. 255-256), who commentated on Avicenna’s Canon of Medicine in 751/1350 (A. Z. ISKANDAR, A Catalogue of Arabic Manuscripts on Medicine and Science in the Wellcome Historical Medical Library, London 1967, pp. 181-182, and plates 26a, 26b [Ms. Or. 119 of his commentary on the Canon is dated 758/1356]; G. C. ANAWATI, Essai de bibliographie avicennienne, Cairo 1950, pp. 204-206; F. SANAGUSTIN, Avicenne (XIe siècle), théoricien de la médecine et philosophe - Approche épistémologique, Damas 2010, p. 45) and was acquainted with the medical works by Ibn al-Nafīs of Damascus (A. Z. Iskandar, “Ibn Al-Nafīs, ῾Alā᾿ Al-Dīn Abū l-Ḥasan ῾Alī Ibn Abī l-Ḥazm Al-Qurashī (or Al-Qarashī),” in: Dictionary of Scientific Biography, pp. 602–606, consulted online at https://www.encyclopedia.com/science/dictionaries-thesauruses-pictures-and-pressreleases/ibn-al-nafis on May 30 2020).

AVICENNA’S SHIFĀ᾿ AND A COMMENTARY ON IT

279

with the library’s catalogue, has been recently argued.54 Since the Ashrafiyya Library, established by the Ayyubid amīr of Damascus alMalik al-Ashraf (d. 635/1237) was later enlarged to gather book material coming from Northern Mesopotamia, Fatimid Egypt, and Syria, the possibility that the Pococke collection or a part of it was hosted at some point in this library, coming from smaller libraries like the school/mausoleum of Ibn al-Labūdī (d. 670/1271), cannot be excluded.55 Why only this single item of the collection moved later to Istanbul – in all likelihood after the Ottoman conquest of Syria in 1516 – is difficult to explain, unless ms. 19 departed from Syria together with some of the other manuscripts of the Pococke Collection presently missing, thus splitting the collection in two distinct locations (sixteen manuscripts in Syria, the other nine elsewhere). But, also under this hypothesis, the random nature of the selection and the rationale of the diaspora would remain problematic. The lost of the frontispieces of three extant manuscripts may prevent access to information relevant in this regard. For the time being, it can be asserted that the Pococke Collection had many potential “access points” during al-Ṣafadī’s lifetime, in Aleppo, Damascus, or elsewhere in Mamluk territories. Some of these venues were due to intellectuals with vast interests and prime political roles like Ibn al-Labūdī, who intended to leave lasting traces in their cultural and social environment: the very idea of founding schools or libraries like the Labūdiyya or the Ashrafiyya attests that influencing cultural life was an important element of their political agenda, making their cultural impact on al-Ṣafadī’s generation arguably significant. This being the case, a version of the Shifā᾿ contained in these repositories 54. HIRSCHLER, Medieval Damascus, p. 27 and n. 35. 55. The relevant passage of Hirschler’s edition of the catalogue of the Ashrafiyya Library (p. 464, l. 16-17; Engl. transl. p. 397, no. 1471) min al-Shifā᾿ li-Ibn Sīnā (“parts of the Shifā᾿ of Ibn Sīnā”) can obviously regard other items of the Pococke collection beside the one labelled here as 19. Hirschler regards the presence of item 19 in the Ashrafiyya Library as likely, marking it with two asterisks (**) at p. 397. If we assume that the part of the Pococke collection referred to in this passage of the Ashrafiyya catalogue was transferred in this library after the death of its last attested owner, Ḥasan b. ῾Alī al-Mutaṭabbib (d. after 751/1350, see above, n. 53), we have good reason to agree with R. ŞEŞEN, Mukhtārāt min al-makhṭūṭāt al-῾arabiyya al-nādira fī maktabāt Turkiyā, Istanbul 1997, p. 893, who dates the catalogue to the VIIIth/XIVth c. Hirschler proposes an earlier date (670s/1270s: see Medieval Damascus, p. 5) for the composition of the catalogue.

280

AMOS BERTOLACCI

had good chances to become a well-known and authoritative version of that work in the region. We should, of course, avoid unwarranted inferences, since al-Ṣafadī might have known the precise composition of the Pococke Collection, if he actually did, also from elsewhere. However, despite this uncertaintly regarding the institutional and social context of Bilād al-Shāmm in the VIIIth/XIVth century, still largely a terra incognita, the diffused presence and resonance in it of the Pococke Collection through its owners can be taken as a plausible fact. In lack of precise documentation, the hypothesis of a connection between al-Ṣafadī and the Pococke collection (through Ibn al-Labūdī, Ḥasan b. ῾Alī al-Mutaṭabbib, or others) remains, at the current state of research, only a reasonable guess. Tentative as it may be, this guess presents nonetheless a considerable explanatory advantage: the extent of the Pococke collection, if known to al-Ṣafadī, helps making sense of the otherwise cryptic mention of the twenty-five volumes in Text 1 [c]. 4. Conclusion The present paper has dealt with two pieces of evidence attesting that at the turn between the Vth/XIth and the VIth/XIIth centuries Avicenna’s Shifā᾿ was copied in twenty-five volumes and was allegedly also commented according to the same number of volumes by one of the main exponents of post-Avicennian theology, Fakhr al-Dīn al-Rāzī. The first piece of evidence is grounded in manuscripts, namely the extant codices of the Bodleian Pococke collection of the Shifā᾿, originally made of twenty-five volumes. The second piece of evidence is much more speculative, being a hypothesis formulated by the VIIIth/XIVth century historian Ṣalāḥ al-Din al-Ṣafadī (d. 764/1363) in the entry on Fakhr al-Dīn al-Rāzī in his most important work of historiography. The Pococke collection was written between Ramaḍān 601/May 1205 and Jumādā I 604/December 1207, when al-Rāzī (d. 606/1210) was still alive. However, we have no attestation that this collection, written (at least in part) in Aleppo, may be linked with Herat in Afghanistan, where al-Rāzī spent the last part of his life. For sure, this series of manuscripts does not contain any significant paratextual apparatus of glosses, which may preserve traces of the commentary on the Shifā᾿ ascribed to al-Rāzī by al-Ṣafadī. The codicological and historical evidence considered in this article, however, though not able to enlarge the present documentation on al-Rāzī’s exegesis on the

AVICENNA’S SHIFĀ᾿ AND A COMMENTARY ON IT

281

Shifā᾿, may nonetheless improve our knowledge of how al-Rāzī approached Avicenna’s masterpiece in philosophy, and of how this latter circulated in the heyday of Mamluk history, through the eyes of the polymath al-Ṣafadī. If the number twenty-five under al-Ṣafadī’s pen in Text 1 is indeed related to the volumes of the Pococke collection, then two implications stem from al-Ṣafadī’s statement. First, by projecting a partition of a copy of the Shifā᾿ like the Pococke collection on the format of al-Rāzī’s commentary, al-Ṣafadī would imply that the commentary by al-Rāzī on the Shifā᾿ had the form of marginalia to Avicenna’s text. Second, by the same token, al-Ṣafadī would attest to the diffusion and impact of the Pococke collection in the Syrian and Egyptian philosophical circles, which he frequented. The first assumption – i.e. paratextuality as the form of exegesis on Avicenna’s Shifā᾿ adopted by al-Rāzī, and as a modus operandi typical of every interpreter of the Shaykh al-ra᾿īs’s masterpiece in pre-Safavid times, when this previous “liminal” exegesis was first fixed in independent commentaries – is being confirmed by recent scholarship.56 Thus, though al-Ṣafadī was probably not directly acquainted with al-Rāzī’s commentary on the Shifā᾿, he might have rightly guessed the format of this commentary. The second assumption – namely the possible connection of al-Ṣafadī with the environment in which the Pococke collection was composed in the century before the one in which he lived, and with the milieu in which it was transmitted to subsequent generations of scholars in Mamluk territories – is a promising path of investigation whose present evidence needs extension and further corroboration. On the one hand, the existence of local authoritative versions of the Shifā᾿ like, in our case, the Pococke collection, prompts the research on the manuscript dissemination of Avicenna’s magnum opus towards a regional history of its diffusion and a careful inspection of its repositories during al-Rāzī’s lifetime.57 On the other hand, outstanding biographers 56. See above, nn. 36-37. 57. BEHRENS-ABOUSEIF, The Book in Mamluk Egypt and Syria, p. 28: “The narrative sources ... rarely mentions libraries or the office of a librarian in their enumeration and description of Mamluk religious foundations in Syria. ... Of course this does not mean that these foundations had no books. Biographical sources indicate that unlike in Cairo, where the academic elite tended to concentrate in the Mamluk princely foundations, the numerous Ayyubid madrasas in Syria maintained their prestigious status and continued to thrive under Mamluk rule, attracting prominent scholars and supplementary endowments.”

282

AMOS BERTOLACCI

and vehicles of intellectual history like al-Ṣafadī, mostly appreciated so far for the information they provide on second-rank Avicennists, emerge also as precious apprisers of the modalities through which al-Rāzī commented on Avicenna. In all these respects, the histories of Avicennism and of Razism look deeply intermingled, as Jules Janssens has pioneeringly foreshadowed in his scholarship. Amos BERTOLACCI IMT Scuola Alti Studi – School for Advanced Studies Piazza S. Francesco 19 It-55100 Lucca [email protected]

PHILOSOPHER AU XIIe SIÈCLE. NOTES AUTOUR D’AVICENNE ET DE SA PREMIÈRE RÉCEPTION LATINE AU MOYEN ÂGE Olga L. LIZZINI Abstract If, as Alain de Libera has written, “the history of Avicennism is first and foremost that of Avicenna latinus,” this is because dealing with translations already means introducing the questions of their reception just as questioning Avicenna’s reception implies the study of translations. By summarizing the major research results concerning the first translations of Avicenna’s texts in Toledo, as well as the “new ideas” elaborated in the first writings that depended on them, and finally, briefly, the historiographical issues related to them, I propose here to review some aspects of the penetration of Avicenna’s metaphysics into “Latin” thought in the 12th century. My remarks offer only a kind of outline; nevertheless, crucial aspects of Avicenna’s early reception and, in a certain sense, even of his philosophy and the “narrative” it implies are presented.

Du point de vue de l’historien de la philosophie médiévale qui est spécialiste de la pensée de langue arabe, le XIIe siècle latin est le siècle où – après les approches éminemment scientifiques du XIe siècle1 – la philosophie arabe commence à être connue par les auteurs du Moyen Âge latin. Plus précisément, c’est au XIIe siècle que, pour la première fois, la philosophie du Moyen Âge latin se trouve à intégrer questions, termes, thèmes et textes qui dérivent de l’effort de traduction, compréhension et élaboration de la pensée grecque, bien sûr – Aristote fut traduit à partir du grec avant qu’il ne le fut à partir de l’arabe2 –, mais 1. Il suffira ici de nommer Constantin l’Africain, sur lequel v. au moins J. BROSSOLET, « Constantin l’Africain (1015-1087) », dans : Encyclopædia Universalis, et les textes d’astronomie et astrologie (v. Ch. BURNETT, Arabic into Latin in the Middle Ages. The Translators and Their Intellectual and Social Context, Furnham / Burlington 2009 et D. JUSTE, Les alchandreana primitifs. Étude sur les plus anciens traités astrologiques latins d’origine arabe (Xe siècle), Leiden / Boston 2007. 2. L. MINIO-PALUELLO, « Aristotele dal mondo arabo a quello latino », dans : Settimane di studio del centro italiano di Studi sull’alto Medioevo, XII, L’Occidente e l’Islam

284

OLGA L. LIZZINI

aussi de la traduction de la pensée grecque telle que le monde de l’Islam classique l’avait élaborée, grâce au travail philosophique original des penseurs qui écrivirent en arabe. On peut à ce propos isoler trois points majeurs. 1. C’est au XIIe siècle qu’on a les premières traductions de textes philosophiques arabes et gréco-arabes – al-Kindī, Isaac Israeli, Ibn Gabirol, al-Fārābī, le De Causis et Avicenne (Ibn Sīnā)3, pour ne nommer que les auteurs et les textes les plus représentatifs – mais aussi les premières versions et élaborations de textes scientifiques (et philosophiques) dont témoignent Adélard de Bath ou Daniel Morley4, et même de textes religieux (on nommera alors la traduction du Coran – la Lex Mahumet pseudoprophete, dont la réalisation avait été encouragée par Pierre le Vénérable5). Les premiers textes nell’alto Medioevo (2-8 aprile1964), Spoleto 1965, t. II, p. 603-637 ; J. BRAMS, « The Latin Aristotle and Medieval Latin Commentaries on Aristotle », dans : Bulletin de Philosophie Médiévale 44 (2002), p. 1-17 ; cf. ID., La riscoperta di Aristotele in Occidente, Milan 2003. 3. En général, v. J. PUIG MONTADA, « The Transmission and Reception of Arabic Philosophy in Christian Spain », dans : C. B. KESSEL (éd.), The Introduction of Arabic Philosophy in Europe, Leiden 1994, p. 7-30 ; D. N. HASSE, « Influence of Arabic and Islamic Philosophy on the Latin West », dans : The Stanford Encyclopedia of Philosophy (Fall 2021 Edition), URL = . Quelques notices sur l’influence d’Isaac Israeli dans : L. LEVIN – R. D. WALKER – S. SADIK, « Isaac Israeli », dans : The Stanford Encyclopedia of Philosophy (Summer 2018 Edition), URL = https://plato.stanford.edu/archives/sum2018/entries/ israeli/ ; pour Ibn Gabirol, v. H. ADLER, « Ibn Ǧabirol and His Influence upon Scholastic Philosophy », Londres 1964 et l’essai de M. BENEDETTO dans : AVICEBRON, La fonte della vita, Milan 2007. 4. V. au moins Ch. BURNETT, Adelard of Bath: An English Scientist and Arabist of the Early Twelfth Century, Londres 1988 ; ID. « The Translation of Arabic Science into Latin: A Case of Alienation of Intellectual Property? », dans : Bulletin of the Royal Institute for Inter-Faith Studies 4 (2002), p. 145-157 ; ID. « The Blend of Latin and Arabic Sources in the Metaphysics of Adelard of Bath, Hermann of Carintia, and Gundisalvus », dans : M. LUTZ-BACHMANN – A. FIDORA – A. NIEDERBERGER, Metaphysics in the Twelfth Century. On the Relationship among Philosophy, Science and Theology, Turnhout 2004, p. 41-66. Pour D. Morley, v. T. RICKLIN, « Die lateinische Entdeckung der Quintessenz: Die Philosophia des Daniel von Morley », dans : LUTZ-BACHMANN – A. FIDORA – A. NIEDERBERGER (éds.), Metaphysics in the Twelfth Century. On the Relationship among Philosophy, Science and Theology, Turnhout 2004, p. 85-112. Pour la science, v. J. MARTÍNEZ GÁZQUEZ, The Attitude of the Medieval Latin Translators towards the Arabic Sciences, Florence 2016 (avec le compte rendu de J. JANSSENS dans : Bulletin of the School of Oriental and African Studies 81 (2018), p. 538-540). 5. Il s’agit de la première traduction du Coran en Occident, par Robert Ketton en 1143 ; sur cet auteur, v. « Robert of Ketton » dans : D. THOMAS – A. MALLETT (éds.), Christian-Muslim Relations. A Bibliographical History, vol. 3 (1050-1200), Leiden / Boston

PHILOSOPHER AU XIIe SIÈCLE

285

d’Avicenne ont été traduits après 1150 à Tolède6. 2. C’est en outre au XIIe siècle que l’on trouve des textes philosophiques rédigés en latin à partir des « nouvelles idées » arabes. Il s’agit notamment des écrits de Gundissalinus – dont la complexité est de plus en plus soulignée par la recherche contemporaine7 – et d’autres textes où la métaphysique avicennienne est évoquée avec des éléments autres (néoplatoniciens, augustiniens etc.) ; c’est par ex. le cas du texte anonyme des Pérégrinations de l’âme après la mort, édité par M.Th. d’Alverny, et du célèbre De causis primis et secundis8. 3. Enfin, c’est également au XIIe siècle qu’une partie de l’historiographie a situé un courant autonome et au moins en partie hétérodoxe de la pensée latine, marqué par l’influence décidée de la philosophie d’Avicenne – l’avicennisme –, et c’est donc le XIIe siècle qui fut, et qui en partie est encore, au centre 2011, p. 508–519 ; Ch. BURNETT, « A Group of Arabic-Latin Translators Working in Northern Spain in the Mid-twelfth Century », dans : Journal of the Royal Asiatic Society (1977), p. 62-108 ; sur le Coran en Occident, v. les deux projets internationaux : « Qur’ān 12-21 » et « The European Qur’ān » : https://quran12-21.org/en/editions/bibliander et https://euqu.eu/the-european-quran . 6. Pour les traductions latines d’Avicenne, v. en général, J. JANSSENS, « Ibn Sīnā (Avicenna), Latin Translations of », dans : Encyclopedia of Medieval Philosophy. Philosophy Between 500 and 1500, Dordrecht 2011 p. 522-527 ; ibid., 2020, p. 798-805. 7. Il y a désormais une bibliographie importante sur Gundissalinus (qui se répète parfois) ; il suffira ici de citer quelques études : A. FIDORA – M. J. SOTO BRUNA, « Gundisalvus ou Dominicus Gundisalvi? Algunas observaciones sobre un reciente artículo de Adeline Rucquoi », dans : Estudios eclesiásticos 76 (2001), p. 467-473 ; A. FIDORA, Die Wissenschaftstheorie des Dominicus Gundissalinus. Voraussetzungen und Konsequenzen des zweiten Anfangs der aristotelischen Philosophie im 12. Jahrhundert, Berlin 2003 (cf. la traduction espagnole : Domingo Gundisalvo y la teoría de la ciencia arábigo-aristotélica, Pamplona 2009); ID., « Dominicus Gundissalinus and the Introduction of Metaphysics into the Latin West », dans : The Review of Metaphysics 66 (2013), p. 691-712 ; ID., « Omnes decepti sunt. Die Metaphysikkritik des Dominicus Gundissalinus (ca. 1150) », dans : G. KRIEGER (éd.), Die “Metaphysik” des Aristoteles im Mittelalter. Rezeption und Transformation, BostonBerlin 2016 ; N. POLLONI, Domingo Gundisalvo, filósofo de frontera, Madrid 2013 ; ID. « Elementi per una biografia di Dominicus Gundisalvi », dans : Archives d’Histoire Doctrinale et Littéraire du Moyen Âge 82 (2015), p. 7-22 ; N. POLLONI, « Aristotle in Toledo: Gundissalinus, the Arabs, and Gerard of Cremona’s Translations », dans : Ch. BURNETT – P. MANTAS (éds), ‘Ex Oriente Lux’. Translating Words, Scripts and Styles in the Medieval Mediterranean World, Cordoue 2016, p. 147-185 ; ID., Domingo Gundisalvo. Una introducción, Madrid 2017; ID., « The Toledan Translation Movement and Gundissalinus: Some Remarks on His Activity and Presence in Castile », dans : Y. BEALE-RIVAYA – J. BUSIC (éds.), A Companion to Medieval Toledo. Reconsidering the Canons, Leiden / Boston 2018, p. 263-280; ID., The Twelfth Century Renewal of Latin Metaphysics. Gundissalinus’ Ontology of Matter and Form, Toronto 2020. 8. Cf. infra.

286

OLGA L. LIZZINI

d’une discussion sur le caractère de l’influence exercée par la philosophie d’Avicenne dans le monde latin9. Mon but est ici minimal. En suivant les lignes que je viens d’indiquer, je voudrais juste revoir quelques aspects de la pénétration de la métaphysique d’Avicenne dans la pensée latine au XIIe siècle ; mes remarques n’offriront qu’une ébauche, une sorte de canevas ; je m’arrêterai néanmoins sur des aspects qui ont été examinés par la recherche et qui sont cruciaux pour comprendre la première réception d’Avicenne et, en un certain sens, même sa philosophie. Une précision apparaît en effet nécessaire : on peut tenter de diviser les aspects que je viens de nommer (les traductions, la philosophie arabo-latine des premiers textes, et l’avicennisme), mais toute enquête autour de l’un ou de l’autre aspect que je viens d’évoquer implique nécessairement les autres. Si, comme Alain de Libéra l’a écrit, « l’histoire de l’avicennisme est d’abord celle de l’Avicenna latinus »10, c’est que s’occuper des traductions signifie déjà introduire les questions de leur réception tout comme s’interroger sur la réception d’Avicenne passe par la question des traductions. Les travaux sur les traductions latines d’Avicenne de Jules Janssens – auquel ce petit travail est dédié avec estime et gratitude – le démontrent clairement11. 1. Les traductions On doit donc brièvement revoir les données bien connues qui concernent les traductions. À part quelques fragments, les textes d’Avicenne traduits en latin font tous partie de l’œuvre majeure du penseur persan, la plus ample et systématique, à savoir la grande summa de philosophie qui porte le titre de Kitāb al-Shifā’ : le Livre de 9. J’ai discuté de la question dans : O. LIZZINI, « Avicennisme latin », dans : Encyclopédie de l’Humanisme méditerranéen, dir. par H. TOUATI, CNRS, 2016, http://encyclopediehumanisme.com/?Avicennisme-latin#outil_sommaire. 10. A. DE LIBERA, « Avicennisme latin », dans : Encyclopaedia Universalis, vol. III, Paris 1990 (v. aussi Dictionnaire de la Philosophie [Les Dictionnaires d’Universalis], Paris 2012). Pour les textes édités, v. la collection de l’Avicenna Latinus : http://www.uai-iua.org/en/ publications?project=89. 11. J’ai présenté une première version de ce travail au colloque « Philosopher au XIIe siècle » au Collège de France (en mai 2017). Avec A. de Libéra, promoteur du colloque, je remercie Ch. Grellard et A. B. Avalos. Je tiens aussi à remercier Massimiliano Lenzi, qui a lu mon texte et partagé avec moi ses remarques.

PHILOSOPHER AU XIIe SIÈCLE

287

la guérison (de l’âme), que le Moyen-Âge latin connut sous le nom de Sanatio ou de Liber Sufficientiae (Sufficientia ou Communia naturalium comme on nommait parfois la physique12 ; Assiphe ou Asschyphe selon la translittération du titre de l’œuvre que l’on retrouve encore comme Essepha, traduit par « Satisfactio », dans le De viris quibusdam illustribus apud Arabes, attribué de manière douteuse à Léon l’Africain, m. 1555)13. De cette summa, au XIIe siècle, les Latins connaissent ou commencent à connaître essentiellement : 1. La « Préface » du Kitāb al-Shifā’ ; elle est constituée d’une Introduction d’al-Juzjānī – élève et secrétaire d’Avicenne – et du prologue à l’œuvre dû à Avicenne lui-même (al-Madkhal, I.1) ; la traduction remonte aux années 1152-1166 et fait partie des textes traduits à Tolède par Avendauth, que l’on identifie désormais avec Abraham ibn Da‘ūd, mort à Tolède vers 118014 ; le texte eut une circulation limitée 12. Sur ce titre, v. G. SALIBA, « Avicenna’s Shifā᾿ (Sufficientia): in Defense of Medieval Latin Translators », dans : Der Islam 94 (2017), p. 423-433. 13. Leo AFRICANUS, De viris quibusdam illustribus apud Arabes, éd. J. H. HOTTINGER, Bibliothecarius quadripartitus, Zurich 1664, p. 246-291. Sur Léon l’Africain, v. au moins K. STARCZEWSKA, « Leo Africanus », dans : D. THOMAS – J. CHESWORTH e.a. (éds.), Christian-Muslim Relations. A Bibliographical History. Volume 6. Western Europe (1500-1600), Leiden / Boston 2014, p. 439-449. Cf. http://referenceworks.brillonline.com/entries/ christian-muslim-relations-ii/leo-africanus-COM_26250 14. Ce serait lui qui, selon quelques sources, aurait transmis le Liber de causis au monde latin ; le texte des premiers manuscrits est, en effet, connu comme Metaphysica Avendauth et Albert le Grand en identifie l’auteur comme étant David Iudaeus : v. Ch. BURNETT, « The Coherence of the Arabic-Latin Translation Program in Toledo in the Twelfth Century », dans : Ch. BURNETT, Arabic into Latin in the Middle Ages, p. 249-288, en particulier p. 265 ; A. FIDORA, « Abraham Ibn Daud und Dominicus Gundissalinus: Philosophie und religiöse Toleranz im Toledo des 12. Jh. », dans : M. LUTZ-BACHMANN – A. FIDORA (éds.), Juden, Christen und Muslime. Religionsdialoge im Mittelalter, Darmstadt 2004, p.10-26 ; A. FIDORA, « Ein philosophischer Dialog der Religionen im Toledo des 12. Jahrhunderts: Abraham Ibn Dawud und Dominicus Gundissalinus », dans : Y. SCHWARTZ – V. KRECH (éds.), Religious Apologetics – Philosophical Argumentation, Tübingen 2004, p. 251-266 ; A. BERTOLACCI, « A Community of Translators : The Latin Medieval Versions of Avicenna’s Kitāb al-Shifā’ (Book of the Cure) », dans : C. MEWS – J. CROSSLEY (eds.), Communities of Learning, Networks and the Shaping of Intellectual Identity in Europe, 1100-1500, Turnhout 2011, p. 37-54 ; A. BERTOLACCI, « The Reception of Avicenna in Latin Medieval Culture », dans : P. ADAMSON (éd.), Interpreting Avicenna. Critical Essays, Cambridge 2013, p. 242-269. Pour l’identification v. aussi M.-Th. D’ALVERNY, « Avendauth? », dans : Homenaje Millás Vallicrosa, vol. I, Barcelone 1954, p. 19-43 ; Cf. K. SZILÁGYI, « Ibn Daud and Avendauth? Notes on a Lost Manuscript and a Forgotten Book », dans : Aleph 16 (2016), p. 10-31 ; G. FREUDENTHAL, « Abraham Ibn Daud, Avendauth, Dominicus Gundissalinus and Practical Mathematics in Mid-Twelfth Century Toledo », dans : Aleph 16 (2016), p. 61-106; en général, v. R. FONTAINE – A. ERAN,

288

OLGA L. LIZZINI

(on n’en connaît que trois copies ; Roger Bacon et Albert le Grand semblent s’y référer15). 2. Une partie de la Logique, notamment le premier traité, l’Isagoge (al-Madkhal), sur les universaux (on en possède treize manuscrits ; il s’agit du seul texte de logique qui fut publié dans l’édition de Venise de 1508). Ce texte fut également traduit à Tolède (1150-1200) par Avendauth16. Les Latins y lurent, parmi d’autres, deux thèses cruciales : la distinction entre la représentation conceptuelle et l’assentiment – connaître ou conceptualiser la quiddité est autre chose que donner l’assentiment à son existence –, et la doctrine – liée à la première – des trois niveaux de considération des quiddités, qui doivent être envisagées différemment s’agissant de leur être même ou bien de leur existence dans les individus ou – encore – de leur existence dans les représentations conceptuelles, à savoir dans l’âme17. Les Latins découvrent – toujours au XIIe siècle – une partie des Analytiques postérieures (Kitāb a-Burhān, II, 7), dont la traduction, qui porte le titre de Summa Avicennae de convenientia et differentia scientiarum, se reconnaît dans le De divisione philosophiae de « Abraham Ibn Daud », dans : The Stanford Encyclopedia of Philosophy (Spring 2020 Edition), URL = . 15. AVICENNA, « Prologus discipuli et capitula », éd. A. BIRKENMAJER, dans : « Avicennas Vorrede zum ‘Liber Sufficientiae’ und Roger Bacon », dans : Revue néoscolastique de philosophie 36 (1934), p. 308-320 réimpression dans : Études d’histoire des sciences et de la philosophie du Moyen Âge. Hommage à Alexandre Birkenmajer par M.-Th. d’Alverny. Choix d’articles par J. B. Korolec, Wrocław 1970, p. 89-101 ; A. BERTOLACCI, « A New Phase of the Reception of Aristotle in the Latin West : Albertus Magnus and His Use of Arabic Sources in the Commentaries on Aristotle », dans : L. HONNEFELDER (éd.), Albertus Magnus und der Ursprung der Universitätsidee. Die Begegnung der Wissenschaftkulturen im 13. Jahrhundert und die Entdeckung des Konzept der Bildung durch Wissenschaft, Berlin 2011, p. 259-276, 491-500 ; A. BERTOLACCI, « Albert the Great and the Preface of Avicenna’s Kitāb al-Šifā’ », dans : J. JANSSENS – D. DE SMET (éds.), Avicenna and His Heritage, Leuven 2002, p. 131-152 ; cf. A. BERTOLACCI, « On the Latin Reception of Avicenna’s Metaphysics before Albertus Magnus : An Attempt at Periodization », dans : D. N. HASSE – A. BERTOLACCI (éds.), The Arabic, Hebrew and Latin Reception of Avicenna’s Metaphysics, Berlin 2011, p. 197-223. 16. Sur cette partie en latin, v. S. DI VINCENZO, « Avicenna’s Isagoge, Chap. I, 12, De Universalibus : Some Observations on the Latin Translation », dans : Oriens 40 (2012), p. 437-476; cf. S. DI VINCENZO, Avicenna. ‘The Healing, Logic: Isagoge’. A New Edition, English Translation and Commentary of the Kitāb al-Madḫal of Avicenna’s Kitāb al-Šifā᾿, Berlin 2021. 17. V. Madkhal I, 3, p. 17, 7-18 et Madkhal, I, 2, p. 15, 1-16, 5. Il existe une très riche bibliographie sur ces passages. Pour les références, v. DI VINCENZO, Avicenna. ‘The Healing, Logic: Isagoge’ ; D. JANOS, Avicenna On the Ontology of Pure Quiddity, Berlin 2020.

PHILOSOPHER AU XIIe SIÈCLE

289

Gundissalinus18. La traduction est donc à situer entre 1150 et 117519. Ici, c’est surtout la question de la division des sciences et de leur subalternatio qui est concernée. Avicenne propose un système du savoir aristotélicien (en permettant ainsi un usage précis de la classification farabienne des sciences20). 3. Encore, les Latins découvrent au XIIe siècle une partie importante (on a même supposé l’intégralité) de la Physique d’Avicenne. Même la version du Kitāb al-Samā‘ al-ṭabī‘ī (Liber primus naturalium, I-III.1, première partie) est à situer à Tolède (1150-1175) et, pour les deux premiers livres de la Physique, ainsi que le prologue et le début du chapitre 1 de la traduction latine (ch. 1 et début du ch. 2 dans le texte arabe), on se réfère à un traducteur anonyme de Tolède21. Les Latins 18. Dominicus GUNDISSALINUS. De divisione philosophiae, éd. A. FIDORA, Freiburg 2007 ; cf. J. JANSSENS, « Le De divisione philosophiae de Gundissalinus: quelques remarques préliminaires à une édition critique », dans : E. CODA – C. MARTINI BONADEO (éds.), De l’antiquité tardive au Moyen Âge: études de logique aristotélicienne et de philosophie grecque, syriaque, arabe et latine offertes à Henri Hugonnard-Roche, Paris 2014, p. 559-570 ; N. POLLONI, « Gundissalinus and Avicenna: Some Remarks on an Intricate Philosophical Connection », dans : Documenti e Studi sulla Tradizione Filosofica Medievale 28 (2017), p. 515-552. 19. La traduction (de parties) de la Rhétorique (al-Khiṭāba), à Burgos, est plus tardive (1240-1256). On trouve des fragments de ce texte (II, 2, et une partie de IV, 1 ; p. 73,7-75 et 206,8-212 de l’édition arabe) dans la traduction de la Rhétorique aristotélicienne d’Hermannus Alemannus, actif au XIIIe siècle ; v. G. CELLI, « Some Observations about Hermannus Alemannus’ Citations of Avicenna’s Book of the Rhetoric », dans : Oriens 40 (2012), p. 477-513. Jusqu’à maintenant, on n’a aucune preuve que la logique ait été traduite en entier ; v. J. JANSSENS, « Albert le Grand et sa connaissance des écrits logiques arabes : une réévaluation du dossier Grignaschi », dans : Ad notitiam ignoti. L’Organon dans la translatio studiorum à l’époque d’Albert le Grand, Turnhout 2013, p. 225-257, qui revoit les hypothèses de M. GRIGNASCHI, « Les traductions latines des ouvrages de la logique arabe et l’abrégé d’Alfarabi », dans : Archives d’Histoire Doctrinale et Littéraire du Moyen Âge 46 (1972), p. 41-107. 20. Cf. la summa Avicennae de convenientia et differentia scientiarum praedictarum qui traite de la subalternatio (le texte correspond aux Analytiques postérieurs d’Avicenne (II, 7) : la théorie avicennienne de la subalternatio permet à Gundissalinus de revendiquer l’existence d’un système hiérarchisé des sciences qui dépendent en fin de compte toutes de la Métaphysique – science supérieure – mais aussi d’intégrer les nouvelles sciences dans ce même système. 21. Pour l’édition du texte, v. les volumes de l’Avicenna latinus dont notamment : Liber primus naturalium: tractatus secundus de motu et de consimilibus, Bruxelles, 2006 ; J. JANSSENS, Avicenna Latinus. Liber primus naturalium. Tractatus tertius. De his quae habent naturalia ex hoc quod habent quantitatem, Bruxelles 2017 ; cf. Ch. BURNETT, « Arabic Philosophical Works Translated into Latin », dans : R. PASNAU (éd.), The Cambridge History of Medieval Philosophy, Cambridge 2014, II, p. 814-822, en particulier p. 818-819 ; sur la Physique d’Avicenne, v. A. LAMMER, The Elements of Avicenna’s Physics, Berlin 2018.

290

OLGA L. LIZZINI

ont également en traduction au XIIe siècle presque l’intégralité des traités II-V de la partie de la Physique (dont le livre sur Les Actions et les passions22). Ce sont ainsi les concepts de nature (ṭabī‘a), de complexion (mizāj), de préparation ou disposition (isti‘dād) de la matière et les enjeux de l’hylémorphisme qui rejoignent le monde latin. Le Kitāb al-Ma‘ādin wa-l-āthār al-‘ulwiyya (correspondant aux Meteorologica d’Aristote, livres A-Γ) est un cas particulier. Il fut d’abord connu de façon partielle : quelques parties du livre I (I.1 et I.5) furent traduites en Latin à Tolède – ou en Angleterre – à la fin du XIIe siècle (ou au début du XIIIe) par Alfred de Sareshel (Shareshill), mais dans cette traduction paraphrastique (extraits de I.1 et I.5) – De mineralibus ou De congelatione et conglutinatione lapidum – le texte connut une circulation énorme23. Il fut ajouté en tant qu’appendice à la traduction La partie correspondant à III. 1, seconde partie – 10 fut traduite à Burgos (1274-80) par Johannes Gunsalvi et Salomon et eut une circulation réduite (avec un seul manuscrit ; v. J. JANSSENS, « The Reception of Avicenna’s Physics in the Latin Middle Ages », dans : A. VROLIJK – J. P. HOGENDIJK (éds.), O ye Gentlemen : Arabic Studies on Science and Literary Culture. In Honour of Remke Kruk, Leiden / Boston 2007, p. 55-64 ; J. JANSSENS, « Ibn Sīnā (Avicenna), The Latin Translations of », dans : H. LAGERLUND (éd.), Encyclopaedia of Medieval Philosophy : Philosophy Between 500 and 1500, Dordrecht 2011, vol. I, p. 522-527 ; J. JANSSENS, « The Physics of the Avicenna Latinus and Its Significance for the Reception of Aristotle’s Physics in the West », dans : A. M. I. VAN OPPENRAAIJ – R. FONTAINE (éds.), The Letter before the Spirit : The Importance of Text Editions for the Study of the Reception of Aristotle, Leiden / Boston 2012, p. 311-330 ; J. JANSSENS, « The Liber primus naturalium, i.e. the Physics of the Avicenna Latinus », dans : Documenti e Studi sulla Tradizione Filosofica Medievale, 28 (2017), p 219-238. Pour d’autres titres et les traités II-V, v. la bibliographie déjà donnée dans : « Avicennisme Latin » cit. 22. Cf. AVICENNA LATINUS, Liber quartus Naturalium. De actionibus et passionibus, Édition Critique par S. VAN RIET, Introduction par G. VERBEKE, Louvain / Leiden 1989. Pour la description du contenu du ms., v. AVICENNA LATINUS, Codices descripsit M.-T. d’Alverny. Addenda collegerunt S. van Riet et P. Jodogne, Louvain / Leiden 1994, p. 86-87. Cf. H. BÉDORET, « Les premières versions tolédanes de philosophie. Œuvres d’Avicenne », dans : Revue néo-scolastique de philosophie 59 (1938), p. 374-400. Pour les autres parties, et Le Ciel et le monde en particulier, v. infra, note. 23. On compte 36 manuscrits et 112 manuscrits de l’Aristoteles Latinus ; v. la transcription par Holmyard-Mandeville in Avicennae de Congelatione et Conglutinatione Lapidum. Being sections of the Kitāb Al-Shifā, the Latin and Arabic texts, edited with an English translation of the latter and with critical notes by E. J. HOLMYARD […] and D. C. MANDEVILLE, Paris 1927 ; E. RUBINO, « Alfredo di Shareshill editore della meteorologia aristotelica », dans : Giornale critico della filosofia italiana (2015), p. 479-496 ; E. RUBINO – S. PAGANI, « Il De mineralibus di Avicenna tradotto da Alfredo di Shareshill», dans : Bulletin de philosophie médiévale 58 (2016), p. 23-87; cf. R. FRENCH, « Teaching Meteorology in Thirteenth Century Oxford: The Arabic Paraphrase», dans : Physis 36 (1999), p. 99-129.

PHILOSOPHER AU XIIe SIÈCLE

291

latine des Météorologiques d’Aristote, et devint ainsi le texte d’Avicenne le plus souvent copié24. Alfred devait être au courant de l’origine avicennienne du texte ; c’est toutefois Albert le Grand qui attribua avec certitude l’œuvre à Avicenne25. Ceci dit, une des traductions les plus importantes réalisées au XIIe siècle – sinon la plus importante – pour l’influence d’Avicenne en Occident (et l’idée, différemment modulée, d’un avicennisme latin qui lui sera liée) est celle du sixième livre de la Physique, Sur l’âme (Kitāb al-Nafs), le Liber de anima seu sextus de naturalibus. La traduction est encore une fois à situer à Tolède (entre 1152 et 1166 : elle est dédiée à l’archevêque Iohannes)26 et fut l’œuvre – ainsi nous informent les manuscrits – d’Avendauth et de Gundissalinus. Sa circulation fut importante (50 manuscrits avec 2 recensions de 31 et 19 mss27). Une traduction partielle de Al-adwiyya al-qalbiyya – De viribus cordis ou De medicinis cordialibus – fut ajoutée à ce texte ; cet ajout devint très célèbre, avec le Canon et le Poème de la médecine ou Cantica (al-Urjūza fī l-ṭibb : un manuel de médecine en poésie, dont la traduction est attribuée parfois à Gérard de Crémone)28. Aux textes de 24. M. MANDOSIO – C. DI MARTINO, « La Météorologie d’Avicenne (Kitāb al-Shifā’ V) et sa diffusion dans le monde latin », dans : A. SPEER – L. WEGENER (éds.), Wissen über Grenzen. Arabisches Wissen und lateinisches Mittelalter, Berlin 2006, p. 406-424. 25. MANDOSIO – DI MARTINO, La Météorologie d’Avicenne, p. 416. Cf. A. ALONSO, « Las traducciones de Juan Gonzales de Burgos y Salomon», dans : al-Andalus 14 (1949), p. 291-319, en particulier p. 306-308. Pour la traduction complète du texte, probablement à Burgos, entre 1274-1280, par Johannes Gunsalvi et Salomon, v. Libri Metheororum. 26. Sur la traduction du L. de anima seu sextus de naturalibus, réalisée à Tolède par Avendauth (Ibn Da‘ūd) et Gundissalinus entre 1152 et 1166, v. D. N. HASSE, Avicenna’s De Anima in the Latin West. The Formation of a Peripatetic Philosophy of the Soul 11601300, Londres / Turin 2000 ; un résumé et quelques titres dans : T. ALPINA, Subject, Definition, Activity. Framing Avicenna’s Science of the Soul, Berlin 2021, p. 27-28. Cf. aussi A.-S. JOUANNEAU, « La naissance tolédane de la ‘philosophie arabe’. Sur les intentions des traducteurs du Liber de Anima de l’Avicenne latin », dans : Les cahiers de l’Islam 1 (2014), p. 39-64. 27. AVICENNA LATINUS. Liber de anima seu sextus de naturalibus, IV-V, Édition Critique par S. VAN RIET, Introduction par G. VERBEKE, Louvain / Leiden, 1968 et I-III, 1972. Pour l’influence du Liber de anima sur le Moyen Âge latin, v. au moins HASSE, Avicenna’s De Anima in the Latin West ; ID., « The Early Albertus Magnus and His Arabic Sources on the Theory of the Soul », dans : Vivarium 46 (2008), p. 232-252 ; M. LENZI Anima, forma e sostanza: Filosofia e teologia nel dibattito antropologico del XIII secolo, Spoleto 2011. 28. Pour al-Adwiya al-qalbiyya (Remèdes pour les maladies cardiaques), v. S. VAN RIET dans: Avicenna Latinus 1968, p. 98*-99*, 116*-118*, 67 n. 65, 187-210 ; cf. T. ALPINA, « Al-Ǧūzğānī’s Insertion of On ‘Cardiac Remedies’ in Avicenna’s ‘Book of the Soul’: the Latin Translation as a Clue to his Editorial Activity on the ‘Cure’? », dans : Documenti e

292

OLGA L. LIZZINI

la physique29 il faut en effet ajouter les textes médicaux dont notamment le Canon de la Médecine (al-Qanūn fī l-ṭibb) traduit – selon l’attribution commune – par Gérard de Crémone, donc encore au XIIe siècle. Les doctrines médicales du Canon, qui fut un texte de référence pour l’enseignement en Europe jusqu’à la fin du XVIIe siècle30, – sont strictement liées à la pensée philosophique d’Avicenne ; l’exemple des propriétés occultes (I, 2, II, 1, chap. 15), qu’Avicenne explique par le biais des idées métaphysiques de la préparation et du flux, peut ici suffire pour l’illustrer31. Quant au Liber de anima, Avicenne y traite des doctrines qui relèvent du De anima d’Aristote, mais il y prend en considération Studi sulla Tradizione Filosofica Medievale 28 (2017), p. 365-400. Une traduction complète du De viribus cordis fut réalisée par Arnauld de Villeneuve (de Villanova) à Barcelone et fut comprise dans les éditions de textes publiés par Andrea Alpago. Mais il y eut peutêtre aussi une traduction tolédane de l’œuvre au XIIe siècle (ou peut être deux ; une traduction partielle et une traduction réalisée par Gérard de Crémone). Quant au Poème de la médecine, également publié par Alpago, il y eut plusieurs traducteurs ; ici on nomme Gérard de Crémone et Armengaud Blaise, dont la traduction circula beaucoup aussi à cause du commentaire d’Averroès qui l’accompagnait (Translatio canticorum Avicennæ cum commento Averrois, 1283) ; v. IBN SĪNĀ  (Avicenne), Urjūza fī-ṭ-ṭibb (Poème sur la médecine), introd. trad. et notes H. JAHIER et A. NOUREDDINE, Paris 1956, avec l’édition du texte et la traduction latine de Gérard de Crémone. 29. Le septième traité (sur Les Végétaux : al-Nabāt), mentionné dans le Catalogue de la Sorbonne de 1338 semble perdu ; le huitième livre sur les animaux (Liber de animalibus : al-Ḥayawān, comprenant Historia animalium, De partibus animalium, De generatione animalium) fut traduit dans une version abrégée (Abbreviatio Avicennae) par Michel Scot et probablement entre 1227 et 1236 (Venise 1508, ff. 29-64) ; la traduction, probablement réalisée à la cour de Frédéric II, pourrait avoir eu à la base un texte déjà abrégé ; pour une analyse en ce sens, v. A. M. I. VAN OPPENRAAIJ, « Michael Scot’s Latin Translation of Avicenna’s Treatise on Animals. Some Preliminary Remarks on the Future Edition », dans : R. BEYERS – J. BRAMS – D. SACRÉ – K. VERRYCKEN (éds.), Tradition et traduction. Les textes philosophiques et scientifiques grecs au Moyen Âge latin. Hommage à Fernand Bossier, Leuven 1999, p. 107-120. 30. Le Canon est organisé en cinq livres : I. Généralités de la médecine (al-Umūr al-kulliyya fī ‘ilm al-ṭibb) ; 2. Pharmacopée (médicaments simples : al-Adwiyya al-mufrada); 3.4. Pathologie spéciale et générale (i.e. les maladies des organes particuliers – al-Amrāḍ al-juz’iyya – et les maladies qui ne sont pas spécifiques d’un membre en particulier – al-Amrāḍ allatī lā takhtaṣṣ bi ’udw bi ‘aynihi) ; 5. Formulaire, i.e. les médicaments composés (al-Adwiyya al-murakkaba wa-l-aqrābādhīn). 31. Cf. N. WEILL-PAROT, Points aveugles de la nature. La rationalité scientifique médiévale face à l’occulte, l’attraction magnétique et l’horreur du vide (XIIIe-milieu du XVe siècle), Paris 2013 ; sur la pénétration des thèses médicales, v. aussi J. CHANDELIER – A. ROBERT, « L’anthropologie des médecins (IXe-XVIIIe siècle) », dans : Revue de Synthèse (2013) ; J. CHANDELIER, Avicenne et la médecine en Italie. Le Canon dans les universités (1200-1350), Paris 2017.

PHILOSOPHER AU XIIe SIÈCLE

293

aussi les discussions issues des Parva Naturalia – ou, plus précisément de son élaboration arabe – notamment en ce qui concerne la prophétie et les visions : la psychologie d’Aristote se trouve ainsi non seulement clarifiée et développée dans ses éléments fondamentaux (la définition de l’âme humaine comme substance et son rapport avec le corps, la théorie de l’intellect), mais aussi soudée à la métaphysique céleste, ce qui permet à la fois d’expliquer la gnoséologie, l’eschatologie et la prophétie elle-même. Plus en détail, c’est grâce à ce texte, que les penseurs du Moyen Âge latin purent s’approprier d’un critère logique qui leur permettait d’exprimer à la fois l’opérativité de l’âme par rapport au corps (en tant que forme et perfection) et sa substantialité, immatérialité et immortalité ; c’est grâce à ce texte, que les penseurs du Moyen Âge latin purent découvrir une analyse minutieuse des cinq sens externes (très importante fut la discussion de la lumière et de la vision) et un examen des cinq sens internes (le sens commun, l’imagination, la puissance imaginative-cogitative, l’estimative et la mémoire), qui se basait sur la distinction entre ‘formes’ (ṣuwar ; formae) et ‘intentions’ sensibles (ma‘ānī ; intentiones) : la connaissance sensible était ainsi pour la première fois considérée comme le moyen pour saisir le monde physique non seulement dans son caractère hylémorphique, mais aussi dans sa « signification », le ma‘nā (ce qui impliquait l’agir), des thèmes que l’on saura développer surtout au XIIIe siècle. 4. La Métaphysique – le Kitāb al-Ilāhiyyāt – autre texte dont l’importance est capitale – fut traduite dans les mêmes années du XIIe siècle (entre 1150 et 1175), par Gundissalinus et encore à Tolède. On possède deux recensions de la traduction latine (avec 15 et 10 manuscrits) et d’autres fragments encore32. Grâce au Liber de philosophia prima 32. La traduction latine fut éditée une première fois à Venise en 1495 et ensuite par les frères Augustiniens de San Giovanni in Verdara à Padoue, dans la collection d’œuvres avicenniennes imprimée à Venise, en 1508, par Ottaviano Scoto. Antérieurement à celle-ci, toujours à Venise, avaient été édités en 1495, la Métaphysique, et en 1507, le Canon ; pour la première, v. Avicenne perhypatetici philosophi : ac medicorum facile primi opera in lucem redacta: ac nuper quantum ars niti potuit per canonicos emendata. Logyca. Sufficientia. De celo et mundo. De anima. De animalibus. De intelligentijs. Alpharabius de intelligentijs. Philosophia prima (réimpression Minerva 1961). L’édition critique de Simone van Riet (avec une introduction doctrinale de G. Verbeke) fut publiée dans la collection de l’Avicenna Latinus dans les années 1977-1980, 1983 ; sur l’influence du texte dans le monde latin on ajoutera aux titres déjà cités, en général, G. C. ANAWATI, « La Métaphysique

294

OLGA L. LIZZINI

sive scientia divina les penseurs du Moyen Âge purent repérer – ou reconnaître et donc élaborer – différents éléments théorétiques fondamentaux par rapport à la compréhension des questions qui étaient pour eux capitales (dont plusieurs venaient des auteurs de l’Antiquité et de l’Antiquité tardive comme Aristote, bien sûr, mais aussi Proclus et Philopon) : la définition de la métaphysique en tant qu’étude de l’être et, en même temps, recherche de Dieu ; l’idée fondamentale d’une coupure ontologique entre Dieu et les créatures, voire la distinction, et donc la composition – absente chez le Principe Premier – de l’essence ou quiddité (māhiyya) et de l’existence (anniyya ou wujūd) dans les êtres créés ; une conception révolutionnaire de l’universel (et à ce propos on peut encore évoquer l’Isagoge et la doctrine de l’indifférence de l’essence), et d’éléments essentiels pour établir, d’un côté, la théorie du primum cognitum et, de l’autre, la conception de l’unité transcendantale33. On ne peut pas énumérer tous les éléments doctrinaux que les auteurs latins purent lire dans la Métaphysique d’Avicenne ; on mentionnera néanmoins l’analyse profonde et articulée de l’idée de la creatio ex nihilo et la théorie de l’émanation qui l’accompagne : non seulement on y trouve une angélologie, mais en outre elle est conçue comme la seule explication non-contradictoire de la dérivation du monde à partir de Dieu. Enfin, à travers l’analyse de l’être, la Métaphysique d’Avicenne offrit non seulement une ontologie couronnée par l’idée de la hiérarchisation des êtres (l’angélologie) et des sciences (la métaphysique gouverne le système du savoir), mais aussi une via maestra pour démontrer l’existence de Dieu. On comprend donc déjà par cette récapitulation sommaire pourquoi l’historiographie a souvent souligné l’entrée d’Avicenne en Europe par des termes grandiloquents : E. Gilson l’avait représentée comme d’Avicenne dans l’Occident latin », dans : AVICENNE, La Métaphysique du Shifā’. Livres I à V, Paris 1978, p. 56-79 ; G. VERBEKE, « Avicenna’s Metaphysics and the West », dans : M. WAHBA (éd.), Islam and Civilization, Le Caire 1982, p. 53-64 ; J. JANSSENS – D. DE SMET (éds.), Avicenna and his Heritage. Acts of the International Colloquium, Leuven / Louvainla-Neuve, September 8 – September 11, 1999, Leuven 2002 ; D. N. HASSE – A. BERTOLACCI The Arabic, Hebrew and Latin Reception of Avicenna’s Metaphysics, Berlin 2011. Une étude complète de son influence et, en général, de l’influence du Kitāb al-Shifā’, au Moyen Âge latin doit encore être réalisée. 33. A. DE LIBERA, « D’Avicenne à Averroès et retour. Sur les sources arabes de la théorie scolastique de l’un transcendental », dans : Arabic Sciences and Philosophy 4 (1994), p. 141-179 ; W. GORIS, Transzendentale Einheit, Leiden / Boston 2015.

PHILOSOPHER AU XIIe SIÈCLE

295

un « point de départ » ; P. Porro a parlé d’un « point de rupture » ; J. Jolivet d’un « tournant »34. 2. La réception. Les traductions et l’influence d’Avicenne Quel est le caractère que l’on peut en général attribuer aux textes qui constituent les sources de ces traductions ? On pourrait dire, premièrement, que les textes de la summa sont liés au schéma traditionnel de l’enseignement de la philosophie et des disciplines qui en étaient proches : ce n’est que dans le Pugio Fidei de Raimòn Martì (m. 1284, on est désormais au XIIIe siècle et chez un auteur pour ainsi dire « orientaliste »), qu’on trouve des traces de la connaissance de quelques ouvrages moins systématiques comme les Ishārāt wa-l-tanbīhāt – Alixarat ou Liber invitationum et exercitationum – et le Kitāb al-Najāt (Amuge)35. Cependant, la pénétration de la philosophie d’Avicenne au Moyen Âge latin, au XIIe siècle, ne fut pas seulement celle des œuvres qu’Avicenne lui-même avait conçues : en premier lieu, parce qu’il y eut une circulation indépendante de certaines parties de l’œuvre avicennienne (la Logique36, la philosophie naturelle37, 34. E. GILSON, « Avicenne et le point de départ de Duns Scot », dans : Archives d’Histoire Doctrinale et Littéraire du Moyen Âge 2 (1927), p. 89-149 ; P. PORRO, « Duns Scot et le point de rupture avec Avicenne », dans : O. BOULNOIS – E. KARGER – J.-L. SOLÈRE – G. SONDAG (éds.), Duns Scot à Paris. 1302-2002, Actes du Colloque de Paris 2-4 Septembre 2002, Turnhout 2004, p. 195-218 ; J. JOLIVET, « Le tournant avicennien », dans : M. LEJBOWICZ (éd.), Une conquête des savoirs. Les traductions dans l’Europe latine (fin XIe siècle - milieu XIIIe siècle). Actes du colloque organisé à la Fondation Singer-Polignac le jeudi 27 novembre 2008, Turnhout 2009, p. 83-99 (repris in J. JOLIVET, Medievalia et Arabica, Paris 2013, p. 329-342). 35. Parmi les auteurs arabes cités par Raimòn Martì on trouve al-Ghazālī et Fakhr al-Dīn al-Rāzī : v. A. CORTABARRÍA-BEITIA, « Les sources arabes de l’Explanatio simboli du dominicain catalan Raymond Martin », dans : Mélanges de l’Institut Dominicain d’Études Orientales 16 (1983), p. 95-116 ; ID., « Avicenne dans le Pugio Fidei de Raymond Martin », dans : Mélanges de l’Institut Dominicain d’Études Orientales 19 (1989), p. 9-16 ; D. TRAVELLETTI, Front Commun. Raymond Martin, al-Ġazālī et les philosophes. Analyse de la structure et des sources du premier livre du Pugio Fidei, Thèse de Doctorat, Université de Fribourg (Suisse), 2011. Cf. Ramón MARTÍ, Pugio fidei, Leipzig, Heirs of Friedrich Lankisch, at the press of the Johann Wittigav’s Widow 1687 ; Texte zur Gotteslehre. Pugio fidei I–III, 1–6. Lateinisch–Hebräisch / Aramäisch–Deutsch. Görge HASSELHOFF (trad.), Freiburg 2014 ; J. JANSSENS, « R. Marti and His References to al-Ghazālī», dans : G. TAMER, Islam and Rationality. The Impact of al-Ghazālī. Papers Collected on His 900th Anniversary, Leiden / Boston 2015, vol. 1, p. 326-344. 36. I. 12 à partir du Kitāb al-Madkhal sur les universaux, II 7 du Kitāb al Burhān. 37. I 1, I 5 et à part II. 6 du Kitāb al-Ma‘ādin wa-l-āthār al-‘ulwiyya.

296

OLGA L. LIZZINI

la métaphysique38) ; et en deuxième lieu, parce qu’il y eut aussi une circulation d’écrits apocryphes39. On se rappellera du fait que l’édition des Augustiniens en 1508 incluait des textes pseudo-avicenniens dont le Liber de causis primis et secundis40, qui – comme Alain de Libéra l’a écrit – contribua à « déformer les deux images d’Avicenne et de l’avicennisme, attribuant notamment à l’un ce qui appartenait à l’autre »41. La pénétration d’Avicenne en Occident au XIIe siècle concerna en somme aussi, ou en fait en premier lieu, ce qu’on était disposé et donc intéressé à lire chez Avicenne ou ce que – chez lui – on aurait voulu pouvoir découvrir. Ainsi, parmi les textes apocryphes on compte des œuvres d’alchimie – dont notamment le De anima in arte alchimiae42. 38. Ilāhiyyāt III.5 – Capitulum de certitudine quiditatis numeri – et le Tractatus V sur les universaux, qui a été conservé dans le seul ms. Vat. Lat. 2186, ff. 64v-70v, sous le titre : “Tractatus logicae”, Avicenna Latinus 1994, p. 92 ; cf. A. BERTOLACCI, « The Reception of Avicenna in Latin Medieval Culture », dans : P. ADAMSON (éd.), Interpreting Avicenna. Critical Essays, Cambridge 2013, p. 242-269. 39. G. STROHMAIER, « Avicenne et le phénomène des écrits pseudoépigraphiques », dans : JANSSENS – DE SMET (éds.), Avicenna and His Heritage, p. 113-122 ; sur le phénomène, v. D. REISMAN, « The Pseudo-Avicennan Corpus, I: Methodological Considerations », dans : J. MCGINNIS (ed.), Interpreting Avicenna. Science and Philosophy in Medieval Islam, Leiden 2004, p. 3-21. 40. Un autre texte aussi, tout en étant présenté comme le deuxième traité de la Physique (sous le titre De coelo et mundo) est apocryphe ; v. O. GUTMAN, Pseudo-Avicenna. Liber Celi Et Mundi: A Critical Edition with Introduction, Leiden / Boston 2003. Le texte a été l’objet de plusieurs études et l’idée d’Alonso selon laquelle le texte aurait été une paraphrase de Thémistius, due à Ḥunayn b. Isḥāq (m. 873) a été questionnée. Sur les hypothèses autour du texte et surtout sur le vrai texte avicennnien, traduit en latin non avant 1274, v. C. CERAMI, « The De Caelo et Mundo of Avicenna’s Šifā’: An overview of its goal, its structure and its polemical background », dans : Documenti e Studi sulla Tradizione Filosofica Medievale 28 (2017), p. 273-329. 41. DE LIBERA, « Avicennisme latin » dans : Encyclopaedia Universalis. 42. Cf. STROHMAIER, « Avicenne et le phénomène des écrits pseudoépigraphiques » : on a en fait trois écrits arabes apocryphes sur l’alchimie attribués à Avicenne et deux falsifications latines, pour une desquelles on peut supposer un original arabe (Epistula ad Regem Hasen de re recta et Liber de anima in arte alchimiae). Cf. Porta elementorum, traité de philosophie naturelle élémentaire préliminaire au De anima alchimique du pseudoAvicenne, qui démontre des liens étroits avec le De elementis de Marius. Sur ces thèmes, v. S. MOUREAU, « Elixir Atque Fermentum. New Investigations about the Link between Pseudo-Avicenna’s Alchemical De Anima and Roger Bacon: Alchemical and Medical Doctrines », dans : Traditio 68 (2013), p. 277-323 ; S. MOUREAU, « Physics in the Twelfth Century: The Porta Elementorum of Pseudo-Avicenna’s Alchemical De Anima and Marius’ De elementis », dans : Archives d’Histoire Doctrinale et Littéraire du Moyen Âge 80 (2013), p. 147-221 ; S. MOUREAU, « Some Considerations Concerning the Alchemy of the De anima in arte alchemiae of Pseudo-Avicenna », dans : Ambix 56 (2009), p. 49-56 ; S. MOUREAU, « Alchemy in the Latin world », dans : H. LAGERLUND (éd.), Encyclopedia

PHILOSOPHER AU XIIe SIÈCLE

297

Avicenne alchimiste est d’ailleurs particulièrement captivant, non seulement parce qu’il n’écrivit aucune œuvre d’alchimie, mais aussi parce qu’au contraire, il éprouvait une véritable aversion intellectuelle à l’égard de cette discipline – tout comme à l’égard de l’astrologie. Si Gundissalinus peut répertorier l’alchimie dans sa divisio scientiarum – « … alquimia, que est sciencia de conversione rerum in alias species », c’est qu’il reprend la liste du De ortu scientiarum, une élaboration latine qui avait été rédigée à partir de l’Enumération des sciences d’al-Fārābī43. Enfin, au moins un élément doit encore être souligné ; il concerne la connaissance d’al-Ghazālī (m. 1111), le grand théologien musulman. C’est en fait à lui aussi que l’on doit la fortune d’Avicenne. Au groupe des textes avicenniens et pseudo-avicenniens connus au XIIe siècle, il faut en effet ajouter la traduction latine des Maqāṣid al-falāsifa, que le Moyen Âge latin connut, encore une fois grâce à la traduction de Gundissalinus, sous le titre de Summa theoricae philosophiae ou de Logica44. La Logica d’al-Ghazālī, qui – comme Janssens l’a démontré – dépend probablement d’une version arabe du Dānesh Nāmeh, voire d’un texte qu’Avicenne aurait lui-même rédigé pour le traduire ensuite en persan45 – fut un vecteur extraordinaire pour la présentation et la pénétration des idées avicenniennes dans la pensée médiévale46. Dans le texte d’Algazel – pour les auteurs du Moyen Âge un disciple d’Avicenne et, comme son maître, lui-même une of Medieval Philosophy, Heidelberg 2011, p. 56-60 et l’édition du texte d’inspiration jābirienne : S. MOUREAU, Le De anima alchimique du pseudo-Avicenne, Florence 2016 ; pour J. RUSKA (dans : Isis 21 [1934], p. 14-51) le texte était un faux du début du XIIe siècle. 43. Le cas du Avicennae ad Hasen Regem Epistula de re recta – en arabe Risālat al-Iksīr – est plus délicat : selon Anawati l’œuvre pouvait bien être d’Avicenne ; cf. M. PEREIRA, « Teorie dell’elixir nell’alchimia latina medievale», dans : Micrologus 3 (1995), p. 103-48. 44. Ch. BURNETT, « Arabic into Latin: the Reception of Arabic Philosophy into Western Europe », dans : P. ADAMSON – R. TAYLOR (éds.), The Cambridge Companion to Arabic Philosophy, Cambridge 2005, p. 370-404. 45. V. J. JANSSENS, « Le Dānesh Nāmeh d’Avicenne: un texte à revoir? », dans : Bulletin de Philosophie Médiévale 28 (1986), p. 163-177 [repris dans: ID., Ibn Sīnā and His influence on the Arabic and Latin World, Ashgate Variorum, 2006] ; J. JANSSENS, « Al-Ġazālī and His Use of Avicennian Texts », dans : M. MARÓTH (éd.), Problems in Arabic Philosophy, Piliscaba 2003, p. 37-49 (repris dans : Ibn Sīnā and His influence on the Arabic and Latin World). 46. En présentant ainsi, aux Latins, un schéma (précisément celui du Dānesh Nāmeh) autre par rapport à celui du Kitāb al-Shifā’: la métaphysique, science de la réalité, précède ici la physique ; celle-ci étudie les réalités matérielles du monde sublunaire qui dérivent

298

OLGA L. LIZZINI

auctoritas – on trouve, par exemple, plusieurs fois le dator formarum, alors que dans la métaphysique d’Avicenne la locution ne se trouve que deux fois (dans IX, 5, 411, 9 et 413,11) ou trois si on considère aussi VI, 2, p. 265, 4, où la cause séparée est « occasio donatrix formarum » ; « la cause qui fait acquérir les formes » – al-sabab al-mufīd li-l-ṣuwar47. Pour décrire l’effet des textes d’al-Ghazālī on pourrait parler donc d’exaspération. En psychologie, les désignations ghazaliennes des facultés de l’âme, des cinq sens intérieurs ou des degrés de l’intellect (in potentia, in habitu, in effectu) ont souvent substitué les appellations avicenniennes, peut-être aussi parce que – comme l’a écrit A. de Libéra – leur clarté et leur cohérence étaient « nées d’un désir de simplification polémique »48. De fait, le texte des manuscrits latins du De anima d’Avicenne inclut parfois des gloses tirées d’al-Ghazālī, et le De anima d’Avicenne était souvent copié avec la traduction (complète ou partielle) de l’œuvre ghazalienne. 3. L’influence Les Latins furent donc introduits à la métaphysique et à la psychologie aristotéliciennes, sinon seulement, sans doute aussi, grâce à l’interprétation avicennienne49 ; c’est donc en ce sens que les textes d’Avicenne concoururent – avec ceux d’autres penseurs arabes, tels qu’al-Fārābī et al-Kindī – à la naissance d’une nouvelle façon de « philosopher »50. des êtres étudiés par la métaphysique tout comme les principes des sciences particulières dérivent des principes dont s’occupe la métaphysique. 47. Cf. Liber de philos. prima, p. 302, 24-25 : Ilāh., aussi X, 1, p. 439, 11-12. 48. DE LIBERA, « Avicennisme latin ». Mais d’autres éléments dérivent d’al-Ghazālī lecteur d’Avicenne ; le thème avicennien de la préparation de la matière devient à travers le théologien musulman un discours sur la noblesse des formes en proportion de la complexion matérielle ; v. M. LENZI, « Vile e gentile. La nobiltà in Dante tra generazione e predestinazione », dans : Studi e Materiali di Storia delle Religioni 86 (2020), p. 698-714, en particulier p. 702-703, et note 20. 49. Le Liber de anima seu sextus de naturalibus et le Liber de philosophia prima furent de facto utilisés par les Latins avant même les textes d’Aristote qui en sont pourtant les sources principales. 50. V. supra et B. NARDI, Sigieri di Brabante nella Divina Commedia e le fonti della filosofia di Dante, Spianate (Pescia) 1912 ; cf. Rivista di filosofia neoscolastica 3 (1911), p. 187195, 526-545 ; 4 (1912), p. 73-80, 225-239 ; E. GILSON, « Les sources gréco-arabes de

PHILOSOPHER AU XIIe SIÈCLE

299

Mais quels sont les textes où la métaphysique d’Avicenne est tout premièrement à reconnaître, au XIIe siècle? Il s’agit – les éléments se mêlent et les distinguer est en effet impossible – des mêmes textes que la critique a (ou avait) indiqués comme essentiels à la lecture ‘avicenniste’ de la philosophie. Il faut tout premièrement mentionner les textes philosophiques de Gundissalinus (m. après 1181), et, en particulier (à part le De anima), le De processione mundi et le De divisione philosophiae. Ce sont ces textes qui ont été principalement (même si pas uniquement) désignés comme « avicennistes »51. Il faut ensuite nommer le Liber de causis primis et secundis (connu aussi sous les titres De intelligentiis ou De fluxu entis ou encore Liber Avicenne in primis et secundis substantiis et de fluxu entis), le texte anonyme, probablement de la fin du XIIe siècle (ou le début du XIIIe siècle) que l’on a déjà évoqué parmi les pseudépigraphes. Et encore, il faut mentionner le texte présenté déjà aux débuts des années ’40 du XXe siècle par M.-Th. d’Alverny : le De peregrinationibus animae apud inferos52. Or, si on sort de la question historiographique de l’avicennisme, comment peut-on juger ces textes au XIIe siècle par rapport à la métaphysique d’Avicenne ? C’est en me posant cette question que je voudrais m’arrêter sur les exemples nommés. Voyons le De processione mundi de Gundissalinus53. Dans ce texte, qui est probablement un texte de la maturité de Gundissalinus, l’influence l’augustinisme avicennisant », dans : Archives d’Histoire Doctrinale et Littéraire du Moyen Âge 4 (1929-30), p. 5-149 ; DE LIBERA, « Avicennisme latin » ; P. PORRO, Prefazione a Avicenna. Metafisica, a cura di O. LIZZINI e P. PORRO, Milan 2006, p. V-XXXVI ; JOLIVET « Le tournant avicennien », p. 83-99 (repris dans JOLIVET, Medievalia et Arabica, p. 329342). Le De anima et la Metaphysica : à l’époque des traductions d’Avicenne, ces deux grands textes d’Aristote, qui avaient en effet déjà été traduits, ne jouissaient pas d’une grande circulation : v. BRAMS, La riscoperta di Aristotele. 51. R. DE VAUX, Notes et textes de l’avicennisme latin aux confins des XIIe-XIIIe siècles, Paris 1934 ; DE LIBERA, « Avicennisme latin ; L. SILEO, « L’Avicenna Latino al tempo dei divieti scolastici del 1210 e 1215. Storia e vicenda storiografica », dans : S. PERFETTI (éd.), Scientia, Fides, Theologia. Studi di filosofia medievale in onore di Gianfranco Fioravanti, Pise 2011, p. 131-171. 52. M.-Th. D’ALVERNY, « Les pérégrinations de l’âme dans l’autre monde d’après un anonyme de la fin du XIIe siècle », dans : Archives d’Histoire Doctrinale et Littéraire du Moyen Âge 13 (1940-1942), p. 239-299. 53. D. GUNDISSALINUS, De processione mundi, éd. G. BÜLOW, « Des Dominicus Gundissalinus Schrift Von dem Hervorgange der Welt », dans : Beiträge zur Geschichte der

300

OLGA L. LIZZINI

d’Avicenne est à reconnaître à plusieurs niveaux. En fait, si, comme la critique récente est en train de démontrer, les sources du texte sont nombreuses (il faut mentionner le De essentiis d’Herman de Carinthie ou Herman le Dalmate Albanois, un autre protagoniste de la rencontre du monde latin avec la culture arabo-islamique au XIIe siècle – il participa à la commission tolédane pour la traduction de textes islamiques –, le Commentarius in Timaeum de Chalcidius, les écrits de Boèce – De consolatione Philosophiae, De Trinitate, De hebdomadibus, De arithmetica – et même des suggestions de Thierry de Chartres et Guillaume de Conches), ses sources principales d’inspiration sont à rechercher chez Ibn Gabirol et chez Avicenne54. Plus précisément, le De processione mundi est une explication élaborée à partir de ces deux auteurs – Avicenne domine en ce qui concerne les prémisses ontologiques du traité – concernant les rapports entre Dieu et le monde créé. Bien que mélangées avec des éléments d’inspiration différente, les instances, pour ainsi dire, avicenniennes peuvent clairement être reconnues : à part le primat ontologique et non temporel de la création (un thème que l’on peut rapprocher de l’idée de la production de l’être chez Proclus et qui correspond aussi à une idée augustinienne) et l’angélologie, qui apparaît, à vrai dire, chez Gundissalinus, comme très complexe, il y a – c’est évident – la distinction entre le nécessaire et le possible. C’est en fait dans la distinction entre le Principe absolument nécessaire, un et simple, et le monde, qui dans les termes avicenniens est « double », que Gundissalinus repère un outil essentiel pour la distinction ontologique entre le Principe et le monde. On a parlé d’une technique d’assimilation et d’altération de Gundissalinus par rapport aux textes d’Avicenne : l’ordre des arguments est parfois changé, quelques mots sont substitués (necesse esse devient en général necessarium) et cela dans un contexte où les citations Philosophie und Theologie des Mittelalters 24/3 (1925), p. 1-56 ; D. GUNDISSALINUS, De processione mundi, ed. M. J. SOTO BRUNA – C. ALONSO DEL REAL, De processione mundi. Estudio y edición crítica del tratado de D. Gundisalvo, Pamplona 1999 ; D. GUNDISSALINUS, The Procession of the World, [trad. par] J. A. LAUMAKIS, Milwaukee 2002. Une analyse textuelle a été offerte par N. POLLONI, « Il De processione mundi di Gundissalinus: prospettive per un’analisi genetico-dottrinale », dans : Annali di Studi Umanistici 1 (2013), p. 25-38; ID., « Gundissalinus on Necessary Being: Textual and Doctrinal Alterations in the Exposition of Avicenna’s Metaphysics », dans : Arabic Sciences and Philosophy 26 (2016), p. 1-28. 54. Aux études déjà citées on ajoutera, A. FIDORA, « On the Supposed ‘Augustinisme Avicennisant’ of Dominicus Gundissalinus », dans : Veritas 47 (2002), p. 387-394.

PHILOSOPHER AU XIIe SIÈCLE

301

peuvent être très synthétiques (une phrase), avoir la longueur d’un paragraphe, ou finalement reproduire des chapitres entiers (dans le De processione mundi on retrouve Ilāhiyyāt I, 6, 7 comme dans le De divisione philosophiae, Burhān II, 7)55. Mais surtout, c’est la théorie gréco-arabe de l’émanation-descente d’une part et de l’ascèse de l’autre, qui pénètre le traité56. La métaphysique, qui est bien sûr le sujet principal de l’œuvre, est présentée dans son moment ascendant et dans son moment descendant, ce qui est en accord avec la méthode analytique que l’on trouve appliquée dans le texte. L’être y est qualifié comme en soi possible ou nécessaire ; or, seul un être nécessaire – dont le texte va démontrer, avec Avicenne (Ilāh. I, 6-7), l’unicité et l’unité – peut être une cause. C’est ainsi, une fois établie la nécessité de la Cause, que le texte déploie une métaphysique du monde créé à partir de la matière et de la forme. Si celles-ci sont créées de nihilo et en dehors du temps, tout ce qui se compose des deux principes est formé à partir d’une composition ou bien d’une mixtion ou encore d’une génération. En ce sens, c’est l’interprétation de la possibilité en tant que potentialité et donc en tant que substrat du passage potentiel-actuel qui permet à Gundissalinus de souder l’hylémorphisme universel d’Ibn Gabirol à la théorie de l’ontologie modale d’Avicenne57. Le Principe est le seul être un, et absolument nécessaire et simple, alors que tout ce qui n’est pas le Principe se doit à la composition de possibilité, voire potentialité ou matière, et forme, et cela, même si certaines compositions donnent lieu à des êtres incorruptibles comme les entités invisibles, les corps célestes et les quatre éléments. L’insistance sur un hylémorphisme foncier doit évidemment être ramené au Fons Vitae d’Ibn Gabirol58 ; 55. POLLONI, « Il De processione mundi di Gundissalinus ». 56. V. au moins M. CHASE, « Quod est primum in compositione, est ultimum in resolutione. Notas acerca de las nociones de ‘análisis’ y ‘síntesis’ en la antigüedad tardía », dans : Anuario Filosofico 48 (2015), p. 103-139. 57. V. POLLONI, Gundissalinus on Necessary Being. Sur le nécessaire, v. aussi R. FONTAINE – St. HARVEY « Jewish Philosophy in the Eve of the Age of Averroism: Ibn Daud’s Necessary Existent and His Use of Avicennian Science », dans : P. ADAMSON (éd.), In the Age of Averroes: Arabic Philosophy in the Sixth/Twelfth Century, Londres 2011, p. 215-228. 58. Pour le Fons Vitae, v. AVICEBRON, La fonte della vita ; Avencebrolis (Ibn Ǧabirol) Fons Vitae, ex Arabico in Latinum Translatus ab Johanne Hispano et Dominico Gundissalino, ed. C. BAEUMKER – G. v. HERTLING, Münster 1892. Sur l’auteur, v. au moins S. PESSIN, Ibn Gabirol’s Theology of Desire. Matter and Method in Jewish Medieval Neoplatonism, Cambridge 2013. Sur la question, N. POLLONI, « Toledan Ontologies: Gundissalinus, Ibn Daud, and the Problems of Gabirolian Hylomorphism », dans : A. FIDORA –

302

OLGA L. LIZZINI

et pourtant, la discussion terminologique sur la matière vient elle aussi – il me semble – directement d’Avicenne59. D’ailleurs, la matière indique chez Avicenne la potentialité, en général, mais aussi la disposition, la préparation spécifique à la réception de la forme. Le causé, qui se trouve du côté de la matière, ne peut absolument se substituer à l’action de la cause (qui est forme et être), mais la métaphysique d’Avicenne offre aux Latins la possibilité de souder les idées augustiniennes des rationes seminales avec l’idée aristotélicienne de la potentialité60. Enfin, le principe avicennien ab uno simplici non venit nisi unum est utilisé par Gundissalinus dans le De processione mundi et c’est en ce sens que le De unitate et uno – où l’influence d’Avicenne est moins sensible – peut être défini comme le texte qui contient, pour ainsi dire, les prémisses hénologiques de la cosmogonie du De processione61. Le texte offre une analyse de l’unité (et des différents genres d’unité), et insiste sur l’idée néoplatonicienne (et proclusienne en particulier) de l’un en tant que principe de l’être. L’unité – qui correspond à la forme et s’oppose donc, en un certain sens, à la matière – est, dans toutes les choses, le vrai principe de l’être et ressemble – en tant que telle – à la lumière. Mais l’influence d’Avicenne est ici moins évidente (la matière deffluit et ce motif proclusien semble presque s’opposer à la conception avicennienne) ; l’inspiration principale du texte a été décelée dans l’ontologie de Ibn Gabirol62. N. POLLONI (éds.), Appropriation, Interpretation and Criticism: Philosophical and Theological Exchanges between the Arabic, Hebrew and Latin Intellectual Traditions, Barcelone / Rome 2017, p. 19-49 ; N. POLLONI, « Gundissalinus on Necessary Being: Textual and Doctrinal Alterations in the Exposition of Avicenna’s Metaphysics», dans : Arabic Sciences and Philosophy 26 (2016), p. 1–28 ; N. POLLONI, « Gundissalinus and Avicenna: Some Remarks on an Intricate Philosophical Connection », dans : Documenti e Studi sulla Tradizione Filosofica Medievale 28 (2017), p. 515-552. 59. Cf. De processione mundi ed. BÜLOW, p. 31, 3-14 et AVICENNA LATINUS, Liber primus …, p. 21, 60-22, 73 60. V. encore POLLONI, The Twelfth Century Renewal of Latin Metaphysics. 61. D. GUNDISSALINUS, De unitate et uno, éd. M. J. SOTO BRUNA – C. ALONSO DEL REAL, De unitate et uno de Dominicus Gundissalinus, Pamplona 2015 ; cf. M. J. SOTO BRUNA, « La ‘causalidad del uno’ en Domingo Gundisalvo », dans : Revista española de filosofía medieval 21 (2014), p. 447-455 et A. DE LIBERA, « Ex uno non fit nisi unum. La Lettre sur le Principe de l’univers et les condamnations parisiennes de 1277 », dans : B. MOJSISCH, O. PLUTA (éd.), Historia Philosophiae Medii Aevi. Studien zur Geschichte der Philosophie des Mittelalters, Amsterdam / Philadelphia 1991, p. 543-560. 62. V. par ex. N. POLLONI, « Il De processione mundi di Gundissalinus: prospettive per un’analisi genetico-dottrinale », dans : Annali di Studi Umanistici 1 (2013), p. 25-38.

PHILOSOPHER AU XIIe SIÈCLE

303

L’élément foncier ou si l’on préfère le dénominateur commun sous lequel l’influence de la métaphysique d’Avicenne au XIIe siècle peut être décelée dans ces textes me semble en effet être celui de la procession elle-même, de l’émanation qui s’accompagne souvent de ce que l’on pourrait appeler une « narration philosophique » (ou un schéma, une structure métaphysique de fond) et qui renvoie à Avicenne là même où les éléments doctrinaux fondamentaux de la pensée avicennienne semblent de fait absents ou mal compris. C’est certainement le cas du De fluxu entis, le texte que Roland de Vaux considérait comme un témoin explicite d’un courant érigénien, où tous les emprunts n’étaient, selon lui, réalisés qu’au seul bénéfice d’Avicenne63. Le thème de l’écrit est l’explication de la relation entre Dieu et le monde, une relation qui est conçue en tant qu’émanation, comme le révèle clairement déjà le titre : le terme fluxus étant d’ailleurs fortement suggestif des thèmes gréco-arabes et d’Avicenne en particulier. Or, comme la critique l’a déjà souligné, les éléments de l’émanatisme avicennien que l’on peut isoler dans ce texte sont plusieurs mais n’ont de sens que grâce à l’utilisation d’autres sources néoplatoniciennes d’inspiration chrétienne, notamment Augustin, Scot Érigène et Grégoire de Nysse. Dans le texte, par exemple, on ne reconnaît pas la distinction logique et ontologique entre nécessité et possibilité (ni celle qui sépare l’essence de l’existence) – voire les distinctions qui constituent le véritable pilier théorique du concept de l’émanation chez Avicenne et qui sont centrales pour l’ensemble de son système. Le processus de l’émanation triadique est néanmoins repris, presque littéralement ; il ne s’agit donc pas de sa structure logique, mais de sa « narration » métaphysique : au Principe des principes suit une cause, qui est le premier créé à savoir, un intellect, et c’est toujours à partir d’un acte de compréhension intellectuelle que les intelligences célestes proviennent, toutes causées, selon le schéma de la chaîne, le niveau inférieur étant celui de l’intelligence ou de l’intellect agent séparé responsable du monde humain. C’est l’activité – et la donation64 – intellectuelle de chaque intelligence qui assure ainsi la procession de l’intelligence qui lui est inférieure et, 63. R. G. DE VAUX, Notes et textes de l’avicennisme latin aux confins des XIIe-XIIIe siècles, Paris 1934. 64. La donation et l’émanation renvoient au même mécanisme et les termes qui les indiquent sont presque des synonymes.

304

OLGA L. LIZZINI

après elle, la forme et le corps de la sphère céleste qui lui appartient65. La création conçue en termes de procession intellectuelle (qui constitue la matière du quatrième chapitre de l’œuvre), laisse reconnaître l’idée avicennienne des intelligences et l’identification entre ce qui est en soi possible et la forme des cieux66 ; chez Avicenne, plus précisément, c’est le possible en tant que nécessaire par autrui qui explique la forme, alors que le possible en soi est la raison du corps des cieux. La procession des intelligences qui est triadique – comme chez Avicenne (« sub omni intelligentia significantur tria in esse »67), – débouche sur l’intelligence « quae est agens in animas nostras » et cela aussi apparaît comme une citation du Liber de philosophia prima avicennien68. Toujours selon les termes avicenniens, les Intelligences et les Âmes des cieux sont assimilées aux Anges69. Encore (on est dans le cinquième chapitre), l’influence d’Avicenne est évidente en ce qui concerne les formes, alors que la discussion à propos de la matière renvoie – une fois de plus – à Scot Érigène70. L’hylémorphisme qui explique le monde de la nature est en fait érigénien, mais le principe avicennien (et plus en général néoplatonicien) selon lequel c’est la forme qui donne existence à la matière est maintenu. D’autres éléments s’expliquent à partir d’autres sources néoplatoniciennes mais se confirment à la lumière de la philosophie avicennienne : dans le sixième chapitre71, par exemple, l’origine de la multiplicité dérive essentiellement du dualisme de l’intelligence (cf. Avicenne et le Liber de Causis). Et le thème de l’unité de l’intelligence et la triade – Intelligence, Âme et Nature – qui constitue le thème du septième chapitre72 et a son explication dans le Liber de Causis et le Fons Vitae (dans le De Causis la création appartient au premier être 65. Comme DE LIBERA (« Avicennisme latin ») l’écrivit, ce « joyau de l’intertextualité médiévale » propose une synthèse de toutes les formes de néoplatonisme » ; les Soliloques d’Augustin (I, 8, 15) ont ici la fonction d’expliquer « la psychologie avicennienne de la connaissance lue à travers Denys et Érigène », mais le néoplatonisme gréco-arabe d’Avicenne sert à son tour à confirmer le néoplatonisme chrétien ; plus en détail, v. LIZZINI, « Avicennisme latin ». 66. DE VAUX, Notes et textes, p. 97-102, en particulier p. 101. 67. DE VAUX, Notes et textes, p. 102 ; cf. Ilāh., IX, 4, p. 406, 13-407, 4. 68. Ilāh., IX, 5, p. 410, 12-411, 4 ; Liber de philos. prima, p. 489. 69. Ilāh., X, 1, p. 435, 6-9 ; la trad. latine in Liber de philos. prima, X, 1, p. 522. 70. DE VAUX, Notes et textes, p. 102-108. 71. DE VAUX, Notes et textes, p. 108-112. 72. DE VAUX, Notes et textes, p. 113-117.

PHILOSOPHER AU XIIe SIÈCLE

305

qui donne l’existence, l’information est le propre de l’intelligence et de la vie, qui donnent la science et le mouvement vital), se trouve comme authentifié par Avicenne, notamment en ce qui concerne la notion de nature (universelle) qui domine la succession de génération et de corruption des êtres vivants. C’est en fait – il me semble – encore une « narration métaphysique » qui peut se lire dans Les pérégrinations de l’âme, le texte édité par Marie-Thérèse d’Alverny, une allégorie philosophique composée en latin, peut-être à la cour de Sicile, peut-être en Catalogne, et datée également de la fin du XIIe siècle ou du début du XIIIe. L’allégorie décrit un voyage que l’âme réalise à partir de ses propres dispositiones. Dans l’âme humaine le texte distingue, en fait, une pars dextera, une pars sinistra et une dispositio media. La dispositio dextera que et paradisius dicitur est divisée en dix ordres, qui se révèlent utiles pour expliquer le voyage de l’âme à travers les dix felicitates du paradis. À chaque ordre de bonheur correspond un degré particulier de joie ou de plaisir. Après dix états de bonheur, l’âme parcourt les dix intelligences « ad similitudinem celorum », chaque intelligence correspondant à l’un des neuf ordres des anges, jusqu’au dixième « qui non est substantia sed essentia prima et est Deus ». De la même manière la « dispositio sinistra que et infernus et tormentum dicitur »73 comprend dix ordres correspondant à autant d’étapes de l’âme. Il y a dix miseriae infernales qui, comme dans l’eschatologie qu’Avicenne décrit dans sa Métaphysique, correspondent aux sphères planétaires74. En donnant l’édition du texte, M.-Th. d’Alverny avait déjà mis en évidence que « le véritable climat » du texte était à rechercher dans la Métaphysique d’Avicenne comme dans la Métaphysique d’al-Ghazālī, 73. Une analogie dans l’Épître des prophéties, attribuée à Avicenne et probablement apocryphe ; un texte inconnu au Moyen Âge latin. 74. D’ALVERNY, « Les pérégrinations de l’âme », p. 294-297. Avicenne (v. par ex. Métaphysique X) place le paradis et l’enfer – s’ils sont corporels, c’est-à-dire conformes à la Loi religieuse – au niveau des sphères célestes, tandis qu’il attribue le paradis et l’enfer spirituels à une dimension intellectuelle ou métaphysique. M-Th. d’Alverny référait ce discours (« Les pérégrinations de l’âme », p. 261) à « La Kaçidah d’Avicenne sur l’âme » (v. B. CARRA DE VAUX, « La Kaçīdah d’Avicenne sur l’âme », dans : Journal asiatique 9/14 (1899), p. 157-173), où on trouvait l’idée de la chute de l’âme, comme si la matérialité elle-même était un enfer. Pour la Qaṣīda fī l-nafs, dont l’attribution à Avicenne est douteuse, v. D. GUTAS, Avicenna and the Aristotelian Tradition. Introduction to Reading Avicenna’s Philosophical Works. Second, Revised and Enlarged Edition, Including an Inventory of Avicenna’s Authentic Works, Leiden 2014, p. 453-456.

306

OLGA L. LIZZINI

c’est-à-dire sa Logica, et dans les textes du néoplatonisme arabe (ou gréco-arabe) qui lui sont proches et donc, principalement : le Liber de causis (cité dans le texte comme Liber de bonitate pura) et le Fons vitae d’Ibn Gabirol. En ce qui concerne notamment Avicenne (le Liber de philosophia prima, mais aussi son Liber de anima), l’influence est évidente dans le thème éthique et psychologique des dispositions de l’âme. La distinction dextera et sinistra rappelle, en fait, la doctrine des deux faces de l’âme (un autre motif de l’influence avicennienne dans le milieu franciscain) qui regardent, l’une, au monde supérieur, l’autre, au monde d’ici-bas. La dispositio moyenne concerne l’âme en tant que liée au corps ; elle semble être proche de la valeur de la vertu chez Avicenne et rappelle donc enfin la conception aristotélicienne du juste milieu75. Mais ce qui est ici fondamental, c’est le lien que le texte établit ou, pour mieux dire, considère comme établi entre la psychologie de l’homme et la métaphysique. C’est le Liber de anima d’Avicenne qui permet de souder la psychologie à la métaphysique. On comprend en ce sens les similitudes des felicitates de l’âme qui appartiennent à l’eschatologie : pour Avicenne, l’âme vertueuse ou parfaite doit dominer les passions du corps, recevoir le flux divin ou encore être satisfaite et calme : prima [...] ut sit anima potens; secunda felicitas [...] ut sciat; tertia [...] ut sit anima virtuosa; quarta ut sit anima luminosa; quinta ut sit anima sedens firmiter locata quod est dicere quiescens ab omni labore; sexta est ut dominetur anima sibi ipsi et obediat sibi ipsi et ut dominans et dominatum sint idem; septima est ut sensu intellectibili sentiat anima fluxum Luminis a divina essentia; octava est ut sit anima picta et formata et figurata omnibus picturis et formis et figuris omnium substantiarum et accidentium [...] et sciat et contempletur ordinem totius veritatis et esse in se ipsa impressum et simulacratum; nona est ut sit diligens has formas et figuras; decima est ut sit anima non comprehendens sed apprehendens in esse suo esse infinitum)76. 75. Ethica vetus (livres II et III de l’Éthique à Nicomaque), Ethica nova (livre I), et ce qu’on appelle Ethica Borghesiana (VII-VIII fragments), sont toutes réalisées à partir du grec au XIIe siècle. 76. D’ALVERNY, « Les pérégrinations de l’âme, p. 285-291. L’enchaînement des gradations de perfection de l’âme énumérées dans le texte rappelle les distinctions avicenniennes : l’âme est puissante (potens) car en premier lieu elle est aussi une puissance (cf. Liber de anima, I, 1) ; elle connaît (elle est sciens), parce que sa perfection spéculative est celle du savoir ; elle est vertueuse, car sa pleine perfection intellectuelle ne peut être séparée de la perfection morale (v. par exemple les passages du Liber de philos. prima, p. 515-517 =

PHILOSOPHER AU XIIe SIÈCLE

307

Mais, surtout, pour Avicenne la perfection de l’âme rationnelle consiste dans la pleine réalisation de l’intellectualité qui, de l’âme, constitue la nature seulement dans le sens qu’elle en constitue l’horizon (l’âme humaine n’est pas une intelligence séparée, mais précisément elle tend à devenir une intelligence). En ce sens, la réalisation de l’intellectualité de l’âme est configurée chez Avicenne comme un « chemin », comme un « voyage » presque, qui a son point culminant dans le fait de devenir un monde et donc un miroir du monde où les formes des choses se trouvent estampillées ou dessinées (comme si elle était une anima picta) et qui s’explique en vertu du lien entre psychologie et métaphysique que l’on a déjà nommé77. D’ailleurs, le thème du voyage de l’âme qui apparaît en ce sens dans la Métaphysique du Kitāb al-Ilāhiyyāt, est un élément récurrent dans la psychologie d’Avicenne ; il est au cœur de ce qu’Henry Corbin avait appelé « le cycle de récits visionnaires », dans lequel est inclus Le récit de Ḥayy ibn Yaqẓān (Qiṣṣat Ḥayy ibn Yaqẓān)78, un texte qui exerça son influence même sur le monde juif : il fut élaboré, encore au XIIe siècle, par le philosophe, homme de science et exégète Abraham ibn Ezra (1089-1164 ou 1167)79 et se retrouve – même si différemment décliné – dans l’Épître du Ḥayy ibn Yaqẓān d’Ibn Ṭufayl80.

Ilāh., IX, 7, p. 429-430) ; elle est lumineuse, c’est-à-dire « éclairée » car sa réalisation maximale vient de l’illumination ; elle est tranquille et apaisée (quiescens) – peut-être parce que, l’âme qui connaît est apaisée (al-nafs al-mutma’inna), ou l’âme qui a dominé ses passions est apaisée ; elle est dominans parce qu’elle domine les passions et donc en un certain sens elle-même (elle est en fait dominans et dominata) ; elle reçoit le flux divin qui est aussi un flux de lumière ; elle est un monde ou un miroir dans lequel il y a le dessin des intelligibles (anima picta : contemplans ordinem totius veritatis et esse in se ipsa impressum) ; et elle est donc une âme qui aime (diligens) et qui en elle apprend (anima apprehendens) l’essence de toutes choses. 77. Ilāh., IX, 7, p. 425-426. Pour l’âme du monde, v. Liber de philos. prima, p. 433, 90 et sqq. et p. 510, 72 et sqq. (cf. Ilāh., VIII, 7, p. 370 et IX, 7, p. 425-426) ; v. O. LIZZINI, « The Pleasure of Knowledge and the Quietude of the Soul in Avicenna (Ilāhiyyāt, VII, 7 and IX, 7) », dans : Quaestio 15 (2015), p. 265-273. 78. Avec L’Épître de l’oiseau (Risalat al-Ṭayr) et l’histoire de Salaman et Absal (Salamān wa Abṣāl). 79. Sur Abraham Ibn Ezra ou Esra, v. au moins T LANGERMANN, « Abraham Ibn Ezra », dans : The Stanford Encyclopedia of Philosophy (Winter 2016 Edition) : URL = . 80. L. GAUTHIER, Roman philosophique de Ḥayy ibn Yaqẓān, Beyrouth 1936 ; repr. Beyrouth / [Paris] 1981 ; IBN ṬUFAYL, Le philosophe sans maître. Traduction d’Étienne Marc de Quatremère, édition du texte, note et préface de J. B. Brenet, Paris 2021.

308

OLGA L. LIZZINI

J’ai mentionné une narration métaphysique. On peut maintenant essayer de préciser le sens de cette locution. Toute générale et en même temps partielle qu’elle soit, elle permet de souligner en quel sens les premiers textes « métaphysiques » du XIIe siècle que l’on a mentionnés utilisent la philosophie d’Avicenne. Plus spécifiquement, la philosophie d’Avicenne est reçue en tant que philosophie qui pose la métaphysique au fondement de l’analyse physique et psychologique81. Avicenne offre aux auteurs du XIIe siècle – au moins dans les exemples que l’on a vus – une idée de métaphysique en tant qu’ontologie. Le sujet de la science « des choses divines » (al-Ilāhiyyāt), en soi séparées de la matière dans la définition comme dans l’existence, est l’ens inquantum est ens, l’existant en tant qu’existant (l’être en tant qu’être : al-mawjūd bi-mā mawjūd), et si, dans son De divisione, Gundissalinus interprète la métaphysique en un sens ontologique, c’est qu’il reprend la définition avicennienne. De plus, Avicenne offre aux auteurs du XIIe siècle une idée de métaphysique en tant que discipline, et Gundissalinus semble être le premier à utiliser le terme dans un sens disciplinaire82. En outre, la métaphysique est pour Avicenne une discipline qui gouverne les autres, qui lui sont toutes soumises (l’idée de la subalternatio se trouve déjà dans le De divisione scientiarum qui reprend la Logique d’Avicenne) ; la métaphysique, dont le sujet est le plus général est aussi la science qui légitime les principes des autres. Enfin, la métaphysique avicennienne inclut l’hylémorphisme d’Aristote dans un schéma vertical et offre ainsi, d’un côté, une explication transcendante de la nature (et en ce sens va-t-on aussi corroborer l’idée classique de l’historiographie selon laquelle le XIIe siècle est le siècle qui découvre la Nature)83 et, de l’autre, une narration dont l’élément clé est l’émanation. 81. M. Th. D’ALVERNY, « Une rencontre symbolique de Jean Scot Érigène et d’Avicenne. Notes sur le “De Causis primis et secundis et fluxu qui consequitur eas” », dans : L. BIELER – J. O’MEARA (éds.), The Mind of Eriugena. Papers of a Colloquium, Dublin 14-18 July 1970, Dublin, 1973, p. 170-181 ; A. FIDORA, « The introduction of metaphysics in the Latin West », dans : The Review of Metaphysics 66 (2013), p. 691-712. 82. FIDORA, « The introduction of Metaphysics in the Latin West ». 83. T. GREGORY, « Natur im Mittelalter (XII Jahrhundert). Natur II. Frühes Mittelalter », dans : J. RITTER – K. GRÜNDER – G. GABRIEL (éds.), Historisches Wörterbuch der Philosophie, Basel 1984, p. 441–447 ; cf. A. MAIERÙ, « Natur III. Hochmittelalter », ibid., p. 447-455 : les idées avicenniennes de nature universelle et nature particulière jouent un rôle au XIIIe siècle.

PHILOSOPHER AU XIIe SIÈCLE

309

Dans un contexte où on avait déjà modulé la philosophie du Moyen Âge (avec les analyses d’Ernest Renan, de Franz Ehrle, de Pierre Mandonnet, de Bruno Nardi), Étienne Gilson proposa dans son étude de 192684 l’idée de « l’augustinisme avicennisant » à laquelle il opposait celle d’ « augustinisme aristotélisant ». Si les auteurs impliqués étaient surtout des penseurs du XIIIe siècle (au XIIe on trouve essentiellement Gundissalinus avec son De anima et, plus tard, le De fluxu entis), c’est que, selon Gilson, c’était à chaque fois essentiellement la place attribuée à l’intellect agent qui permettait de distinguer les auteurs selon les deux nuances de la catégorie évoquée (l’augustinisme aristotélisant ou avicennisant) : tout en préservant la théorie augustinienne de l’illumination (ce qui l’autorisait à parler d’ailleurs d’augustinisme), les premiers penseurs reconnaissaient à l’âme humaine dans sa singularité l’activité de l’intellect (qui était en ce sens « agent »), alors que les autres, qui suivaient de près Avicenne, séparaient l’âme humaine de l’intellect agent, qui était « actif » ou « agent » au-delà de la dimension humaine. Toute la question de l’augustinisme avicennisant (et de l’avicennisme) n’était, en effet, pour Gilson, qu’une lecture de la réception de la psychologie d’Aristote dans le cadre de la doctrine augustinienne de l’illumination. Contre cette idée de l’augustinisme avicennisant le père Roland de Vaux proposa la catégorie de l’avicennisme. Les Notes et textes sur l’avicennisme latin aux confins des XIIe-XIIIe siècles furent publiées en 1934. Avec l’édition de deux textes que l’on a déjà mentionnés – l’anonyme De causis primis et secundis et de fluxu qui consequitur eas, et la fin du De anima attribué à Gundissalinus – de Vaux y présentait une étude (les Notes), rédigée en sept brefs chapitres, où il s’efforçait de défendre l’existence d’un véritable « avicennisme latin » qu’il situait avant l’entrée en Occident de la pensée d’Averroès et la « crise » qui en fut la conséquence. Gundissalinus et l’auteur anonyme du De fluxu entis auraient été témoins de l’existence des véritables avicennistes (qu’on aurait donc dû distinguer des avicennisants85), qui, à partir de la seconde moitié du XIIe siècle, avaient estimé pouvoir reconnaître chez Avicenne la conception d’une harmonie foncière entre pensée philosophique et pensée religieuse. 84. GILSON, « Pourquoi Saint Thomas a critiqué Saint Augustin ». 85. DE VAUX, Notes et textes, p. 38.

310

OLGA L. LIZZINI

J’ai essayé ailleurs de présenter l’histoire, les enjeux et les critiques des notions d’augustinisme avicennisant et d’avicennisme (aux noms de Gilson et de de Vaux il faut en fait en ajouter d’autres). Je ne vais pas m’y arrêter à nouveau. En l’état actuel des recherches – et c’est sur ce point que je voudrais conclure – ce qu’on peut suggérer c’est une distinction fondamentale dans la réception de la philosophie avicennienne entre le XIIe et le XIIIe siècle. Le premier représente le moment des traductions, de l’entrée d’Avicenne en Occident et donc – sans vouloir réduire l’impact d’Avicenne sur le Moyen Âge latin ni aux limites d’un courant, ni à celles d’un siècle – de « l’avicennisme » de de Vaux, de « l’avicennisme augustinisant » de Jolivet86 ou de l’« avicennismo agostinistico » que Bruno Nardi proposait déjà en 191287 ; ou – dans les termes que j’ai évoqués ici – d’un « avicennisme » où le thème de la « narration métaphysique » est central. Le second est le siècle qui put intégrer la philosophie d’Avicenne même en dehors du contexte franciscain ou augustinien et qui, justement, en l’intégrant et en l’élaborant, retint d’Avicenne ce qui pouvait servir à une cause théorétique autonome (même si limitée par l’adhésion au modèle de la foi) et proposa la discussion des thèmes avicenniens (l’être, la quiddité, la nature, la matière), même de façon indépendante de cette (ou de toute) « narration ». Olga L. LIZZINI Faculté des Arts, Lettres, Langues et Sciences humaines, Département des Études Orientales - Université Aix-Marseille 29 avenue Robert Schuman F-13621 Aix-en-Provence [email protected]

86. J. JOLIVET, « L’augustinisme avicennisant au XIIe siècle : un effet de mirage », dans : Chora 2 (2004), p. 5-20 [repris dans: JOLIVET, Medievalia et Arabica, p. 313-327]. 87. NARDI, Sigieri di Brabante ; E. BERTOLA, « È esistito un avicennismo latino nel Medio Evo ? », dans : Sophia 35 (1967), p. 318-334 ; 39 (1971), p. 278-320, en particulier p. 306, pour cette expression à propos de Guillaume d’Auvergne.

HENRY BATE ON AN APPARENT CONTRADICTION IN AVICENNA’S DOCTRINE OF THE SOUL Carlos STEEL Abstract In the third Part of his philosophical encyclopedia Speculum divinorum, Henry Bate points to an “apparent contradiction” in Avicenna’s views on the soul. Whereas he generally subscribes to Aristotle’s definition that the human soul is the first perfection of an organic body, he also claims that it is not related to the body as its form. In this contribution, I examine with Henry Bate how we can explain this apparent contradiction. Bate’s reference to Avicenna comes in his search for a middle position between the opposing views on the soul of Siger of Brabant and Thomas Aquinas. To find a way out of this fruitless debate he insists on the fundamental difference between two modes of formal causality. Whereas the material form is nothing but the act of its substrate-matter, the immaterial form communicates its being to matter in such a way that it remains self-subsistent. The ambiguity in Avicenna regarding the definition of the soul as “form” can be overcome if one distinguishes between two radically different modes of formal causality.

Scholars often have noticed that Avicenna seems to be reluctant to consider the human soul in its essence as the form of the body and prefers “perfection” over “form” to characterise the soul. Avicenna’s interpretation of the Aristotelian definition of the soul stands in a long tradition of reading Aristotle in a (Neo)-Platonic perspective. Interestingly, medieval authors, when reading Aristotle, referred to Avicenna’s authority to qualify Aristotle’s definition, most conspicuously Albert the Great.1 In this contribution for our Avicennian friend Jules, I intend to examine a less noticed discussion of Avicenna’s 1. On the early reception of Avicenna’s treatise on the Soul among the Latins, see C. BAZÁN, “The Human Soul: Form and Substance? Thomas Aquinas’ Critique of Eclectic Aristotelianism,” in: Archives d’histoire doctrinale et littéraire du Moyen Âge 64 (1997), pp. 95-126, and D. N. HASSE, Avicenna’s De anima in the Latin West. The Formation of a Peripatetic Philosophy of the Soul 1160-1300, London / Turin 2000.

312

CARLOS STEEL

views on the soul. In the third Part of his monumental philosophical encyclopaedia Speculum divinorum, Henry Bate points to an apparent contradiction (contrarietas) in what Avicenna says about the soul.2 For, as he notices, in the first part of his De anima, Avicenna subscribes to Aristotle’s definition: “humana anima est perfectio prima corporis naturalis instrumentalis.”3 In the second part, however, he writes that the human soul is not related to the body as its form and does not need an organ ready for its activities: “anima humana non habet se ad corpus ut forma (...) nec eget ut praeparetur sibi membrum.”4 Finally, in the fifth part Avicenna demonstrates that “the human soul is not impressed in a body” (“non est impressa in corpore”) and that it is not its form.5 Let us examine with Henry how we can explain this apparent contradiction. 1. The context Once Aristotle’s short chapter on the agent intellect became known to Latin scholars, it gave rise to endless controversies. For, as Bate nicely observes, “the words of the Philosopher on this topic contain so much ambiguity and he has taken in his way of speaking so much precaution that everyone, even those with contrary opinions, believes that his interpretation is concordant with the Philosopher’s words.”6 The third part of the Speculum contains a fascinating discussion on the nature of the agent intellect, which, in many ways, continues the lively debate on this topic at the university of Paris around 1270.7 2. See Speculum, II, 23. I quote the Speculum in the Leuven edition, started by E. VAN VYVER (Parts I-III) and continued by C. STEEL and G. GULDENTOPS. 3. AVICENNA, De anima, I, c. 5, p. 80, 12-16. All references to Avicenna are to the edition of S. VAN RIET in the series Avicenna Latinus. 4. AVICENNA, De anima, II, c. 1, p. 113, 44-47. 5. AVICENNA, De anima, V, c. 2, p. 101, 87-88. See also I, 1, p. 20, 31-33: “ipsa certe non est forma materiae nec est in materia: forma etenim quae est in materia est forma impressa in illa et existens per illam”; cf. 45-47 and p. 22, 64-71. 6. Speculum, III, 4, p. 169, 86-90: “In hac namque materia sermones Philosophi tantam ambiguitatem continent, atque tali quidem usus est cautela in modo loquendi quod unusquisque, etiam contrarie opinantium, intentionem suam verbis Philosophi consonam arbitretur.” 7. See on this debate, A. DE LIBERA, L’Unité de l’intellect. Commentaire du De unitate intellectus contra averroistas de Thomas d’Aquin, Paris 2004; C. BAZÁN, “Radical Aristotelianism in the Faculties of Arts. The case of Siger of Brabant,” in: L. HONNEFELDER – DE

A CONTRADICTION IN AVICENNA’S DOCTRINE OF THE SOUL

313

At that time Bate was a young student in Paris and he must have known personally the main antagonists in this debate, Siger of Brabant and Thomas Aquinas. When Bate started working on the Speculum, some 15 years later, he lived outside the university context in Mechelen,8 but had access to an impressive library offering ample material to continue the discussion on the intellect.9 With chapter III, 15 begins a long “dialectical investigation” (“inquisitio disputativa”) on how the intellect perfects a human being (as its “form” or as its “motor”?). This discussion goes on meandering with multiple digressions until the end of V, 16. Before presenting his own view Bate first discusses two radically opposing opinions. Some philosophers maintain with Aristotle that the intellect is united to a body as its “form” and that it is “the act of that body”: “Visum autem est quibusdam quod intellectus est actus corporis [...]” (pp. 208, 5-209, 30). Against this view many arguments are put forward: the most important comes in a long quotation taken from Siger of Brabant, here called “quidam vir famosus” (p. 211, 87). Other arguments are taken without acknowledgment from objections formulated by Thomas against the view that the intellect is form of the body (see Contra Gentiles, II, q. 56). For all these reasons, other philosophers, including especially Siger, concluded that the intellect cannot be united to the human body as its form, but only as its “mover” (motor), “as a sailor is in his ship” (p. 212, 26-30). Chapter 16 offers a refutation of this view with arguments mostly taken from Thomas Aquinas. R. WOOD – M. DREYER (eds.), Albertus Magnus und die Anfänge der Aristoteles-Rezeption im lateinischen Mittelalter von Richardus Rufus bis zu Franciscus de Mayronis, Münster 2005, pp. 585-629; G. GULDENTOPS, “Beyond Averroism and Thomism: Henry Bate on the Potential and the Agent Intellect,” in: Archives d’histoire doctrinale et littéraire du Moyen Âge 69 (2002), pp. 115-152. 8. The Speculum is dedicated to Guy d’Avesnes bishop of Utrecht, who was elected to that position in 1301. Bate must have started this huge enterprise much earlier, probably at his return to Mechelen from Orvieto in 1292. The last additions to the Speculum are dated 1310. On Bate and his encyclopaedia, see G. GULDENTOPS, “Henry Bate’s Encyclopaedism,” in: P. BINKLEY (ed.), Pre-Modern Encyclopaedic Texts, Leiden / New York / Köln 1997, pp. 227-237. 9. Besides Thomas’ De unitate intellectus Bate had a rare copy of Siger’s De anima intellectiva (there are only three manuscripts known of this text and Bate’s copy does not share their errors). Besides Themistius’ commentary also used by Thomas, Bate had a personal copy of Moerbeke’s translation of Philoponus’ De intellectu, not known to Thomas (only three manuscripts of this translation are preserved).

314

CARLOS STEEL

In the next chapter (17), Siger is given the opportunity to defend himself against Thomas’ objections. This defence, as Bate says, makes him formulate views that contradict his own position. In chapters 19-20 Bate develops his own position in the debate (“Dissolutio quaestionis propositae”) after having prepared his argument in a long digression in chapter 18 on the distinction and coincidence of formal and efficient causality. Having used Thomas against Siger, Henry launches a sharp attack against Thomas himself in chapters 21 and 22. From chapter 23 on, Bate discusses the different arguments that had been put forward for the first view, namely that the soul is the form of the body. After a long digression on the nature of matter, the discussion is concluded in V, 13. Then follows the examination of the arguments for the second view, that the soul is the mover of the body (V, 14-V, 16).10 Bate’s explanation how Avicenna’s apparent contradiction should be interpreted is situated in chapter III, 23, where he discusses the first argument for the first position. However, Avicenna’s view on formal causality came already under scrutiny in the previous chapters. 2. With Thomas against Siger of Brabant Having presented the two opposing views on the intellective soul, Bate first discusses the second opinion: “Some hold the view that the intellect is not united to the human body as its form, but only as its mover, as a sailor is united to his ship, having been inspired by what Plato said.”11 This is undoubtedly the position of the “famous” Siger of Brabant, since Bate quotes in what follows from his De anima 10. For an analysis of the argument in parts III-V see my introduction to the edition of Parts IV-V: On the Nature of Matter on the Intellect as Form of Man, Leuven 1993, pp. IX-XII. 11. Speculum, III, 16, pp. 212, 27-213, 30: III, 15, pp. 26-30: “Hinc ergo opinati sunt quidam quod intellectus nequaquam unitur humano corpori ut forma, sed ut motor tantum, quemadmodum nauta navi, occasionem etiam accipientes ex sermonibus Platonis.” The example of the sailor is found in ARISTOTLE, De an., II, 1, 413a 9, but is not attributed to Plato. As the editor notices, Bate clearly follows what Thomas asserts at the beginning of ScG, II, q. 57: “Plato igitur posuit, et eius sequaces, quod anima intellectualis non unitur sicut forma materiae, sed solum sicut motor mobili, dicens animam esse in corpore sicut nautam in navi.” On the attribution of this comparison to Plato, see the note of R.-A. Gauthier in the Leonine Edition of the Sententia De anima, II, 2, p. 76, 153.

A CONTRADICTION IN AVICENNA’S DOCTRINE OF THE SOUL

315

intellectiva.12 In this treatise Siger defended his view that the intellect is the perfection of its body as its “mover” (“motor”), as is clear from this passage: The Philosopher indicates in On the Soul II that the intellect, as it is separable from the body, is not an act of the body, or, if it is an act of the body, that it is its act just as the sailor is act of his ship, that is, that it is the perfection of the body, yet separated from the body in its being, though it is united with it in acting, as is the sailor with his ship.13

In chapter 16, Bate brings forward a number of arguments against Siger’s view that the intellective soul is the “mover” of the body (“At huius quidem opinionis absurditas faciliter comprobatur”). Remarkably, all arguments are taken from Thomas’ treatise De unitate intellectus (1270), though this work was not a refutation of De anima intellectiva – which Siger wrote around 1272, to reply to Thomas’ criticism –, but was an attack on Siger’s earlier views. As is mostly the case, Bate does not acknowledge his debt to Thomas in this chapter. These are the arguments: (1) “Nam ut scribitur 9° Metaphysicae...” (213, 36-41) = same quotation in Thomas III, 178-185: “Primo quidem per hoc quod dicit Philosophus in IX Methaphisice...” (2) “Item, quorum actus in alterum seu in motum transeunt,...” (213, 46-53) = Thomas III, 208-218: “Tertio, qui in his quorum actiones in alterum transeunt....” One may notice that Bate modified the example “Sortes intelligeret” into “Henricus intelligeret aut Guido” (3) An objection “At vero dixerit forsan aliquis quod motum ab intellectu...” (214, 55-59) = Thomas III, 219-224: “posset ergo aliquis sic dicere quod motum ab intellectu...” 12. On Siger of Brabant and his doctrine about the intellect, see J.-B. BRENET, “Siger de Brabant et la notion d’operans intrinsecum : un coup de maître?,” in: Revue des sciences philosophiques et théologiques 97 (2013), pp. 3-36 with extensive bibliography in n. 1. Unfortunately, the contribution of Bate in this debate is never acknowledged. 13. De anima intellectiva, III, p. 79, 38-42 (ed. C. BAZÁN): “Innuit Philosophus in secundo De anima quod intellectus, cum sit separabilis a corpore, non est actus corporis, aut si est actus corporis, quod est actus eius sicut nauta navis, hoc est, quod est perfectio corporis, in esse tamen suo a corpore separatus, licet in operando unitus, ut nauta navi.” This quotation is not found in the Responsio of Siger, but is an argument against the opposing view that the intellect is united to the body as its form. There is no doubt, however, that Siger shares this view. See further in his Responsio, p. 85, 85-89.

316

CARLOS STEEL

(4) Reply to the objection with a reference to De anima II 2 and Themistius (p. 109, 68-71): “Sed tunc procedendum est secundum doctrinam Aristotelis in 2° De anima ...” (214, 59-66) = Thomas, III, 224-234: “Huic autem dicto Aristotelis resistit in II De anima...quod exponens Themistius...” After the quotation from Themistius, Thomas explains the “intentio Philosophi” (III, 235-250): the same argument is found in the next section in Bate, 214, 68-215, 80. However, Bate first inserts a quote from Averroes (De anima, II, 24): “Et Commentator super eodem: Actio, inquit agentis est illud quod existit in recipiente, et est forma”, and then reformulates Thomas’ exposition, thus giving the impression that the argument that follows also comes from the Commentator. (5) “Demonstratione quidem igitur universali ex doctrina Philosophi hoc declaratur.” The demonstration that follows “ex actibus principia actuum cognoscimus...” (215, 81-95) is a patchwork of passages taken from De unitate intellectus, at the beginning of chapter III, 8-40, with additional quotations from De anima. (6) Only the concluding section (215, 96-216, 13) has no parallels in Thomas. It offers a long quotation from Averroes, De anima II, 37 where the Commentator concludes that the soul is “forma corporis.” Bate qualifies “anima” as “intellectiva”, thus forcing Averroes to say what he did not intend to say. Follows a reference to Ethics, I, 4, where it is said that just as the soul is in the body as its formal cause, so the intellect in the soul.14 In Chapter 17 (“Qualiter veritatis praemissae contrarium opinantes involvunt se ipsos et sibi ipsis contradicunt”), Siger is given an opportunity to defend himself against Thomas’ objections. As Bate observes, this defence brings him to formulate views that contradict his own position: In their attempt to escape somehow the dangerous but inevitable attacks of these demonstrations, those people who ascertain that the intellect is 14. Speculum, III, 16, p. 216, 5-8: “Hinc Philosophus in 1° Ethicorum: Non sicut omnino separatum ens, sicut motor solum existens, sed ut aliud in alio, puta sicut visus, inquit, in corpore sive toto quidem.” The editor could not identify this quotation, but see Eth. Nic. I, 4, 1096b27-29: “Set certe ei quod est ab uno esse, vel ad unum omnia contendere, vel magis secundum analogiam; sicut enim in corpore visus in anima intellectus et aliud utique in alio.”

A CONTRADICTION IN AVICENNA’S DOCTRINE OF THE SOUL

317

united to us only as a mover, affirm the following, as if they were ignorant of their own voice.15

What follows is a long quotation from Siger’s De anima intellectiva, in which Siger, in his attempt to escape Thomas’ criticism, reformulates his view on the intellect as “motor.” His argument goes as follows. Although the intellect is by its nature separate from the body, in the activity of thinking, it is “intrinsically operating with regard to the body”; however, operations of intrinsic operators may be attributed to what is composed of the intrinsic operator and that in which the latter operates intrinsically; therefore, these intrinsic movers may be called “forms and perfections” of the bodies with regard to which they operate.16 J.-B. Brenet hails the notion of a “motor intrinsecus” as “un coup de maître” which allows Siger to take a third way between materialism and strict dualism.17 The intellective soul is not a material form but is and remains in its essence a separate substance. However, in thinking it enters a relationship to the body, as it operates upon the phantasmata. The intellect does not operate from without, as a craftsman upon a material, but from within. As it is an intrinsic moving principle, it may be considered a part of the whole composite (this human being), and even its “form” (insofar as the moving principle 15. Speculum, III, 17, p. 216, 17-20: “Importunam itaque et inevitabilem harum demonstrationum invasionem aliqualiter evadere conantes, hi quidem nobis intellectum asserunt uniri ut motorem tantum, propriam vocem ignorantes dicunt.” 16. De an. int., c. 3, p. 85, 80-85: “Intellectus in intelligendo est operans intrinsecum ad corpus per suam naturam, operationes autem intrinsecorum operantium, sive sint motus, sive sint operationes sine motu, attribuuntur compositis ex intrinseco operanti et eo ad quod sic intrinsece operatur, immo etiam apud philosophos intrinseci motores, vel intrinsece ad aliqua operantes, formae et perfectiones eorum appelantur.” 17. See BRENET, “Siger de Brabant,” p. 9: “Siger s’évite à ses yeux le double écueil du matérialisme et du dualisme strict: l’intellect n’est pas une forme matérielle, non plus qu’une substance ontologiquement séparée qui n’entrerait dans aucune forme de composition naturelle avec le corps; dans son rapport intellectuel au corps, il est un opérant intrinsèque qui fait advenir, en dépit de la coupure dans l’être, une unité sui generis: l’homme, lequel se voit attribuer par transfert métonymique l’opération de cet acteur du dedans qu’est pour lui l’anima intellectiva. L’operans intrinsecum apparaît de ce point de vue comme le concept majeur d’une troisième voie, fidèle à Aristote, qui parvient à articuler l’idée nécessaire de la séparation in esse à celle de l’intellect comme ‘partie’ de l’âme, comme partie de l’homme, et même, conformément à la définition générale de l’âme donnée dans le De anima, fût-ce en un sens large et presque équivoque, à celle de l’intellect comme ‘forme’ et ‘perfection’.”

318

CARLOS STEEL

is also formal cause). That is the reason why “the philosophers” define the soul as “act of the body.” Against Thomas’ objection that one could not explain that “this human thinks” if the intellect is not connected to the body as its form, Siger answers that the whole can be denominated from one of its parts. If the intellect as “intrinsecum operans” is part of the composite (“this human being”), one may also say that “this human being thinks.” Take the example, Siger says, of sight: We say that a human being sees, though sight only occurs in the eye18 and not in other parts of that human, for instance in his foot; nor is it true to say that his foot sees; and unless the eye in which alone sight occurs, is united with the other parts, it would be impossible to attribute sight to this whole composed of eye and other parts. In this sense [one can say that] a human being thinks, though the act of thinking occurs only in the intellect, not in the body; hence neither the body thinks, though the body has sense perception; the human being however thinks owing to a part (of his being) just as the human being sees owing to a part. Yet the mode of union of the part that sees to the other parts in the whole is different from the mode of union of the part that thinks to the other parts in the whole that thinks. Nevertheless, this union is a sufficient condition to attribute to the whole what belongs to its part.19

Even if the intellect is only united “in operando” with the body, this union “secundum naturam ipsius” is sufficient to consider the intellect as a part of the whole human being, as it is “intrinsecus operans.” And if it is a part, this whole human being can be said to think. 18. Cf. ARISTOTLE, De an., I, 4, 408b 25-27. 19. De an. int., c. 3, pp. 85, 90-86, 1: “Unde considerandum quod hominem ipsum dicimus videre, cum tamen visio sit in solo oculo et non sit in aliis partibus hominis, ut in pede ; nec est verum dicere pedem videre ; et nisi oculus in quo solo est visio unionem haberet ad alias partes, non esset attribuere cuidam toti ex oculo et partibus aliis videre. Sic et homo intelligit, cum tamen intelligere sit in solo intellectu et non in corpore ; unde nec corpus intelligit quamquam corpus sentiat ; homo autem ipse intelligit secundum partem, sicut videt secundum partem. Modus tamen unionis partis videntis ad alias partes in toto vidente alius est quam modus unionis partis intelligentis ad alias partes in toto intelligente. Sufficiens tamen est unio ad hoc ut quod parti inest, per partem toti attribuatur.” See also c. 3, p. 87, 17-23: “dicendum quod homo est homo per intellectum, nec tamen propter hoc oportet alteram partem huius compositi uniri alteri parti ut figura cerae unitur, sed sufficit quod praedicto modo uniatur ut totum compositum ab eo denominetur. Et est attendendum quod, cum homo denominetur intelligens ab opere intellectus, et a substantia eius habet denominationem. Quod enim denominatur ab eius accidente, et ab eius substantia.”

A CONTRADICTION IN AVICENNA’S DOCTRINE OF THE SOUL

319

Bate is less convinced than Brenet that Siger displays here a “coup de maître.” On the contrary. As he says, “those people adduce an example that is absolutely unsuitable for what they intend (to say) and thus acknowledge their error.”20 For they say that the whole may be denominated from its part, as when it is said that “a human is healed, when his eye is healed.”21 Thus they conclude that the intellect, which is “motor” of the human being, may be united to the human being as a “part” from which it can be denominated. But, as Bate comments, a moving principle never becomes a part of what is moved. And to ridicule Siger’s view he gives the trivial example of a pen moved by a human person. Though this human person who is its “motor” is thinking, his writing pen is not. This counterexample, however, misses the point: for Siger insists that the intellect is an “intrinsecus movens”, and the person who uses the pen certainly is not. Finally, “coming to the end (of their investigation) forced by the truth” (“ad metam ducti a veritate coacti” 217, 43) these people too, Bate concludes, have to admit that the intellective soul, if it is an intrinsic mover, must be considered as “the perfection and form of the body”, as is clear from what Siger says: “dicendum quod anima intellectiva perfectio corporis est, secundum quod intrinsecum operans ad corpus perfectio et forma corporis habet dici” (p. 87, 33-34). In fact, as Bate explains, the argument about an intrinsic principle that is moving the whole makes no sense unless that principle is considered as a formal principle. Therefore, the intellect cannot only be regarded as the “mover” of the body, for the moving cause, insofar as it is moving, can never become a part of what is moved. And if it 20. Speculum, III, 17, p. 217, 27-28: “Insuper et ad erroris sui confessionem exemplum quoddam ad hoc adducunt, intentioni suae prorsus inconveniens.” 21. “nam homo sanatur cum sanatur oculus’ (p. 217, 29-30). Bate’s editor, E. Van de Vyver, could not find this example in Siger’s text and supposed that Bate may have read it in another now lost work of Siger: “Fortasse allegat Bate opus Sigeri perditum (de Intellectu?) vel hactenus ineditum.” I do not share this supposition. It is true that one does not find the example of the healing of the eye in Siger, but the more appropriate example of “seeing.” The example of the healing of the eye for denomination of the whole from the part comes from Aristotle and was well known by medieval scholars (see Phys., V, 1, 224a 25-26; Metaph., K, 11, 1067b 4: “sanatur corpus cum sanatur oculus’). This may explain why Bate formulated Siger’s example in this manner. However, the argument for which the example is used is found in Siger: it is possible to denominate the whole from its parts. Just as one can say that this human sees because his eye sees, one can say this human thinks because his intellect is thinking (see texts quoted in footnote 19).

320

CARLOS STEEL

were a part, it should be its formal principle since it can neither be an accident of it, nor the matter of what is moved. This is exactly what we call a formal principle, Bate says, namely an intrinsic moving principle.22 To explain how such an intrinsic formal principle can move the body, while keeping its transcendency, Bate refers to the celestial bodies, which are moved by immaterial immobile principles that are their forms, but never enter into composition with their bodies. As he explains later in the Speculum in the examination of the movement of the celestial spheres: “their motor is intrinsic in what is moved and yet separable and outside the extension of the body and thus similar to what the soul is and the separate forms”.23 Brenet rightly points to a similar cosmological background for Siger’s doctrine of the soul as “intrinsic motor.”24 Much further in the Speculum, in chapters V, 14-17, Bate discusses, one by one, the different arguments in favour of Siger’s position, which had been put forward in III, 15. “To be sure, those arguments which seem to prove that the intellect cannot at all be the form of a human being, but only its mover, must be refuted, for they are sophistic” (V, 14, p. 168, 7-9). In order to undermine Siger’s arguments Bate introduces his own distinction between material and immaterial forms. When the soul is understood as an immaterial form keeping its transcendency towards the body, Siger’s arguments against this explanation will be dissolved (see V, 14).25 22. See Speculum, III, 6, p. 178, 16-23: “Agens igitur intellectus aliquid nostri est non corporis; ergo aliquid animae, non motor tantum, ut infra apparebit, quia scilicet motor, secundum quod motor tantum, non est aliquid eius quod movet, aut si aliquid eius est, cum non sit accidens illius, quia praeter ipsum esse potest seu habet esse, neque materia similiter, ut demonstratum est supra, necessario erit principium formale; principium enim operationis intrinsecum formale principium vocamus.” See also XX, 29, p. 146, 43-48: “Motor autem quomodocumque dicatur appropriatus mobili, per se tamen, secundum quod motor simpliciter consideratus, inquam, non est aliqua pars essentiae seu quidditatis nec essentia tota nec aliquid proprie spectans ad eius quid est esse quod movetur ab ipso ; principium etenim intrinsecum rei motae non est, sed extrinsecum.” 23. See Speculum, XXI, 4, p. 199, 21-22. : “Declaratum etiam est quod motor hic intrinsecus est mobili suo necnon et cum hoc separabilis et extra magnitudinem ac per hoc animae similiter se habens et formae separatae seu abstractae (...).” See also further on p. xxx. 24. See BRENET, “Siger de Brabant”, pp. 29ff. 25. This distinction is not found in De anima intellectiva. See, however, Quaestiones in III De anima, q. 4, p. 13, 74-76; q. 9, p. 26, 8. More on this distinction further, on p. xxx.

A CONTRADICTION IN AVICENNA’S DOCTRINE OF THE SOUL

321

3. With Siger against Thomas Aquinas Chapter 21 is entitled “Dissolutio rationis cuiusdam super quam sustentatus est Thomas expositor.”26 In this chapter Bate criticizes Thomas’ view on the intellect as expounded in De unitate intellectus.27 Although the human soul is in its essence the act of the body, the intellect, which is an inherent part of the soul, transcends the body, as it has no need of bodily organs to come to actuality: “quod intellectus sit aliquid animae quae est actus corporis; ita tamen quod intellectus animae non habeat aliquod organum corporale.” (I, 465-467) Already during his life, Thomas’ position has come under sharp criticism, as it seems to come to an impossible conclusion.28 How could a power of a soul that is form and act of a body, transcend that body? How could a power achieve what goes beyond the essence of the soul to which it belongs? Siger formulated the objection as follows: It is impossible that some substance is united to matter and that the power of this substance is separated from matter. If the substance of the intellective soul had its being united to matter so that it gives being to matter, then its act of thinking, which belongs to its substance, would occur in some part of the body, as sight in the eye, what the Philosopher denies.29

Although Thomas says that this “appears most clearly and undoubtedly from the words of Aristotle”, he seems to have had doubts himself, as he adduces as confirmation of his conclusion examples taken from natural phenomena. “For we see in many cases that a form is the act of a body mixed of elements and yet has a power that is not the power of any of its elements, but surpasses them” (I, 472-479): thus, a magnet has the power to attract iron, a power that surpasses the capacities of the mixture of elements of which it is the form; 26. On Bate’s attitude towards Thomas Aquinas see G. GULDENTOPS, “‘Famosus expositor...’, on Bate’s (anti-)Thomism,” in: Recherches de théologie et philosophie médiévales 72/1 (2005), pp. 191-223, esp. pp. 200-201 on Bate’s rejection of Thomas’ doctrine of the soul as form of the human body. 27. I quote De unitate intellectus in the Leonine edition. For translated texts I used and adapted R. MCINERNY, Aquinas against the Averroists. On There Being Only One Intellect, West Lafayette (IN) 1993. 28. On the early reactions to Thomas, see DE LIBERA, L’Unité de l’intellect, pp. 22-23, 102ff. Bate’s contribution to this debate is not mentioned. 29. De an. int., c. 3, p. 82, 2-7.

322

CARLOS STEEL

similarly, the jasper has a power to coagulate blood. Bate adds himself the example of the scamonny that can reduce jaundice, an example he probably knows through Avicenna’s Canon (II, 2, 636). And many more examples could be adduced to show that plants and animals may have powers that surpass the powers of the elements of which they are composed. However, as Bate observes, “what said does not lead to a necessary conclusion nor does it dissolve the arguments that oppose this conclusion. For examples only may persuade, they never conclude with necessity.”30 And Bate knows how to undermine the persuasive force of an example by indicating dissimilarities between the example and what one intends to explain with it. As he argues, there is nothing special about a form being the act of a body and yet surpassing the nature of the body and its composing elements. In fact, this is a characteristic of all material forms, as Aristotle shows in Metaphysics, VII, 17, 1041b 11-31. But the case of the intellect, which in its activities may surpass all material conditions, is quite different.31 From an examination of its activity one must necessarily conclude that the intellect itself, from which proceeds this activity, cannot be a (material) form or act of this body. For everything only acts insofar as it is in act. Therefore, if the intellect is a power of the soul, which itself is an act of the body, its activity too will communicate with the body. What Thomas claims, that “a form of which the activity does not communicate with the body is itself the act of a body” is thus utterly impossible, as Siger already said in his response to Thomas. Having concluded that Thomas’ argument is worthless (“nihil valet”), Henry turns in the next chapter (22) to another formulation of Thomas’ view. Finally, “forced by truth (“a veritate coactus” – same phrase used for Siger above) the “famous interpreter” Thomas comes to a formulation of his views that is more acceptable to Bate. This new formulation is found in a section where Thomas replies to some objections (“rationes vero quas in contrarium adducunt non est difficile solvere” III, 364-365). Opponents had claimed that, if the 30. Speculum, III 21, p. 236, 64-65: “exempli enim facultas non est nisi persuasio, nihil de necessitate concludens”, cf. THEMISTIUS, De an., p. 64, 80-81: “quamvis persuasionum debilissimum sit exemplum.” 31. See DE LIBERA, L’Unité de l’intellect, p. 103, on a similar critique by the anonymous of Giele.

A CONTRADICTION IN AVICENNA’S DOCTRINE OF THE SOUL

323

intellect is an act of the body, it is a material form and thus not exempt of corporeality, which makes it incapable of having universal knowledge. Thomas answers that he does not admit “that the human soul is form of the body according to its intellective power.” And against the objection that a power of the soul cannot be “more immaterial” than its essence, his answer is “that this would be the case if the essence of the human soul were form of matter in such a way that it did not exist by its own being, but only in virtue of the being of the composite.” The human soul, however, has a mode of existence by itself. “Although matter communicates somehow with it, it never fully comprehends it, as it is greater in dignity than the capacity of matter. Therefore, nothing prevents such a form having some operation or power to which matter does not attain.”32 Bate quotes these answers of Thomas in extenso. Yet he remains unsatisfied as Thomas’ qualified position still is self-contradictory. That the intellective soul is not a “forma materialis” anticipates what Bate himself will say. However, Thomas continues to compromise, as he gives the soul characteristics of both the material form and the immaterial form. For the soul is said both to exist by itself without matter and to have its being as the act of a material substrate. In his rejection of Thomas’ qualified view Bate draws upon Siger’s objection against Thomas, that “the ratio essendi in virtue of which something has a being united with matter and the ratio essendi in virtue of which something is separate and subsists by itself are so opposed to one another that they could not belong to the same being.”33 As we have seen, Bate criticized Siger’s view with arguments deriving from Thomas, and now he criticizes Thomas with arguments taken from Siger. The result is a remarkable dialectical disputation. What Henry does not say, but certainly must have noticed, is that Thomas, as he does himself, refers to Avicenna’s authority on the soul. Thomas, however, does not quote Avicenna to show his inconsistency in his doctrine of the soul, but as an argument ex auctoritate for his own interpretation of Aristotle. After having expounded the 32. De unit. int., III, 398-401. 33. De an. int., c. 3, p. 80, 64-66: “Unde anima intellectiva non potest habere rationem per se subsistentis et cum hoc unum facere cum materia et corpore in essendo” quoted by Bate as the main argument against those scholars who consider the soul to be the “form” of the body in Speculum, III, 15, p. 211, 96-98.

324

CARLOS STEEL

views of the Greeks (Themistius, Theophrastus, Alexander), Thomas turns to the Arabs, and first to Avicenna. The first text he quotes comes from De anima, V, 2. Avicenna first distinguishes the practical and the theoretical intellect (the former needs a body for its activities, the latter not always nor completely) and then concludes that the human soul is not identical with any of those powers, “but is what has those powers” and is “a solitary substance, that is, something that exists by itself and has an aptitude for these activities.” This text is not quoted by Bate, though he should have liked it. For with this insistence on the soul being a “solitaria substantia” Avicenna clearly interprets Aristotle in a Platonic, more dualistic sense. Even if the soul is the origin of its powers and activities, it is not identical with them in its essence. If Bate may have liked it, one wonders why Thomas selected exactly this quotation in support of his quite different view on the soul. The two subsequent quotations added by Thomas are also found in Bate. First, it is said that the soul is the “perfectio prima corporis naturalis instrumentalis” insofar as it has an ability to perform acts. And yet Avicenna also declares “quod anima humana non se habet ad corpus ut forma.” Thomas hastens to qualify that this statement only holds “when the soul is considered in what is proper to it, its intellective power.”34 As said, it is remarkable that Thomas refers to Avicenna to corroborate his claims that the intellective soul is the act of the body, though having a power transcending the body. After all, we noticed that Avicenna hesitates to consider the soul as the form of the body. Even if Avicenna endorses the Aristotelian definition of the soul, he qualifies it by adding “secundum quod attribuitur ei actiones”, that is, the definition only holds insofar as the soul is related to the body, not as it is in itself. Whereas Thomas seems to smoothen the differences between Avicenna and Aristotle, Albert had a different approach. Unsurprisingly Henry quotes him at the end of his criticism of Thomas, and together with Albert his student Ulrich of Strasbourg. “Ulrich asserts in his book On the Supreme Good that the intellect cannot be 34. Bate may have been inspired in his selection of texts by reading De unitate intellectus. Yet, his Avicenna quotations do not depend on Thomas’ text, as he uses a manuscript similar to S, different from the Avicenna text Thomas used. Thus, he reads II, 1 p. 80, 14 electione] cum electione; 15 meditando]+ vel consultando with ST; p. 113, 45 sibi praeparetur] inv. with S.

A CONTRADICTION IN AVICENNA’S DOCTRINE OF THE SOUL

325

the act of a body and the venerable Albert too admits this in some places.” Bate’s editor refers in footnote to many passages where Albert makes such a claim, and more examples could be added.35 In recent years several studies have been published on Albert’s doctrine on the intellect and its relation to Avicenna. As Dag Hasse says, Albert “follows Avicenna in teaching that one must distinguish what the soul is in itself from what it is in relation to the body” and he shares “Avicenna’s contention that Aristotle’s analysis of the soul was focused on the function and not the essence of the soul.” This Avicennian background is most evident in Albert’s discussion of the definition of the soul in De homine.36 Following Avicenna Albert shows that Aristotle’s definition of the soul as the form of the living body does not define what the soul is in itself but only “secundum quod principium emanandi a se affectiones.”37 4. Digression on the causes Having distanced himself from both opponents in the debate, Bate has to develop his own views, avoiding the Scyla of materialism and Charybdis of dualism. Quite a challenge not only for the author, but also for the reader who tries to follow him on his meandering path! To prepare his own solution Bate enters a long digression in chapter 18 on the differences of the four modes of causality, how they are related and how they may coincide in reality. This digression will help him to better distinguish two types of formal causes, the material and the immaterial, and to understand in what sense the soul is the formal cause of the body.38 At the end of the previous chapter (17) Bate had argued against Siger that an intrinsic moving principle cannot just be the efficient cause but must also be considered as the formal cause of the body it 35. cf. De Caelo, I, tr. 3, c. 4, ed. Colon., V/1, p. 62, 47-51; Physica, VIII, tr. 2, ed. Colon., IV, p. 606, 78-80. 36. See De homine, ed. Colon., XXVII, II, p. 31, 45ff.. 37. See on Albert and his relation to Avicenna D. N. HASSE, “The Early Albertus Magnus and his Arabic Sources on the Theory of the Soul,” in: Vivarium 46 (2008), pp. 232-252. 38. “Capitulum 18: Digressio quaedam occasione praemissorum circa distinctas proprietates causarum et coincidentiam earum ac formalis causae duplicitatem.”

326

CARLOS STEEL

moves. He further observed with Averroes that, in the case of immaterial beings, the efficient, formal and final causes coincide. Thus, the separate forms of the celestial spheres are not only their forms, but also as forms the cause of their movement and their finality.39 Bate insists that, in this coincidence, the formal cause is the most fundamental and has priority over the efficient and final cause, though, in reality, they coincide. This is how Averroes explains it: Eternal beings do not have an agent cause but by analogy and from the four causes they only have the formal and final cause; and even if they have some sort of agent cause, it will be only insofar as it is a form for something and maintains it.40

Avicenna, however, as Bate observes, seems to have held a different view on the causes, since he gives primacy to the agent cause over the formal cause, considering the “agent”, not the “form” as the “cause of being.” Bate quotes from Metaphysics, VI, 1: By agent cause [we mean] the cause which bestows an existence that is other than itself. In other words, the essence of the agent is not, by primary intention, to be the receptacle of that which acquires from it its existence neither is it what gives form to it, but it is so that it has in itself the power of that being and not accidentally.41

In contrast to Aristotle, who considers the agent cause primarily as the source of change and movement, Avicenna understands agency as 39. Speculum, III, 17, p. 219, 85-93: “Unde in substantiis intellectualibus triplex genus causae coincidit…” The editor refers to AVERROES, Metaph. XII, 36, 318K and De Subst. Orbis, c. 2, 6F-G. 40. “Res enim eterne non habent agens nisi secundum similitudinem, neque habent ex quatuor causis nisi formalem et finalem; et si habuerint aliquid quasi agens, non erit nisi in quantum est forma illi et conservans ipsum.”(De Caelo, IV, 1, ed. CARMODY, p. 654, 16-19. 41. Metaph., VI, 1, p. 191, 10-12, ed. VAN RIET: “Agens vero est causa quae acquirit rei esse discretum a seipso, scilicet ut essentia agentis secundum primam intentionem non sit subiectum illius esse quod acquiritur ab eo nec informetur per illud, sed ita ut in seipso sit potentia illius esse non accidentaliter.” The formulation “nec informetur per illud” is ambiguous. I take “illud” to refer to the agent cause, whereas the subject of the verb “informetur” is the being that is caused by the agent. Avicenna explains that the agent cause neither becomes the material cause or the formal cause of that which it makes exist. For my translations from the Latin Avicenna, I make use of Michael Marmura’s translation from the Arabic (Avicenna. The Metaphysics of Healing, Provo 2005), adapting it where needed to the sense of the Latin translation. In this case I also used the Italian translation and annotation of A. Bertolacci.

A CONTRADICTION IN AVICENNA’S DOCTRINE OF THE SOUL

327

cause of being, the first cause or creator standing somehow as model of this creative agency.42 Albert shares a similar view on the priority of the agent cause over the formal, as appears from a long text Bate quotes from his De causis et processu universitatis (I, tr I, ch. 11).43 Notwithstanding these authorities, Bate argues that, properly speaking and according to Aristotle, only the formal cause can be regarded as the “cause of being” as “ratio ipsius quod quid erat esse”, whereas the efficient cause is, as Aristotle defines it, the principle that initiates movement (“unde motus principium”). He also rejects Avicenna’s view that the metaphysician uses another notion of agent causality than the natural philosopher. Avicenna believes that Aristotle, when arguing in a metaphysical investigation, considers the efficient cause not only “as the cause of motion”, as is the case in physics, but also “as the principle and giver of being (“principium essendi et datorem eius”) such as the creator of the world is.” For Bate, on the contrary, the efficient or agent cause is both in a metaphysical and in a physical context, primarily the cause of motion, as Aristotle defines it “unde motus principium.” If we consider God as the “principle of being”, it must be insofar as He is the ultimate formal cause of the universe. Of course, in the case of the Creator and of all immaterial principles there is a perfect coincidence between formal and agent causality, which is not the case in the natural world. “Therefore, such an agent or efficient cause is not only the principle of movement insofar as it is an agent or efficient cause, but also the cause of being insofar as it is formal cause.” Henry observes that Avicenna himself adheres to such a view when he distinguishes between “the cause that prepares” by causing accidental changes in a subject so that it may receive the form and “the cause that gives a being its perfection”44. What gives 42. See K. RICHARDSON, “Avicenna’s Conception of the Efficient Cause,” in: British Journal of the History of Philosophy 21 (2013), pp. 220-239, in particular pp. 223-225. 43. On Albert’s understanding of the efficient cause and his debt to Avicenna, see E. GILSON, “Notes pour l’histoire de la cause efficiente,” in: Archives d’histoire doctrinale et littéraire du Moyen-Âge 37 (1962), pp. 7-31; W. DUNPHY, “Albert and the five Causes,” in: Archives d’histoire doctrinale et littéraire du Moyen-Âge 41 (1966), pp. 7-21. See besides the chapter in De Causis, Metaph., V, c. 3, “a digression on the number and modes of causes”: “intentio efficientis est inducere esse et dare ipsum post non-esse” (ed. Colonienis, p. 211, 25-29, quoted by DUNPHY, p. 10). 44. Speculum, III, 18, pp. 221, 44-222, 61; cf. AVICENNA, Liber primus naturalium, c. 10, p. 87, 27-35, ed. VAN RIET.

328

CARLOS STEEL

perfection, is what bestows the specific form and this is not a cause of movement. Of course, nothing prevents us from considering some efficient cause as the cause of being of something “insofar as the formal cause coincides with the efficient” (p. 222, 61-69). Similarly, on the question of formal agency, Avicenna seems, as Bate notices, to make contradictory statements. For in some texts, he limits formal agency to what is found in natural, composite beings. Thus, he writes that “the form is the essence which, when held in matter, constitutes the species”45 and again that “the form when it is taken as one of the principles, is related to what is composed out of it and matter, because it is a part of that which it constitutes in act”46 and again “by formal cause we understand nothing else than the cause which is part of the being of the thing, through which the thing is what it is in actuality.”47 All this seems to indicate that Avicenna regarded as form in the proper sense only the form in relation to a material substrate. If this is the case, it is difficult to designate immaterial substances, such as souls, as forms. This is explicitly excluded in a passage in Metaphysics II, 1 quoted by Bate: Each substance is either body or other than body. If it is other than body, then it is either part of a body or it is not a part of a body but is something altogether separable from bodies. If it is part of a body, then either it is its formal or its material part. If it is separated and not a part of a body, then either it is somehow attached to bodies in terms of moving them – and this is called “soul” – or it is separate from material things in all respects and this is called intellect.48

Bate concludes: From these and similar texts it seems that it is Avicenna’s opinion that no intellect and no immaterial substance can be or can be called a form in the proper sense, as they are separate essences, but only causes of movement (“motores”), though in an improper sense these too may be called forms.49 45. “forma dictur essentia quae, quando fuerit habita in materia, constituet speciem.” (cf. Sufficientia, I, c. 10, ed. Venet., 1508, f. 19vaF). 46. AVICENNA, Liber primus naturalium, c. 10, p. 93, 49-51, ed. VAN RIET. 47. Metaph., VI, 1, p. 291, 10-12, ed. VAN RIET: “Dico igitur quod nos non intelligimus esse causam formalem nisi causam quae est pars essentiae rei per quam est res id quod est in effectu.” 48. Metaph., I, 1, p. 68, 76-83, ed. VAN RIET. 49. Speculum, III, 18, p. 224, 34-28.

A CONTRADICTION IN AVICENNA’S DOCTRINE OF THE SOUL

329

However, in other texts quoted, Avicenna has no problem in using the term “form” in a much larger sense. In Metaphysics, VI, 4, Avicenna enumerates different usages of the term “form.” The first meaning is that of a being in actuality that is ready to act: “in this sense, the separate substances are forms.”50 And Bate also refers to Metaph., IX, 4, where Avicenna distinguishes two sorts of “forms and perfections of bodies”: “either forms whose existence is in the materials of the bodies (…) or forms that subsist by themselves and because of the materials of bodies, as is the case of the souls.” Of the soul it is said that “it is appropriate to its body (“appropriatur corpori”) insofar as it has its activity because of that body and in it.”51 Avicenna also endorses the view that “the soul of each heavenly sphere is its perfection and its form and not a separate substance, for otherwise it would be an intellect, not a soul.”52 After all, one may ask why Bate launched this long digression on the causes in a discussion about the intellective soul. The reason is that he attempts to solve the unfruitful discussion about whether or not the soul is “form” of the body. In his view, there is no doubt that the soul is the formal cause of the living body, and not just its “motor” or efficient cause. However, in the case of immaterial beings, the formal, efficient and final causes coincide in reality, though the formal cause has primacy. The ambiguity in Avicenna regarding the definition of the souls as “form” can be overcome if one distinguishes between two radically different modes of being a form. This distinction is made in chapters 19-20 where Bate sets out his own solution of this quaestio vexata on the intellective soul. 5. Bate’s own solution Bate starts chapter 19 with a general explanation of the endless confusion in this discussion: Concerning the aforementioned question, one should pay attention to the fact that opinions that seem to be contradictory and arguments that argue to both sides [of a question] may contain some truth and that 50. Metaph., VI, 4, pp. 324, 13-325, 38, ed. VAN RIET. 51. Metaph., IX, 4, p. 485, 25-32, ed. VAN RIET. 52. Metaph., IX, 4, p. 484, 11-17, ed. VAN RIET.

330

CARLOS STEEL

there was in them no other reason for error than an ignorance or negligence to distinguish some equivocation of the terms “act” and “form.” Although regarding generic terms equivocations often remain unnoticed,53 one should not neglect to notice the difference between the separable and the inseparable form: I mean the material form and the immaterial or abstract form.54

In what follows, and also later at the end of Part V, Bate repeatedly explains the fundamental difference between those two modes of formal causality. The material form is produced from the potency of matter (“educta de potentia materiae”) through the material causes working upon matter and it has its being in matter as its act. When the composite being ceases to exist, the material form too does no longer exist. The immaterial form, by contrast, does not come forth from matter, but “from outside”, it does not have its being in a subject but exists by itself. It never enters into a composition with the body but holds it and perfects it in a “separate manner.” When the body it is related to ceases to exist, the immaterial form remains what it always was. As Aristotle says of the intellective soul, “having been separated, this alone is just what it is” (III 5, 430a 22). Henry also refers to what Aristotle says in Metaph., XII, 3, 1070a 13-14 that “in some cases” the forms continue to exist even when their subject is corrupted. Aristotle says “in some cases”, not in all cases, to leave the possibility open that there are separate forms, such as the intellective soul. Aristotle, Bate says, calls this immaterial and separable form “quod quid erat esse” and “ratio” and “exemplum sive species” and “exemplar seu causa exemplaris et paradigma”.55 Speaking in a Platonic manner, one could say that the body “participates” in the being of the separable form: “illud autem esse Platonici vocant esse per communicationem seu participationem” (235, 32-33). It is clear that Bate reads a Platonized Aristotle, assuming that the forms or 53. “Licet circa genera plerumque lateant aequivocationes”: see ARISTOTLE, An. post., II, 13, 97b 30-31; Phys., VII, 4, 249a 22-25. Bate often refers to this “dictum”, see V, 12, p. 165, 15; V, 15, p. 175, 78; VI, 8, p. 27, 99. 54. Speculum, III, 19, p. 228, 27-34. 55. Speculum, III, 18, p. 227, 16-20: “alius autem est [sc. modus] immaterialis formae, quam Philosophus non solum quod quid erat esse describit, ac rationem esse dicit, sed exemplum sive speciem et exemplar seu causam exemplarem et paradigma propter immaterialitatis et separabilitatis rationem et essentiam.” On Bate’s explanation of the difference between material and immaterial forms, see GULDENTOPS, “‘Famosus expositor...’,” pp. 201ff.

A CONTRADICTION IN AVICENNA’S DOCTRINE OF THE SOUL

331

causes of beings are primarily and ultimately the separate paradigmatic causes. To be sure, the material form too is a cause of being, but in an inferior mode and depending upon the causality of the separate form: “amborum modorum formae sunt essendi causae, licet non penitus eadem ratione univoca, sed [...] analogice dicta” (227, 20-23). Whereas the material form has no being separate from matter and is nothing but the act of matter, the immaterial form communicates its being to matter in such a way that it remains self-subsistent and separable and never becomes the act of a substrate-body.56 One might object that such a form remains external and separate from the body it is supposed to inform. Understanding the soul in this way thus seems to lead to an unfortunate dualism. But Bate replies that “in reality and in another sense” such a form “is more intrinsically and more intimately” (“magis intrinsecus et intimius” [227, 6]) connected to its body than a material form. In fact, what Aristotle said about the soul and its relation to the body, that they are not really two different things, but one and the same reality numerically one (cf. II, 1, 412b 6-9), only makes sense when the form is seen as an immaterial self-subsistent principle and its subject has no other being than that communicated to it by its from (233, 76-79). This is the case of the intellective soul: though entering “from without”, its causality is “naturally prior” and more comprehensive than that of the material form. Hence, this form is no less the cause of being and of being one for the body than the material act; on the contrary, it is more perfectly, more intimately and more indivisibly cause of being and unity.57 56. Speculum, V, 16, p. 178, 70-72: “Immaterialis vero forma suum esse taliter communicat materiae quod per illam non sustentatur, sed per se quidem subsistens est, et quantum ad hoc a materia separabile, ut iam dictum est saepe, ab aliis autem dicitur hoc esse non immersum materiae, sicut esse materialis formae immersum dicitur materiae.” 57. Speculum, V, 5, p. 135, 78-84: “Ob hoc autem non imperfectius est corpori causa essendi ens et unum quam materialis actus, immo perfectius, intimius, magis intrinsecus et indivisibilius inexistit ei entitatis et unitatis causa, ut visum est, praesertim cum secundum naturalem entium ordinem materiali actu hominis in genere causarum formalium sit intellectus prior, naturaliter principalior et magis universalis ac per se causa.” VI, 22, p. 83, 45-47: “Intellectus autem, licet de foris adveniat, non minus tamen sed magis unitur et intimius humano corpori quam anima sensitiva corpori animalis aut alia materialis forma suae materiae ex cuius educitur potentia.” See also IV, 43, pp. 114-115, 26-55; VII, 17; XI, 7 and XIV, 5.

332

CARLOS STEEL

In Part V, chapters 4-5, Bate explains at length how we have to understand this difficult thesis. The human body has a material form produced from the potencies of matter under the influence of diverse natural agents, including the celestial bodies. The material form acts as an efficient cause transforming matter to constitute a being similar to it (as a human generates a human), whereas the immaterial form – the intellective soul – is the principal cause of its being, as it is the formal cause of its being and maintains it in being. Once united to the intellective form, the material form has no being on its own, but “participates” in the being of the immaterial and incorporeal form.58 This superior form is thus the ultimate cause of being and unity of the human individual, though it becomes somehow “incorporated and individualised” in this particular human being as “Guido” or “Socrates”, because of the influence of subordinated material forms.59 Whereas the material form that makes this a particular human being has no being besides this particular, the intellective form, which subsumes the material act and grants it its being, can exist separately from this material actualisation. And yet this transcendent form can be said to be “proper” (propria) to each of us, as Averroes says about the agent intellect in De anima, III, 36: “because that through which something performs its proper activity is its form, and we perform our proper activity through the agent intellect, the agent intellect must be form in us.”60 Like Siger, Bate finds a confirmation of his understanding of the intellective soul as the transcendent form of the body in what Averroes says of the forms of the celestial bodies (as explained in De substantia orbis, 2). Although these forms are immaterial and remain separate, they are intimately united to the spheres they move.61

58. Speculum, III, 20, p. 233, 76-78: “immaterialis forma et subiectum eius, licet ens actu sit, non tamen sunt duo entia actu, sed unum numero, cum subiectum non sit ens per se sed per illam.” 59. See Speculum, IV, 5, p. 134, 73ff.: “Proportionaliter quidem quoque modo, ut esse hominis in Sorte aut Guidone (…).” 60. See Speculum, III, 6, p. 178, 23-27 with reference to AVERROES, In De anima, III, 36, pp. 499, 587-500, 590. Bate quotes the full text of Averroes in Speculum, XVI, 2, p. 246. 61. Bate repeatedly points to this analogy. He notices, however, a difference. The material substrate of bodies in the sublunary realm has some mode of being, before it is connected to the immaterial form: the immaterial form subsumes this being of the material

A CONTRADICTION IN AVICENNA’S DOCTRINE OF THE SOUL

333

6. Conclusion It is not always easy to determine what exactly Bate’s own position is between the numerous quotations, digressions, repetitions in redundant periods. He disapproves of Siger but is also sharp (and often unfair) in his criticism of Thomas. When he has to express his own view he keeps hiding behind a mass of authorities he quotes. What is most remarkable in his argument is his insistence on the priority of the formal cause of being. The intellective soul has a role similar to the paradigmatic forms, which are the causes of all beings by participation. Bate’s Platonic understanding of the agent intellect is most clearly explained in the introductory chapter of Part III: Through the agent intellect in the act of thinking the same happens according to the Philosopher as what happens through the ideas of the Platonists, as if the intellect were itself the separate ideas.62

Further on, at the end of Part V, Bate concludes that one may also consider the divine intellect as the primary formal cause of all intellectual activities.63 The divine form comprehends in itself in a transcendent way all immaterial forms, and as the formal cause of the first heaven, it is also the first cause of being and thus creator of the whole universe.64 It has taken a long way to follow Bate on his own quest of truth. What, then, about Avicenna and his views on the soul? Bate returns to that question in chapter 23, where he begins discussing the arguments in favour of the soul as form of an instrumental body. He shows that there is no problem in accepting Aristotle’s definition form. In the case of the celestial spheres, the substrate body would be nothing when its form is separated from it (see Speculum, III, 20, p. 233, 83-92). 62. Speculum, III, 1, p. 157, 63-66: “accidit per intellectum agentem in actione intelligendi illud idem fieri secundum Philosophum quod per ideas Platonicorum, ac si intellectus esset ipse quidem ideae separatae.” Bate seems to be inspired by Averroes’ dictum at the end of In De anima, III, 18 (p. 440, 96-98): “Et omnia dicta ab Aristotele in hoc sunt ita quod universalia nullum habent esse extra materiam, quod intendit Plato. Quoniam si ita esset, non indigeret ponere intellectum agentem”. Yet, he does not abandon the Platonic separate forms but introduces them in the agent intellect. 63. See Speculum, V, 18 “Qualiter […] secundum Aristotelicam philosophiam simul et Platonicam intellectus non solum ille qui proprius sed et conditor seu universalis est causa formalis homini licet remota.” 64. On God as the first formal cause of the universe, see Speculum, XIV, 5, and GULDENTOPS, “Beyond Averroism,” pp. 144-150 and ID., “‘Famosus expositor...’,” pp. 218-219.

334

CARLOS STEEL

of the soul as form of the body if one does keep in mind that “form of the body” is an equivocal phrase: the intellective soul is form of the body in quite a different sense than the vegetative and sensible soul. Only if one pays attention to the different senses of being “form”, one may explain the different positions of Avicenna which at first seem contradictory.65 Carlos STEEL De Wulf Mansion Centrum, Hoger Instituut Wijsbegeerte – KU Leuven Kardinaal Mercierplein 2 B-3000 Leuven [email protected]

65. “quae quidem contrarietas aliter quam secundum praedictam distinctionem concordari non posset” (p. 241, 92-93).

NATURA INTENDIT SPECIEM: A BRIEF RECEPTION HISTORY OF AN AVICENNIAN TOPOS FROM PETER OF SPAIN TO FRANCESCO PICCOLOMINI Guy GULDENTOPS La nature ne se soucie pas de l’individu, mais de l’espèce [...].1 Abstract The Avicennian dictum “nature intends the species” has been interpreted in different ways. This paper analyzes the most interesting late-medieval and Renaissance texts in which the dictum is cited. While major 13th-century thinkers (such as Peter of Spain, Albert the Great, and Thomas Aquinas) use it in defense of a species-centered view on reality, Duns Scotus and John Baconthorpe propound a more individualist interpretation (focusing on the vague individual); and while the dictum is often cited in support of a realist theory of universals, it is also adduced by nominalists. In the Renaissance, Alessandro Achillini and Francesco Piccolomini argue for an individualist understanding of Avicenna’s view on the intention of nature, but their interpretation does not supersede the classical idea that the species is prior and superior to the individual (as is evidenced by Pomponazzi, Javelli, Galileo, and the Coimbra Jesuits).

1. The Venerable Bede on Avicenna In a fine essay on “The Reception of Avicenna’s Physics in the Latin Middle Ages,” Jules Janssens points out that Adam of Bocfeld and Roger Bacon paraphrase a famous passage from book I of the Sufficientia, where the Persian Peripatetic argues that nature aims at producing 1. A. THIBAUDET, Trente ans de vie française. III. Le Bergsonisme, Tome 2, Paris 1923, p. 135. Albert Thibaudet’s sentence is a rather banal reformulation of one of Henri Bergson’s fundamental intuitions; see his L’évolution créatrice, Paris 1908, p. 55: “[...] chaque espèce, chaque individu même ne retient de l’impulsion globale de la vie qu’un certain élan, et tend à utiliser cette énergie dans son intérêt propre [...]. L’espèce et l’individu ne pensent ainsi qu’à eux, – d’où un conflit possible avec les autres formes de la vie.”

336

GUY GULDENTOPS

non-concrete individuals.2 In a parallel passage from book VI of his Metaphysics, Avicenna articulates his view more clearly. Sublunary individuals are perishable, but since they reproduce themselves, there is a persisting succession of individuals, and this renders the species perpetual. Hence it is obvious that the goal of nature is only to preserve what is termed ‘the vague individual’ (i.e., an indeterminate x instantiating the species X). As Deborah Black has stressed, this downplaying of the concrete individual in the physical realm is closely linked up with Avicenna’s view of divine foreknowledge: “Just as God’s providence is perfected so long as he knows particulars ‘in a universal way’, so is the natural order completed so long as the species are exemplified in merely vague individuals.”3

2. See J. JANSSENS, “The Reception of Avicenna’s Physics in the Latin Middle Ages,” in: A. VROLIJK – J. P. HOGENDIJK (eds.), O ye Gentlemen. Arabic Studies on Science and Literary Culture. In Honour of Remke Kruk, Leiden / Boston 2007, pp. 55-64, esp. 59-60. Cf. AVICENNA, Liber primus naturalium, I, 1, ed. S. VAN RIET, Louvain-la-Neuve / Leiden 1992, pp. 8, 53 - 9, 71 (= The Physics of The Healing [...] Books I & II. A parallel EnglishArabic text translated [...] by J. MCGINNIS, Provo 2009, p. 7). On Adam’s use of Avicenna’s ideas about nature, see also S. DONATI, “Physica I, 1: L’interpretazione dei commentatori inglesi della Translatio vetus e la loro recezione del commento di Averroè,” in: Medioevo 21 (1995), pp. 75-255, esp. 202-206 and 221; on Bacon’s interpretation, cf. infra, pp. 345-346. 3. AVICENNA, Liber primus naturalium, I, 1 (cf. n. 2), and Liber de philosophia prima sive scientia divina, VI, 5, ed. S. VAN RIET, Louvain / Leiden 1980, pp. 334, 31 - 335, 52 (= The Metaphysics of The Healing. A parallel English-Arabic text translated [...] by M. E. MARMURA, Provo 2005, pp. 226-227). On these passages, see O. LIZZINI, “Wuğūd-Mawğūd/Existence-Existent in Avicenna. A key ontological notion of Arabic philosophy,” in: Quaestio 3 (2003), pp. 111-138, esp. 134-137; D. L. BLACK, “Avicenna’s ‘Vague Individual’ and Its Impact on Medieval Latin Philosophy,” in: R. WISNOVSKY – F. WILLIS – J. C. FUMO – C. FRAENKEL (eds.), Vehicles of Transmission, Translation, and Transformation in Medieval Textual Culture, Turnhout 2011, pp. 259-292, esp. 260262 (quotation from p. 262); J. JANSSENS, “What about Providence in the Best of All Possible Worlds? Avicenna and Leibniz,” in: P. D’HOINE – G. VAN RIEL (eds.), Fate, Providence and Moral Responsibility in Ancient, Medieval and Early Modern Thought, Leuven 2014, pp. 441-454, esp. 449-450. On the Aristotelian background of Avicenna’s view, see P. ADAMSON, “On Knowledge of Particulars,” in: Proceedings of the Aristotelian Society 105 (2005), pp. 273-294; on the Alexandrian background, see M. RASHED, Essentialisme. Alexandre d’Aphrodise entre logique, physique et cosmologie, Berlin / New York 2007, pp. 252-254; on Philoponus’ influence, see P. LETTINCK, Aristotle’s Physics and Its Reception in the Arabic World, With an Edition of the Upublished Parts of Ibn Bājja’s Commentary on the Physics, Leiden 1994, pp. 96-97, and, more skeptically, A. LAMMER, The Elements of Avicenna’s Physics. Greek Sources and Arabic Innovations, Berlin / Boston 2018, pp. 54-55 and 62-64.

NATURA INTENDIT SPECIEM

337

In Latin scholasticism, Avicenna’s view on the intention of nature4 was always understood in the light of a creationist re-interpretation of Aristotle’s teleology (which was encapsulated in the sentence “Deus et natura nihil faciunt frustra”).5 Moreover, Avicenna’s complex view was epitomized in a compact adage that could easily be cited in various contexts: “natura intendit speciem.” Its polysemic meaning is brought out in the Axiomata philosophica, a scholastic florilegium published under the name of the Venerable Bede. Among the propositions on natura, the compiler of this pseudepigraph cites the dictum “nature only intends the species, not the individual” and attributes it to Avicenna, though it does not occur in the Avicenna Latinus in this precise form. The compiler also adds two short explanations: the dictum either means that “common nature [i.e., universal nature, which influences all natural things], like the heavenly world, only aims at perpetuating the species, not at perpetuating the individual”; or, as others believe, Avicenna makes this claim “because he assumes real universals in a Platonic manner.”6 4. On his ideas about finality in nature, see K. RICHARDSON, “Avicenna on Teleology: Final Causation and Goodness,” in: J. K. MCDONOUGH (ed.), Teleology. A History, Oxford 2020, pp. 71-89, esp. 85 (with n. 35), where she argues that according to Avicenna “natural active principles are directed toward ends that promote the survival of the bearers of those principles.” On her reading, his teleology subordinates the individual to the species: “From this point of view, the survival of a plant or an animal is good. And the survival of the species is even better, since species persist through eternity.” 5. See e.g. Auctoritates Aristotelis, 3, 18, ed. J. HAMESSE, Louvain / Paris 1974, p. 161, 16 (based on ARISTOTLE, De caelo, I, 4, 271a33). On this principle, see J. G. LENNOX, Aristotle’s Philosophy of Biology. Studies in the Origins of Life Science, Cambridge 2001, pp. 205-220; M. LEUNISSEN, Explanation and Teleology in Aristotle’s Science of Nature, Cambridge 2010, esp. pp. 64-65, 129-130, 169-172, and 211-217. On teleology in medieval philosophy of nature, see e.g. C. L. BARNES, “Natural Final Causality and Providence in Aquinas,” in: New Blackfriars 95 (2014), pp. 349-361; R. PASNAU, “Teleology in the Later Middle Ages,” in: MCDONOUGH (ed.), Teleology. A History, pp. 90-115, esp. 91-100. 6. Axiomata Philosophica Venerabilis Bedæ [...], Ex Aristotele et alijs præstantibus Philosophis diligenter collecta [...], Coloniæ, Sumptibus Bernardi Gualtheri, 1623, pp. 237238 (= PL 90, Paris 1862, col. 1018C-D): “Natura solùm intendit speciem et non indiuiduum. Auicenna [...] natura non indiuidualis, sed communis, sicut cœlum solum intendit speciem semper continuare, et non indiuiduum. Alio modo alijs dicendum videtur, quod hoc dicat Auicenna ponendo vniuersalia realia more Platonico. Et ideò de hoc eius dicto non curandum.” On this florilegium, see J. HAMESSE, “Du manuscrit à l’imprimé : l’évolution d’un florilège philosophique du XIIIe au XVIIe siècle,” in: S. CAROTI – R. IMBACH – Z. KALUZA – G. STABILE – L. STURLESE (eds.), « Ad Ingenii Acuitionem ». Studies in Honour of Alfonso Maierù, Louvain-la-Neuve 2006, pp. 127-145, esp. 137-141; EAD., “Les instruments de travail philosophiques et théologiques, témoins de l’enseignement et

338

GUY GULDENTOPS

The aim of this paper is to document how the Avicennian topos was used in different philosophical and theological contexts by various authors from the thirteenth to the sixteenth century. As will hopefully become clear, the explanations outlined in the Axiomata philosophica were not entirely uncontroversial. First, I will present the topos as it was construed by three major thirteenth-century thinkers: Peter of Spain, Albert the Great, and Thomas Aquinas. In a second section, its function in Roger Bacon’s Communia naturalium, in the Correctorium literature, and in Duns Scotus’s criticism of Aquinas will be described. The third section will be devoted to John Baconthorpe’s prolix discussion of the priority of the individual and to his defense of Avicenna as “the best interpreter of Aristotle on this issue.” In the fourth section, we will explore how the dictum was used in the quarrel between the so-called realists and nominalists (from Walter Burley and John Buridan to Hugolinus of Orvieto and Petrus Nigri). The fifth section will highlight some Renaissance philosophers who continued the medieval discussions of the Avicennian topos. In the last section, I will concentrate on Francesco Piccolomini’s account of the centuries-old debate on “what is primarily known according to nature,” a debate in which he assigned a pivotal role to Avicenna as a mediator between Alexander of Aphrodisias and Duns Scotus. At the end of this story, we will have to revisit the thesis that Latin scholasticism heralded the supremacy of the individual over the species. 2. A key assumption of thirteenth-century philosophy and theology: “Nature intends the species” In his Questions on Aristotle’s Zoology, Peter of Spain briefly alludes to the dictum. At the beginning of his commentary on book II of De partibus animalium, he discusses several questions that are connected with Aristotle’s theory of causes. In his answer to the question “how does the goal coincide with the form?,” he claims that a universal de l’influence des ordres mendiants à l’époque de la papauté d’Avignon,” in: K. EMERY, Jr. – W. J. COURTENAY – S. M. METZGER (eds.), Philosophy and Theology in the Studia of the Religious Orders and at Papal and Royal Courts, Turnhout 2012, pp. 601-628, esp. 612; and my “Nemo/Nihil dat quod non habet : Fortune d’un topos de Platon à Derrida,” in: J. HAMESSE – J. MEIRINHOS (eds.), Les Auctoritates Aristotelis, leur utilisation et leur influence chez les auteurs médiévaux. État de la question 40 ans après la publication, Barcelona / Madrid 2015, pp. 269-315, esp. 287 (with n. 54).

NATURA INTENDIT SPECIEM

339

agent cause intends something that is different from what it produces: “Universal nature intends the species and produces an individual, and in this manner the form and the goal do not coincide, since the goal is related to the species, whereas the production is related to the individual.” In the case of particular agent causes, however, the formal cause coincides with the final cause, since a particular agent cause intends and produces an individual (Socrates, for instance, generates the particular being he intends to generate, namely his particular son).7 A somewhat similar view is found in Albert the Great’s Questions on Aristotle’s Zoology. There, Albert claims that “universal nature intends to conserve the entire universe and its parts, and because species, not individuals, are the [constitutive] parts of the universe, universal nature principally intends to conserve the species.” Particular nature, however, primarily intends to produce a being similar to the procreating individual.8 Elsewhere in the same work, Albert observes 7. PETER OF SPAIN, Questiones super libro ‘De Animalibus’ Aristotelis, XII, 5, ed. F. NAVARRO SÁNCHEZ, Farnham 2015, pp. 287, 185 - 288, 192 (cf. ARISTOTLE, De part. an., II, 1, 639b10-13): “[...] natura uniuersalis intendit speciem et operatur indiuiduum, et hac uia forma et finis non coincidunt. Nam finis est respectu speciei, operatio autem respectu indiuidui. Sed agens particulare sicut homo intendit generare hominem, et sicut est agens particulare sic intendit [speciem] et operatur indiuiduum et non speciem, unde intendit generare hunc hominem scilicet filium suum [...]. Et loquendo de tali agente coincidunt forma et finis, quia idem est quod generatur et quod intenditur.” I put “speciem” between square brackets because it makes no sense; instead of deleting it, one might also add “” between “intendit” and “speciem”. On Peter’s questions (which apparently influenced Albert the Great), see also J. F. MEIRINHOS, “Petrus Hispanus Portugalensis? Elementos para uma diferenciação de autores,” in: Revista Española de Filosofía Medieval 3 (1996), pp. 51-76, esp. 70-71; M. DE ASÚA, “Medicine and Philosophy in Peter of Spain’s Commentary on De animalibus,” in: C. STEEL – G. GULDENTOPS – P. BEULLENS (eds.), Aristotle’s Animals in the Middle Ages and Renaissance, Leuven 1999, pp. 189-211; F. NAVARRO SÁNCHEZ, “La obra médica de Avicena en las Questiones super libro De animalibus de Petrus Hispanus,” in: Enrahonar. Supplement Issue 2018, pp. 387398. On the distinction between universal and particular nature, see T. BORSCHE and B. HOPPE, “Natura universalis/particularis,” in: Historisches Wörterbuch der Philosophie, Bd. 6, Darmstadt / Basel 1984, cols. 509-518, esp. 511-514. 8. ALBERT, Quaestiones super De animalibus quas reportavit fr. Conradus de Austria, XV, 2, resp., ed. E. FILTHAUT, Münster i. W. 1955, pp. 260, 69 - 261, 6. In this question, Albert discusses “whether the generation of the female is intended by nature.” On this question, see Th. W. KÖHLER, Homo animal nobilissimum. Konturen des spezifisch Menschlichen in der naturphilosophischen Aristoteleskommentierung des dreizehnten Jahrhunderts, Teilbd. 1, Leiden 2008, pp. 493-494; I. M. RESNICK, “Albert the Great on Nature and the Production of Hermaphrodites. Theoretical and Practical Considerations,” in: Traditio 74 (2019), pp. 307-334, esp. 318-319 (for similar views, see e.g. BONAVENTURE, Commentaria in quatuor libros Sententiarum, IV, 44, dub. 2, Ad Claras Aquas, Typographia

340

GUY GULDENTOPS

that “nature orders nourishing toward the preservation of the individual, while directing sexual activity toward the preservation of the species.” Eating, drinking, and sex are accompanied by “the sweetest pleasures,” but “just as nature intends the preservation of the species more than that of the individual, it has also instilled a more intense pleasure into sexual activity than into the activity of the nutritive faculty.”9 In his Physics commentary, Albert elaborates his view on the intention of nature in greater detail. An individual comprises two aspects: on the one hand, the perfection of being according to its specific nature, and on the other the supposit that underlies this nature. What nature intends per se when it produces things is the perfect existence (esse perfectum) of the species, and that is why Aristotle holds that “nature intends divine being.”10 According to Albert, this idea enables us to correctly interpret a dictum he attributes to Averroes: “the work of nature is the species.” If nature intended the individual, its product would be destroyed, and if it intended a universal class, such a generic being would be sufficient. Both of these consequents are false, and the conditional subclauses are counterfactual. Nature, then, intends the perfect and perpetual existence that inheres in a species. This, Collegii S. Bonaventurae, p. 918, and THOMAS AQUINAS, Summa theologiae, I, 99, 2, ad 1; cf. infra, n. 36). Albert does not mean that “species are parts of the universal” (pace Resnick), but rather that the universe is constituted of species; this does not prevent the universe from encompassing individual things as parts of the whole (cf. ALBERT, Physica, III, 2, 14, ed. P. HOSSFELD, Münster i. W. 1987, p. 194, 13-15). For the idea of the similarity between what generates and what is generated, cf. Ph. W. ROSEMANN, Omne agens agit sibi simile. A “Repetition” of Scholastic Metaphysics, Leuven 1996, esp. pp. 196-201. On Albert’s attitude toward Avicenna, see A. BERTOLACCI, “‘Averroes ubique Avicennam persequitur’: Albert the Great’s Approach to the Physics of the Šifā’ in the Light of Averroes’ Criticisms,” in: D. N. HASSE – A. BERTOLACCI (eds.), The Arabic, Hebrew and Latin Reception of Avicenna’s Physics and Cosmology, Boston / Berlin 2018, pp. 397-431. 9. ALBERT, Quaestiones super De animalibus, V, 3, resp., p. 155, 7-14. This passage is also quoted by G. TOEPFER, “Arterhaltung,” in: ID., Historisches Wörterbuch der Biologie. Geschichte und Theorie der biologischen Grundbegriffe, Bd. 1: Analogie – Ganzheit, Stuttgart / Weimar 2011, pp. 132-140, at p. 133 (Toepfer does not refer to Avicenna in this context). For the commonplace idea that sexual activity culminates in the most energetic corporeal pleasure, cf. e.g. THOMAS AQUINAS, Sententia libri Ethicorum, VII, 11, ed. Leon. 47, Romae 1969, p. 425, 98-101 (based on Eth. Nic., VII, 12, 1152b17-18); Summa theologiae, I-II, 34, 1; II-II, 53, 6, resp.; 152, 1, resp.; 153, 4, resp.; 180, 2, resp. 10. ALBERT, Physica, I, 1, 6, p. 12, 67-83 (this chapter is also mentioned by BLACK, “Avicenna’s ‘Vague Individual’,” p. 262, n. 7). Albert alludes here to De gen. et corr., II, 10, 336b27-32 and De an., II, 4, 415a26-b1.

NATURA INTENDIT SPECIEM

341

however, does not mean that the plurality of individuals serves no purpose at all.11 Rather, as Avicenna shows, nature “intends the vague individual, which represents the species, such as some human being or some ox, because this [type of] individual is perpetual and perfect in its being.”12 Here too, Albert adds the distinction between universal and particular nature. The goal intended by universal nature (i.e., by the heavens with their parts, constellations, and motions) is the divine being that is “attained in the totality of particular [beings] through the succession [of individuals].” The goal of particular nature is the concrete transitory individual, which is “attained through one single act of reproduction.” Universal nature produces a concrete supposit “only accidentally, namely in order that it be the substrate of divine being.”13 For Albert, then, it is clear that “the species is worthier than the individual” and therefore “the preservation of the species is better than that of the individual.”14 Similar views are expressed by Thomas Aquinas. Let us look at four texts where he relies on the Avicennian topos.15 In question 1 of 11. ALBERT, o.c., p. 12, 84-94. Cf. Averroes’s Physics commentary, where it is said that “it is the intention of this book to treat of universal things constituted by nature” (Aristotelis De Physico Auditu libri octo cum Averrois [...] Commentariis, I, 4, Venetiis, Apud Iunctas, 1562, fol. 7G); see also J. FRITSCHE, Methode und Beweisziel im ersten Buch der „Physikvorlesung“ des Aristoteles, Frankfurt a. M. 1986, p. 331, n. 412. Albert’s rephrasing of Averroes is bound up with the theorem “the work of nature is the work of intelligence” (on this, see L. HÖDL, “Opus naturae est opus intelligentiae. Ein neuplatonisches Axiom im aristotelischen Verständnis des Albertus Magnus,” in: F. NIEWÖHNER – L. STURLESE [eds.], Averroismus im Mittelalter und in der Renaissance, Zürich 1994, pp. 132-148); Albert somehow anticipates Linnaeus’s well-known dictum “naturae opus semper est species et genus” (see C. LINNAEUS, Philosophia Botanica, Editio Quarta, studio Curtii SPRENGEL, Halae ad Salam, Sumtibus C. A. Kümmel, 1809, p. 177 [§ 162]). 12. ALBERT, o.c., p. 13, 1-4. Cf. AVICENNA, Liber primus naturalium, I, 1 (cf. n. 2). On the meanings of ‘vague individual’ in scholasticism, see E. J. ASHWORTH, “Medieval Theories of Singular Terms” (esp. §§ 2 and 9), in: Stanford Encyclopedia of Philosophy: https://plato.stanford.edu/entries/singular-terms-medieval (accessed in March 2022). 13. ALBERT, o.c., p. 13, 5-25. 14. ALBERT, Commentarii in Librum IV Sententiarum, 26, C, 10, ad 4, ed. S. BORGNET, Paris 1894, p. 112b. 15. On Aquinas’s debt to Avicenna in general, see J. F. WIPPEL, “The Latin Avicenna as a Source for Thomas Aquinas’s Metaphysics,” in: ID., Metaphysical Themes in Thomas Aquinas. II, Washington, D.C. 2007, pp. 31-64 (Wippel does not discuss the influence of the Avicennian dictum); see also R. C. TAYLOR (ed.), Aquinas and Arabic Philosophy (= American Catholic Philosophical Quarterly 88/2 [2014]), in particular J. JANSSENS, “A Survey of Thomas’s Explicit Quotations of Avicenna in the Summa contra Gentiles,” ibid., pp. 289-308.

342

GUY GULDENTOPS

Quodlibet VIII, Aquinas discusses how God’s ideas are related to individual creatures. In his response, he argues that God’s ideas “regard creatures not only in respect of the nature of the species but also in respect of the singularity of the individual, though primarily in respect of the nature of the species.”16 According to Edward Booth, Aquinas here connects his Aristotelian ontology with his Pseudo-Dionysian theodicy and thus adds “a very pointed correction” to Avicenna’s thesis, since “in the individual being the divine mind and will holds everything together in unity.”17 However, nothing in this question suggests that Aquinas here ‘corrected’ (or wanted to correct) Avicenna, and one may doubt Booth’s thesis that according to Aquinas “the intention of nature bears simultaneously on Socrates and the human species, but the species as found in the individual instance of Socrates.”18 Aquinas actually underlines that “a model first regards what the producer first intends in the product.” Since every producing cause primarily intends what is most perfect in its product, and since “the most perfect aspect of each individual is its specific nature,” which brings both individual matter and generic forms from an imperfect state to completion, it follows that “nature first intends the most specific species.” This means that “nature does not principally intend to generate Socrates, otherwise Socrates’ death would destroy the order and intention of nature.” Aquinas’s claim that “nature intends to generate [a] human being in Socrates” focuses on the species rather than on the individuals that instantiate it. Accordingly, Aquinas concludes that “the exemplar in God’s mind primarily regards the specific nature of any creature.” The individuals enter the divine mind only in a quasi-indirect way. Still, given that “the first thing intended is the last thing accomplished,” it is clear that what is generated first is a concrete individual (the species human being is generated only 16. THOMAS AQUINAS, Quodlibet VIII, 1, resp., ed. Leon. 25, Roma / Paris 1996, p. 54, 38-41. 17. E. BOOTH, Aristotelian Aporetic Ontology in Islamic and Christian Thinkers, Cambridge 1983, p. 264. For a similar interpretation, see F. INCIARTE, Forma formarum. Strukturmomente der thomistischen Seinslehre im Rückgriff auf Aristoteles, Freiburg / München 1970, p. 144: “Die Tatsache, daß diese gemäß einem Ziel von Gott gesetzte Ordnung auf die Arten ausgerichtet ist, besagt jedoch nicht, daß sie gleichsam über den Individuen schwebt [...]” (referring to Quodlibet VIII, 1, 2, Quodlibet VIII, 3, 2, and Summa contra gentiles, II, 24). 18. BOOTH, ibid.

NATURA INTENDIT SPECIEM

343

through the generation of concrete individual human beings).19 In this entire argument, Aquinas adopts the Greco-Arabic naturalist view on the ontological priority of the species.20 In an often quoted passage from his Questions on the Soul (more precisely in the question as to “whether a separated soul knows all natural things”), Aquinas argues that separated souls only have a general knowledge of some individual things. In this text, he explicitly claims that the perfection of nature consists in the perpetuation of the species: Individual things do not belong to the perfection of nature for their own sake but for something else, namely in order that the species, which nature intends, might be preserved in them. For nature intends to generate [a/the] human being and not this [concrete] human being, except insofar as [a/the] human being can only exist as this [concrete] human being. [...] Hence to know the species of things belongs to intellectual perfection, but not the knowledge of individual things, except perhaps accidentally.21 19. THOMAS AQUINAS, Quodlibet VIII, 1, resp., pp. 54, 42 - 55, 83 (= Thomas Aquinas’s Quodlibetal Questions, transl. by B. DAVIES – T. NEVITT, Oxford 2019, pp. 56-57; translation slightly modified). For the idea that the first thing intended is the last thing produced, cf. Auctoritates Aristotelis, 12, 49, ed. HAMESSE, p. 236, 69-70 (based on Eth. Nic., III, 5, 1112b23-24) and THOMAS AQUINAS, Summa contra gentiles, III, 66, ed. Leon. 14, Romae 1926, p. 188a23-a1. 20. Cf. J. A. AERTSEN, “Einleitung: Die Entdeckung des Individuums,” in: J. A. AERTSEN – A. SPEER (eds.), Individuum und Individualität im Mittelalter, Berlin / New York 1996, pp. IX-XVII, esp. XIV: “Aus dieser Antwort wird ersichtlich, wie stark bei Thomas die griechische Überordnung des Allgemeinen noch weiterwirkt.” (Aertsen does not refer here to Avicenna.) 21. THOMAS AQUINAS, Quaestiones disputatae de anima, 18, resp., ed. B.-C. BAZÁN, Roma / Paris 1996, p. 157, 273-287 (= Questions on the Soul, transl. by J. H. ROBB, Milwaukee 2005, p. 216-217; translation slightly modified): “Singularia namque non sunt de perfectione naturae propter se, set propter aliud: ut in eis saluentur species quas natura intendit. Natura enim intendit generare hominem, non hunc hominem, nisi in quantum homo non potest esse nisi hic homo. [...] Vnde et cognoscere species rerum pertinet ad perfectionem intelligibilem, non autem cognitio indiuiduorum, nisi forte per accidens.” (This passage influenced later authors: see e.g. ALPHONSUS TOSTATUS [d. 1455], Commentaria in Septimam Partem Matthæi, Venetiis, Apud N. Pezzana, 1728, p. 379 [quæstio 589].) On this passage, see F. MARTY, La perfection de l’homme selon Saint Thomas d’Aquin. Ses fondements ontologiques et leur vérification dans l’ordre actuel, Roma 1962, pp. 151-152; J. A. AERTSEN, Nature and Creature. Thomas Aquinas’s Way of Thought, Leiden 1988, p. 108; B. C. BAZÁN, “The Creation of the Soul according to Thomas Aquinas,” in: K. EMERY, Jr. – R. L. FRIEDMAN – A. SPEER – M. MAURIÈGE (eds.), Philosophy and Theology in the Long Middle Ages, Leiden 2011, pp. 515-569, esp. 552-553. See also Quaestiones disputatae de potentia, 10, 2, ad 12, ed. P. M. PESSION, Torino / Roma 1949, p. 261b; Summa contra

344

GUY GULDENTOPS

Aquinas’s devaluing of the knowledge of individual things and their singular properties obviously rests on the assumption that the species occupies a higher natural rank than the individual. This assumption is also presupposed in a question from the first part of the Summa theologiae. While speculating about angels and their individuation, Aquinas argues that every angel is a single species. In this context, he remarks that “the good of the species outweighs the good of an individual.” This is exactly the reason why a plurality of angelic species is better than the unrealized alternative of a multiplication of individual angels within one species.22 Furthermore, he argues that “since the numeric multiplication [i.e., the multiplication of numerically distinct individuals] can be continued infinitely, it is not intended by the agent [i.e., God].” The idea behind this inference is, of course, that what the agent cause intends is its goal or end, and this end cannot be an endless series. Therefore “the perfection of the nature of angels requires a multiplication of species, not a multiplication of individuals within one species.”23 With regard to human beings, Aquinas might seem to present an individual-centered interpretation of the intention of nature.24 However, as he explains in his question on whether there would have been procreation in paradise (another question from the Prima pars), it is gentiles, III, 59, ed. Leon. 14, p. 164a50-b2; and Summa theologiae, III, 2, 5, resp. (= Vol. 48: The Incarnate Word, 3a. 1–6, transl. by R. J. HENNESSEY, London / New York 1976, pp. 54-55). 22. THOMAS AQUINAS, Summa theologiae, I, 50, 4, resp. and ad 3. For the idea that “the good of the species is more divine than the good of the individual,” see also Super Sent., IV, 26, 1, 2, arg. 3, where he refers to ARISTOTLE, Eth. Nic., I, 1, 1094b9-10: “[...] while to do so [i.e., to achieve the good] for one person [...] is satisfactory enough, to do it for a nation or for cities is finer and more godlike” (Nicomachean Ethics, Translation [...] by C. ROWE, Oxford 2002, p. 96); cf. Summa contra gentiles, II, 45, ed. Leon. 13, Romae 1918, p. 372 (b26-32), and III, 136, ed. Leon. 14, p. 412 (a8-9). 23. THOMAS AQUINAS, Summa theologiae, I, 50, 4, ad 4. On this question, see e.g. J. H. WRIGHT, The Order of the Universe in the Theology of St. Thomas Aquinas, Rome 1957, p. 92; I. IRIBARREN, “Angelic Individuality and the Possibility of a Better World: Durandus of St Pourçain’s Criticism of Thomas Aquinas,” in: I. IRIBARREN – M. LENZ (eds.), Angels in Medieval Philosophical Inquiry. Their Function and Signification, London / New York 2008, pp. 45-59, esp. 50; G. PINI, “The Individuation of Angels from Bonaventure to Duns Scotus,” in: T. HOFFMANN (ed.), A Companion to Angels in Medieval Philosophy, Leiden 2012, pp. 79-115, esp. 84-94. 24. Cf. T. HOFFMANN, “Ideen der Individuen und intentio naturae. Duns Scotus im Dialog mit Thomas von Aquin und Heinrich von Gent,” in: Freiburger Zeitschrift für Philosophie und Theologie 46 (1999), pp. 138-152, esp. 141-142.

NATURA INTENDIT SPECIEM

345

only because of the immortality of the human souls that “nature, or rather the Maker of nature, intends the multitude of individuals for their own sake.” According to Aquinas, God alone creates individual souls and He alone instituted procreation in order to guarantee the multiplication of human beings. Hence it is somewhat inaccurate to say that nature intends the multitude of human individuals (that is the reason for inserting the correctio “or rather [...]”). Nature primarily aims at “what is everlasting and perpetual,” i.e., at the species and at individual beings insofar as they are imperishable; nature does not primarily attend to those individuals that exist only in time and temporarily.25 These four texts demonstrate that in his metaphysical analysis of creatures Aquinas attaches less value to the singularity of individual beings than to their species or specific forms.26 3. Debating the basic assumption: from Roger Bacon to Duns Scotus (and some Scotists) The view that nature intends the species soon became disputed. One of the philosophers who qualified it was Roger Bacon. In a much quoted passage from his Communia naturalium, he claims that universal nature “primarily intends and produces the individual,” while the individual is “absolutely prior” according to the intention of particular nature, i.e., according to the natural power that governs the individual. However, the significance of Bacon’s seemingly individualist claim should not be overrated, since he also affirms that “nature 25. THOMAS AQUINAS, Summa theologiae, I, 98, 1, resp. (= Vol. 13: Man Made to God’s Image, 1a. 90–102, transl. by E. HILL, London / New York 1964, p. 152). 26. Cf., more in general, G. MENSCHING, “Zur Neuentdeckung des Individuationsprinzips im 13. Jahrhundert und seinen Antinomien bei Thomas von Aquin und Johannes Duns Scotus,” in: AERTSEN – SPEER (eds.), Individuum und Individualität, pp. 287-302, esp. 291. In his eschatology, Aquinas also relativizes the prominence of the concrete individual, since he contends that God has predetermined the number of the predestinate “on the basis of the proportion of the principal parts [i.e., the proportion of human beings and angels] to the good of the universe” (see Summa theologiae, I, 23, 7, ad 2). The human individual appears to be nothing but a part that is subsidiary to the order of the whole of reality. On this passage, see E. KURZ, Individuum und Gemeinschaft beim Hl. Thomas von Aquin, München 1932, pp. 99-100: “Der Mensch ist trotz seiner Würde Teil [...]. Also auch der gute Mensch steht nicht individualistisch für sich allein, auch er [...] empfängt sein Schicksal vom Ganzen.”

346

GUY GULDENTOPS

as the power directing the universal species [...] principally produces and intends the universal.”27 Still, Bacon’s view may have influenced other Franciscan thinkers who give precedence to the individual. Aquinas’s abovementioned argument concerning the multitude of angels was criticized by William de la Mare in the Correctorium fratris Thome. According to the Franciscan theologian, Aquinas’s view that the perfection of nature does not require a multiplication of individuals is “erroneous” for two fundamental reasons. (1) Aquinas’s position implies that God’s commandment “Be fruitful and multiply!” would be absurd (He would have commanded something which He did not want or which no animal could intend, since the generation of animals does not lead to a multiplication of species, but only proliferates individuals). (2) In creating human beings, God intended the multiplication of the elect and the restoration of the fall of the angels, and that is also what a human being ought to intend in the act of procreation. This goal, however, can only be attained “through the multiplication of individuals according to matter.”28 Both of these arguments show that William was mainly led by theological considerations to question Aquinas’s metaphysical subordination of the individual. 27. See ROGER BACON, Liber primus communium naturalium, II, 3, 1, ed. R. STEELE, Oxford s.a., p. 94, 14-26. Cf. H. HÖVER, “Roger Bacons Hylomorphismus als Grundlage seiner philosophischen Anschauungen (Fortsetzung von Bd. XXV),” in: Divus Thomas. Jahrbuch für Philosophie und spekulative Theologie 26 (1912), pp. 2-53, esp. 25-26; MENSCHING, “Zur Neuentdeckung des Individuationsprinzips,” pp. 288-289; C. PANTI, “«Natura non intendit nisi quinque digitos». Caso, contingenza e mostruosità nelle Questiones supra octo libros Physicorum e nei Communia naturalium di Ruggero Bacone,” in: Rivista di Storia della Filosofia 68 (2013), pp. 65-94, esp. 85-86; C. CRISCIANI, “Universal and Particular in the Communia Naturalium: Between ‘Extreme Realism’ and ‘Experientia’,” in: P. BERNARDINI – A. RODOLFI (eds.), Roger Bacon’s Communia Naturalium. A 13th Century Philosopher’s Workshop, Firenze 2014, pp. 57-82, esp. 63-65; C. TRIFOGLI, “Avicenna’s Physics in Roger Bacon’s Communia naturalium,” in: HASSE – BERTOLACCI (eds.), The Arabic, Hebrew and Latin Reception of Avicenna’s Physics and Cosmology, pp. 433-457, esp. 441-443. 28. See P. GLORIEUX, Les premières polémiques thomistes: I. Le Correctorium Corruptorii « Quare ». Edition critique, Kain 1927, pp. 69-70. (A third objection against Aquinas claims that his argument entails that one could never intend to divide a line, since it could be infinitely divided; this objection is irrelevant to our topic.) On Quidort (d. 1306), see C. JONES, “John of Paris: Through a Glass, Darkly?,” in: ID. (ed.), John of Paris. Beyond Royal and Papal Power, Turnhout 2015, pp. 1-31; and my “Two « Platonic » Scholastics on the Soul’s Presence in the Body: John Quidort and Giles of Viterbo,” in: Archives d’Histoire Doctrinale et Littéraire du Moyen Âge 82 (2015), pp. 69-95, esp. 76-84.

NATURA INTENDIT SPECIEM

347

In his rejoinder to the Correctorium, the Dominican John Quidort argues that William misrepresents Aquinas’s position. According to Quidort, Aquinas “does not mean that the multiplication of individuals is by no means intended.” Aquinas rather wants to say that “it is intended, by any agent whatsoever, not for the sake of itself [i.e., for the multitude of individuals], but [...] for the sake of the species in order that it be preserved in the succession of individuals, since the species cannot be perpetuated in one single corporeal individual.”29 In support of this view, Quidort cites a number of philosophical authorities. First of all, Aristotle holds that “animals have been endowed with the procreative power in order that divine being, i.e., the existence of the species, be preserved.”30 More generally, he maintains that “all beings strive to be, have a tendency to preserve their being, and do everything they do for the sake of this [self-preservation].”31 Other arguments derive from the Arabic Peripatetic tradition. Algazel holds that “the multiplication of individuals within a species occurs only for the sake of the species.” Quidort seems even to be willing to endorse Averroes’s claim that divine providence takes care of individuals only indirectly, namely insofar as this “would be true if it were to be understood teleologically, i.e., with regard to the universal good of their own species or of another species.”32 Quidort strengthens this 29. JOHN QUIDORT OF PARIS, Le Correctorium Corruptorii « Circa », ed. J.-P. MULLER, Roma 1941, p. 74, 20-26. For parallel responses, see Le Correctorium Corruptorii « Quare », pp. 70-71, and P. GLORIEUX, Les premières polémiques thomistes: II. Le Correctorium Corruptorii « Sciendum ». Edition critique, Paris 1956, pp. 72-73. 30. QUIDORT, o.c., p. 74, 26-30. Cf. ARISTOTLE, De an., II, 4, 415a28-b1; AVERROES, Commentarium magnum in Aristotelis De anima libros, II, 34, ed. F. S. CRAWFORD, Cambridge (Mass.) 1953, pp. 182, 49 - 183, 56. 31. QUIDORT, o.c., p. 74, 30-31. Quidort refers to De caelo; Muller notes “locum non inveni” and refers to De gen. et corr., II, 10, 336b27 sq. However, it is more likely that Quidort has De caelo, I, 12, 283a13-17 in mind: cf. THOMAS AQUINAS, In De caelo, I, 29, 5, ed. Leon. 3, Romae 1886, p. 116a: “omnia appetunt esse.” 32. QUIDORT, o.c., p. 74, 31-37. Cf. ALGAZEL, Metaphysics, I, 4, 3, 6, ed. J. T. MUCKLE, Toronto 1933, p. 118, 4-7; AVERROES, In Metaph., XII, 37, in: Aristotelis Metaphysicorum Libri XIIII Cum Averrois [...] Commentariis, Venetiis, Apud Iunctas, 1562, fol. 320vI-K. On Averroes’s view, see R. TAYLOR, “Providence in Averroes,” in: D’HOINE – VAN RIEL (eds.), Fate, Providence and Moral Responsibility, pp. 455-472, esp. 464-467 (on the commentary on Metaphysics, XII); M. POSTI, Medieval Theories of Divine Providence, 1250– 1350, Leiden 2020, pp. 59-63; on the Latin reception of Averroes’s view, see C. MARTIN, “Providence and Seventeenth-Century Attacks on Averroes,” in: P.J.J.M. BAKKER (ed.), Averroes’s Natural Philosophy and Its Reception in the Latin West, Leuven 2015, pp. 193-212, esp. 197-201.

348

GUY GULDENTOPS

argument with quotations from Avicenna’s Metaphysics. According to Avicenna, the generated individual beings are infinite in number and “are not the essential goal of nature,” since the essential goal of nature is “the stability of being,” which cannot be preserved in one concrete perishable individual. Hence nature primarily aims at the persistence of natural species and of non-concrete individuals.33 Finally, Quidort connects the Avicennian claim about nature with his Neoplatonic theory of God’s gradual self-communication (in accord with the axiom that everything is received in the manner of the recipient34). Since divine goodness cannot pour out itself in one single creature, it communicates itself to a plurality of creatures in different degrees, according as it is communicable to creatures. However, individuals of one species do not partake of divine goodness in different degrees (among individuals of one species, there is no priority or posteriority in respect of perfection; they constitute “only a chronological order,” as Algazel observes). This, then, is the fundamental reason why nature (or God) does not principally intend to engender or to multiply individuals belonging to one species.35 After defending Aquinas’s position, Quidort attempts to refute William’s arguments against Aquinas. William’s reference to Genesis 1: “Be fruitful and multiply!” is misleading: although God “intended the number of the elect, the restoration of the [fallen] angels, and the perpetuation of the species in many individuals,” this does not imply that human beings, when reproducing themselves, also intend all of these goals; it suffices for them to intend their individual good. At this point, Quidort refers to Avicenna’s Metaphysics and Physics, where universal nature (“which seeks the good of the species”) is distinguished from particular nature (“which seeks the good of the individual”). Furthermore, Quidort divides universal nature into “naturing Nature” (identified with God) and “natured nature” (cautiously identified with the celestial power that shapes the active powers of 33. QUIDORT, o.c., pp. 74, 37 - 75, 47. Cf. AVICENNA, Liber de philosophia prima, VI, 5, ed. VAN RIET, p. 334, 31-42. 34. On this principle, see C. D’ANCONA COSTA, Tommaso d’Aquino. Commento al « Libro delle Cause », Milano 1986, p. 272. 35. QUIDORT, o.c., p. 75, 47-62. For the idea that individuals of one species do not differ according to the priority of perfection, he refers not only to ALGAZEL, Metaphysics, I, 1, 6, ed. MUCKLE, p. 41, 4-9, but also to ARISTOTLE: An. post., II, 13, 97b30; Phys., VII, 4, 249a22-25; and Metaph., III, 3, 999a12-13.

NATURA INTENDIT SPECIEM

349

sublunary substances). On the basis of these distinctions, he points out that “what occurs against the intention of particular nature often occurs in accord with the intention of universal created nature.” Moreover, when universal created nature directs a process toward one determinate goal, naturing Nature can “direct it toward another superior goal.” This difference of goals is illustrated with the Aristotelian example of the generation of females: even though females are “against the intention of particular nature” (according to Aristotle, they are “failed males,” “due to a bad disposition of matter”), universal nature aims at the generation of both males and females “for the sake of the species.” Accordingly, the multiplication of individuals belonging to one species is “perhaps not intended by the particular nature of the individual,” since this individual nature only “intends to ejaculate the superfluous residuum of the digestion for the sake of its individual good, except perhaps inasmuch as this particular nature is shaped by universal nature.” Universal natured nature, to which the reproductive power of the individuals is subordinated, intends to generate “some [indeterminate] individual” in order to perpetuate the species. In addition to this, naturing Nature (i.e., God) directs the multiplication of individuals “toward a higher goal, namely the restoration of the fall of the angels or the sharing in His glory or something else.” These divine goals, however, are quite different from what human individuals intend in their sexual activities. In short, then, the multiplication of individuals is not intended as a goal in itself, since it could be infinitely continued. Still, it can serve “as [a means leading] toward the goal,” i.e., with a view to the divine goals just mentioned.36 36. QUIDORT, o.c., pp. 75, 63 - 76, 95. Cf. AVICENNA, Liber de philosophia prima, VI, 5 (cf. n. 33); Liber primus naturalium, I, 7, ed. VAN RIET, p. 66, 29-35 and pp. 68, 64 - 69, 86. On naturing Nature and natured nature, cf. J. M. MARLING, The Order of Nature in the Philosophy of St. Thomas Aquinas, Washington, D.C., 1934, pp. 9-12; K. HEDWIG, “Natura naturans/naturata,” in: Historisches Wörterbuch der Philosophie, Bd. 6, Darmstadt / Basel 1984, cols. 504-509, esp. 504-505. Quidort attributes the distinction of the two levels of nature to “Tullius”; Cicero, however, does not use these medieval terms, but Quidort may have been influenced by twelfth-century Platonists who mentioned Cicero’s definitions of ‘nature’ (cf., e.g., WILLIAM OF CONCHES, Dragmaticon philosophiae, I, 7, ed. I. RONCA, Turnhout 1997, p. 30, 26-31; see also A. SPEER, Die entdeckte Natur. Untersuchungen zu Begründungsversuchen einer “scientia naturalis” im 12. Jahrhundert, Leiden 1995, p. 189). For the view of the female as a mas occasionatus, cf. ARISTOTLE, De gen. an., II, 3, 737a25-28 (Quidort may have been influenced by Albert’s Quaestiones super De animalibus, where it is also argued that the generation of a female does not accord with the intention of particular nature; cf. n. 8).

350

GUY GULDENTOPS

Quidort’s defense of Aquinas clearly leads him to emphasize the ontological inferiority of the individual. In his view, the individual is subordinated not only to the species, but also to universal natured nature and to divine providence, which does not seem to be deeply interested in the individual as such. Even though Aquinas is convinced that God knows all creatures in their individuality,37 his understanding of divine knowledge and providence was sharply attacked by Duns Scotus. As Scotus observes in his Sentences commentary, the claim that “the individual does not belong to the intention of nature” seems to be contradicted by the claim that “divine providence is primarily concerned with individuals.” To avoid this contradiction, Scotus rejects the former claim. If nature, directed by God, produces individuals, and if providence works “not only in the species but [also] principally in individuals,” it necessarily follows that “the intention of nature is concerned not only with the species but [also] principally with the individuals that are to be produced.”38 At least on this topic, Scotus puts forward a strongly individualist interpretation of the Avicennian topos. In his Questions on the Soul, by contrast, Scotus embraces a view that is more in line with Avicenna’s own position. There Scotus notes that “nature primarily intends to produce a nature [i.e., a common essence] in some supposit, and that [i.e., what nature primarily intends to produce] is a vague individual.” In this context, Scotus draws on Avicenna’s authority in support of his non-Thomistic theory of how individuals 37. See e.g. THOMAS AQUINAS, Summa theologiae, I, 14, 11, resp.; I, 15, 3, ad 4. Cf. G. T. DOOLAN, Aquinas on the Divine Ideas as Exemplar Causes, Washington, D.C., 2008, pp. 124-133; V. CORDONIER, “La doctrine aristotélicienne de la providence divine selon Thomas d’Aquin,” in: D’HOINE – VAN RIEL (eds.), Fate, Providence and Moral Responsibility, pp. 495-515, esp. 499-506. 38. DUNS SCOTUS, Reportata Parisiensia, I, 36, 4, esp. § 31, ed. T. B. NOONE, in: ID., “Scotus on Divine Ideas. Rep. Par. I-A, d. 36,” in: Medioevo 24 (1998), pp. 359-453, at p. 436 = The Examined Report of the Paris Lecture: Reportatio I-A. Latin Text and English Translation, Vol. Two, [ed. and transl. by] A. B. WOLTER – O. V. BYCHKOV, St. Bonaventure 2008, p. 420, § 115 (I modified Wolter and Bychkov’s translation; in particular, their translation of intentio nature as “nature’s effort” is inexact). On Scotus’s view, see É. GILSON, John Duns Scotus. Introduction to His Fundamental Positions, transl. by J. G. COLBERT, London / New York 2019, pp. 232-233; T. B. NOONE, “Aquinas on Divine Ideas: Scotus’s Evaluation,” in: Franciscan Studies 56 (1998), pp. 307-324, esp. 319-320; HOFFMANN, “Ideen der Individuen und intentio naturae,” pp. 147-149; and R. CROSS, Duns Scotus on God, Aldershot / Burlington 2005, pp. 77-83.

NATURA INTENDIT SPECIEM

351

can be known (the first stage in this process is the imagining of a vague individual toward which the intellect first turns its attention).39 Scotus’s preference for Avicenna’s view that nature intends the vague individual influenced later Scotist discussions on divine ideas. In his lengthy question on divine ideas, the early-fourteenth-century Franciscan John of Bassoles inveighs against those who deny the existence of ideas of individuals. He employs the very same argument Scotus used against Aquinas’s view on the intention of nature, and adds that “it is astonishing that nature, while intending second substances or species per se, would accidentally intend first substances, which are substances to the highest degree and in the truest sense of the word.” Furthermore, he argues against the view that “nature intends individuals only in order to conserve the species.” According to Bassoles, this view is obviously false in the case of angels; and in general, nature intends the species more for the sake of the individuals than the other way round, “because there is something more perfect in the individual than in the species.” Even though nature intends the generation of individuals in order to conserve the species, “[it does so], not without the individual, but rather in the individual and ultimately for the sake of the individual.”40 This argument rests on Scotus’s theory that, ontologically speaking, the hecceitas, i.e., that which characterizes the individual, is “the utmost perfection” of this being.41 Bassoles thus clearly 39. DUNS SCOTUS, Quaestiones super Secundum et Tertium De anima, II, 22, ed. C. BAZÁN – K. EMERY, Jr. – R. GREEN – T. NOONE – R. PLEVANO – A. TRAVER, Washington, D.C. / St. Bonaventure 2006, p. 237, 2-5. On this question, see BLACK, “Avicenna’s ‘Vague Individual’,” pp. 280-284. On Scotus’s theory of human knowledge of singulars, see e.g. R. CROSS, Duns Scotus’s Theory of Cognition, Oxford 2014, pp. 18-22, 43-57, and 74-77. 40. Opera JOANNIS DE BASSOLIS [...] In Quatuor Sententiarum Libros (credite) Aurea [...], , Venundantur a F. Regnault et I. Frellon, Parisiis, 1517, I, d. 36, q. 3, a. 2, fols. 193vb-194ra. One of John of Bassoles’s targets is Henry of Ghent: cf. Quodlibet II, 1, resp., ed. R. WIELOCKX, Leuven 1983, pp. 6, 76 7, 5; Quodlibet IX, 2, ad 2, ed. R. MACKEN, Leuven 1983, p. 45, 51-55. On John of Bassoles, see W. COURTENAY, “Early Scotists at Paris: A Reconsideration,” in: Franciscan Studies 69 (2011), pp. 175-229, esp. 202-207. 41. DUNS SCOTUS, Lectura, II, 3, 1, ed. L. MODRIĆ et al., Civitas Vaticana 1982, pp. 282, 30 - 283, 7 and p. 283, 23-25; Ordinatio, II, 3, 1, ed. C. BALIĆ et al., Civitas Vaticana 1973, p. 479, 1-2 and 19-23, and p. 483, 18-20. On this view and its background, see AERTSEN, “Einleitung: Die Entdeckung des Individuums,” pp. XII-XIII; O. BOULNOIS, Lire Le Principe d’individuation de Duns Scot (Ordinatio II, d. 3, p. 1), Paris 2014, pp. 207208; A. LAZELLA, The Singular Voice of Being. John Duns Scotus and the Ultimate Difference, Fordham 2019, pp. 168-171.

352

GUY GULDENTOPS

takes up Scotus’s individualist interpretation of Avicenna’s view on the intention of nature. Some three centuries later, the Observant Peter of Poznań repeats Bassoles’s argument in his commentary on Scotus’s Sentences commentary. Although he does not agree with all elements of Bassoles’s doctrine of divine ideas, he does not criticize his individualist understanding of the Avicennian topos.42 4. John Baconthorpe’s defense of Avicenna The eclectic fourteenth-century Carmelite John Baconthorpe digresses on the Avicennian dictum in his Trinitarian theology.43 In his commentary on distinction 5 of book I of Lombard’s Sentences, he examines “whether generation primarily ends in something similar or rather in something distinct.” As Baconthorpe notes, this question has its remote roots in Henry of Ghent’s claim that the Father “generates [the Son as] something that is absolutely and really identical and distinct only because it derives its being from something else.”44 (Another source is Scotus’s thesis that “generation is both distinctive and assimilative.”45) According to Baconthorpe, however, this question makes sense only with regard to the generation of creatures and it is pointless to discuss it with regard to the divine persons.46 42. Commentaria in Doctrinam D. S. Ioannis Duns Scoti [...] Auctore [...] PETRO POSVenetiis, Ex Officina Marci Ginammi, 1627, I, dist. 35, disputatio quinta, p. 521a. On Piotr Bieliński, see J. SCHMUTZ, Scholasticon: https://scholasticon.msh-lse.fr/ Database/Scholastiques_fr.php?ID=1042. 43. On Baconthorpe (d. ca. 1348), see e.g. Th. JESCHKE, “Scrivere la storia della filosofia nel trecento. Giovanni Baconthorpe e il dibattito sulla beatitudine nell’atto riflessivo (ca. 1293-1320),” in: Giornale critico della filosofia italiana 92/1 (2011), pp. 57-75; R. L. FRIEDMAN, Intellectual Traditions at the Medieval University. The Use of Philosophical Psychology in Trinitarian Theology among the Franciscans and Dominicans, 1250–1350, Leiden / Boston 2013, pp. 768-777. 44. HENRY OF GHENT, Summæ Quæstionum Ordinariarum [...] Tomos Posterior, 58, 1, Vænundatur in ædibus IODOCI BADII, 1520, fols. 124vN-125rO = Summa (Quaestiones ordinariae), art. LVI-LIX, ed. G. A. WILSON – G. J. ETZKORN – B. GOEHRING – L. N. ETZKORN, Leuven 2021, pp. 105, 248 - 106, 286. 45. DUNS SCOTUS, Lectura, I, 7, 1, ed. C. BALIĆ, Civitas Vaticana 1960, pp. 492, 20 493, 3. Scotus adds that “nature first intends the similar before intending the distinct, and this [i.e., the similar] is the vague individual, not the fully articulated or distinct individual.” In this question, Scotus argues that the generation of the Son is a univocal generation. 46. BACONTHORPE, In Sent., I, 5, 1, 5, in: DOCTI RESOLUTI IO. BACHONIS [...] Quæstiones in Quatuor Libros Sententiarum, & Quodlibetales, a R. P. IO. CHRYSOSTOMO MARASCA NANIENSE,

NATURA INTENDIT SPECIEM

353

Applied to creatures, the question can have two answers. (1) The first answer compares the generation of creatures with the generation of divine persons. If the word ‘similar’ refers to “a substantial form that is the beginning and the end of generation,” and if ‘distinct’ refers to “a relation that is entailed by the substantial form as the beginning and the end of generation,” the question can easily be answered: in the realm of creatures, as for God, the similar has priority. (2) If one focuses exclusively on the generation process in creatures, it is evident that the word ‘distinct’ denotes an individual, whereas ‘similar’ denotes the nature of a species. Accordingly, one can ask whether generation primarily ends in a concrete individual or rather in the specific nature. With regard to God, this question is meaningless, since the deity or divine nature is not a species that comprises different individuals. Still, if one wants to save Henry’s speculations, one might say that “just as man generates and is generated only because this man [generates] and this man [is generated], likewise God generates and is generated only because this God, i.e, the Father, [generates] and this God, i.e., the Son, [is generated].” Even though this latter claim is true only “according to our way of understanding,” it is clear that in this sense “something distinct is generated primarily.” Yet, Baconthorpe stresses that he holds this view only because of the individuality of every act: “Just as an act is necessarily concerned with singulars, so it must necessarily intend something distinct as its end.” According to the “intention of nature,” however, generation “primarily ends in something similar in both realms [i.e., in creatures and in God].” To buttress this claim, Baconthorpe cites Avicenna’s Metaphysics VI, 5, where he says that “the primary intention of nature is the permannent existence of human nature or of a similar nature or of a perpetual non-designated individual,” and “this intention is the perfect cause of the actuality of universal nature,” which accidentally requires “an infinite series of individuals.”47

[...] dilucidatæ, Conciliationibus [...] locupletæ, Cremonæ, Apud M. A. Belpierum, 1618, p. 133aA. 47. BACONTHORPE, o.c., p. 133aB-E. Baconthorpe cites AVICENNA, Liber de philosophia prima, VI, 5, ed. VAN RIET, p. 334, 33-34 and 38-42. Instead of “illa intentio est causa perfectiva actionis naturae uniuersalis,” the edition of Baconthorpe has: “illa intentio est causa perfecta actualitatis naturæ vniuersalis.”

354

GUY GULDENTOPS

However, Baconthorpe observes that this view is apparently contradicted by Aristotle, who in book II of De anima “seems to hold that owing to the intention of nature it is not the species that is primarily intended as the essential goal of nature.” Since Aristotle says that “it is most natural for living beings to produce something else,” he apparently means that “a perpetual individual is primarily intended owing to the intention of nature and not only owing to a feature [i.e., the individuality] of an act.” A confirmation of Aristotle’s alleged prioritization of the individual is found in Averroes. In his commentary on De anima, he claims that “since divine providence could not achieve perpetual permanence in an individual, it mercifully achieves it on the level of the species.” This obviously means that nature primarily intends the individual, at least if the primary intention of nature is to be identified with its essential goal. To make this point clearer, Baconthorpe explains the distinction between essential and accidental goals. While the essential goal of nature is “the goal that is absolutely intended,” the accidental goal is “what is intended under a [determinate] condition, namely because the primary and essential goal cannot be reached.” On the basis of this distinction, one has to conclude that nature primarily intends the individual. According to Aristotle and Averroes, then, it seems clear that “absolutely speaking, nature intends to perpetuate itself through generation in what it generates.” If nature is said “to perpetuate itself in the species because none of the corruptible beings is perpetual,” this must mean that the species is only accidentally intended by nature.48 Baconthorpe, however, is well aware that this Averroean argument misconstrues Aristotle’s species-centered teleology. Therefore, he embarks on a defense of Avicenna’s view, averring that “on this issue Avicenna is the best interpreter of Aristotle and not an antagonist with a contrary opinion.” In order to prove this, Baconthorpe points out that the phrase ‘individual intended by nature’ can be understood in two ways. Either it refers to the concrete individual that is subsumed under the species (e.g., the concrete son whom Socrates intends to generate); or it refers to the individual “that is the entire species,” i.e., what Avicenna calls “the perpetual non-designated individual.” In 48. BACONTHORPE, o.c., pp. 133aE-133bB. Cf. ARISTOTLE, De an., II, 4, 415a28-b1; AVERROES, In De an., II, 34, ed. CRAWFORD, pp. 182, 49 - 183, 56 (cf. n. 30).

NATURA INTENDIT SPECIEM

355

that sense, Averroes’s interpretation is also correct: “What nature primarily and principally intends is the perpetual individual.” From all this, Baconthorpe infers a twofold conclusion: (1) nature primarily intends the nature of the species (identified with the perpetual nondesignated individual); (2) consequently, owing to the intention of nature, generation primarily ends in something similar rather than in something distinct.49 Still, as Baconthorpe notes, Aristotle and Avicenna are confronted with a serious difficulty: if the primary and essential goal of nature is the perpetuation of the species, and if nature cannot achieve this without a perpetual individual, two un-Aristotelian consequences will follow, namely (1) that “nature will reach its goal only accidentally,” and (2) what is even more absurd, that “nature essentially intends what is impossible.” Baconthorpe offers a sophisticated reply to this objection. What nature primarily and essentially intends can be considered in two ways: either according to the order of perfection (i.e., the order according to which the best is always principally intended); or according to the order of generation (i.e., the order according to which something is actualized). If one considers the first order, “it is evident that the nature of the species or the perpetual individual that is identical with the entire species is more perfect than any [concrete] individual subsumed under the species.” In this sense, the essential goal of nature is the nature of the species or the perpetual individual. However, according to the order of generation, which moves from the imperfect toward the perfect, the concrete individual is what is primarily intended, whereas the species is just accidentally intended. Given the distinction between the order of perfection and the order of generation, Baconthorpe is also able to show that nature does not aim at anything impossible. If nature primarily intends the species according to the order of perfection, this natural intention is not in 49. BACONTHORPE, o.c., p. 133bB-D: “Respondeo quod Auicenna in isto facto est optimus expositor Aristotelis et non contrarius opinator. [...] indiuiduum potest esse intentum a natura dupliciter: uno modo quod sit indiuiduum determinatum hoc vel illud, puta aliquod indiuiduum sub specie, scilicet quod Sortes intendit hunc filium suum generare [...]; alio modo quod natura non intendit generare vnum indiuiduum non contentum sub specie, sed que sit tota species, et illud vocat Auicenna indiuiduum perpetuum non designatum. Per hoc ad argumentum dico quod verum est quod vult, quod natura primo et principaliter intendat indiuiduum perpetuum et non indiuiduum contentum sub specie, quod sit distinctum ab alio indiuiduo, sed indiuiduum quod sit tota species.”

356

GUY GULDENTOPS

vain, since it “always can have the nature of the species or the perpetual individual as its intended object.”50 According to this order, nature’s “supreme perfection” is the perpetual existence of the species. While nature cannot produce this immediately or without individuals, “that is no wonder, since producing belongs to a different order, namely to the order of generation.” In God, however, there is only the order of perfection, according to which “the Son is generated in an equally primary way as the Son and as God, because He is similar [to the Father], both with regard to the intention of nature and to the order of the act [of generating].” Still, if we apply our way of speaking about generation to God, we can say that according to the intention of nature He is primarily generated as the similar and as God, while He is generated primarily as the Son and as distinct in respect of the individuality of the act of generating.51 In this intricate question, Baconthorpe sides with Avicenna and assumes that the species or the vague individual is more noble than the concrete individual. As Chrysostomus Marasca, Baconthorpe’s seventeenth-century editor, observes, this view seems to be in contradiction with what Baconthorpe claims in his commentary on distinction 3 of book I of the Sentences, where “he holds that the individual is more perfect than the universal.” However, according to Marasca, both views can be reconciled. He presents two arguments to back up this harmonization. (1) In the commentary on distinction 3, the individual is compared with the species as such: in this manner, the individual is more noble, because “the concept of the individual virtually contains the concept of the species, and the individual is something complete and ultimate.” In the commentary on distinction 5, by contrast, the individual is compared with the species inasmuch as it subsumes all individuals. In this sense, the species is more perfect, since “it encompasses a larger number of perfect beings and the generation of this or that individual in particular is directed toward the species,” which is perpetuated by the totality of individuals. (2) The individual excels the species “in beingness and completion, because the individual cannot be contracted [i.e., determined by additional particularizing limitations] or receive an addition, as the species can.” The 50. BACONTHORPE, o.c., p. 134aA-E. 51. BACONTHORPE, o.c., p. 134aE-bD.

NATURA INTENDIT SPECIEM

357

species, however, surpasses the individual “in extension and inasmuch as it is an essential and constitutive part of the universe, of which the individual is only an accidental part.”52 Marasca’s comments on Baconthorpe’s views highlight the ambivalence of both theologians toward the species and the individual. Depending on the perspective from which one looks at the individual and the species, either can have a sort of priority. 5. The Avicennian dictum in the quarrel between realists and nominalists The importance of the Avicennian dictum in the debate between realists and nominalists is underlined by a late-fifteenth-century document related to the Wegestreit in Ingolstadt. This document catalogues the issues on which the moderni contradict the doctrine of the reales, and also mentions the question as to “whether a species is produced through natural generation, since the author of the [Book of ] Causes asserts that nature intends the species, and what kind of species that is, whether it is a concept or rather a real nature.”53 52. MARASCA, Conciliatio XVII, in: DOCTI RESOLUTI IO. BACHONIS [...] Quæstiones in Tertium, & Quartum lib. Sententiarum, & Quodlibetales, a R. P. IO. CHRYSOSTOMO MARASCA [...] dilucidatæ, Conciliationibus [...] locupletæ, Cremonæ, Apud M. A. Belpierum, 1618, p. 783b. For the context of the question on the “first known according to the priority of perfection,” see W. GORIS, “Between Unity and Perceptibility – Richard Conington and the Concept of Being,” in: EMERY – FRIEDMAN – SPEER – MAURIÈGE (eds.), Philosophy and Theology in the Long Middle Ages, pp. 189-211, esp. 205, n. 40, and ID., “The Critique of the Doctrine of God as First Known in the Early Carmelite School,” in: EMERY – COURTENAY – METZGER (eds.), Philosophy and Theology in the Studia of the Religious Orders and at Papal and Royal Courts, pp. 493-526, esp. 519-526. For the view that the individual is something perfect or ultimate, see BACONTHORPE, o.c., I, 3, pp. 96bB-97aA, esp. 96bD-E; cf. supra, n. 41, and ARISTOTLE, Metaph., X, 9, 1058b8-10, as well as THOMAS AQUINAS, In duodecim libros Metaphysicorum Aristotelis expositio, X, 11, ed. M.-R. CATHALA – R. M. SPIAZZI, Taurini 1964, p. 502b (§ 2132) and Qu. disp. de anima, 1, resp., ed. BAZÁN, p. 7, 197-200. For the idea that the species are the essential parts of the universe, cf. e.g. THOMAS AQUINAS, Scriptum super libros Sententiarum, II, 3, 1, 4, ad 3, ed. P. MANDONNET, Parisiis 1929, p. 98; II, 17, 2, 2, ad 6, p. 433; Summa theologiae, I, 23, 7, resp. (cf. supra, n. 8). 53. See the “Ingolstädter Schriftstücke über den Wegestreit (Wende des 15. und 16. Jahrhunderts),” in: F. EHRLE, Der Sentenzenkommentar Peters von Candia, des Pisaner Papstes Alexanders V. Ein Beitrag zur Scheidung der Schulen in der Scholastik des vierzehnten Jahrhunderts und zur Geschichte des Wegestreites, Münster i. W. 1925, p. 337. The reference to the Liber de causis is puzzling; perhaps one should conjecture “Auicenna” instead of “autor causarum” (unfortunately, this unique MS., Munich, Universitätsbibliothek, 2° Cod. 482, was destroyed during World War II). On this document, see M. HOENEN,

358

GUY GULDENTOPS

In this section, we will look at Walter Burley and some of his followers and critics. In a digression within his second commentary on chapter 1 of book I of the Physics, Burley argues against Ockham’s conceptualist view that “a species is nothing but some simple concept that exists in the mind, since outside the intellect there is only the singular [thing].”54 In Burley’s opinion, there exists “something outside the soul that is not a singular thing.” He adduces three complex arguments for this view, the first of which is based on the Avicennian thesis that nature primarily intends a species. Given that what nature primarily intends is also produced by nature, and since this product is a natural and extramental thing, the conclusion must be that a species is something extramental. Burley is aware that the Avicennian topos is disputable. Therefore, to strengthen his argument, he addresses two objections. (1) The first objection is that “nature primarily intends an individual thing [...] and Avicenna was wrong in claiming that it is the species that is intended by nature.” Burley’s reply to this objection consists of two arguments. (i) The intention of nature that inheres in a particular inanimate agent, for instance the intention of this concrete fire, always remains identical; unlike the purpose of a free agent, it never changes. If fire a intends to generate fire b, it follows that as soon as b has been generated, a will no longer have a natural intention (or if it still had the intention to generate anything, it would intend to generate b, but in that case the intention of nature would be pointless, since b already exists and cannot again be produced by nature). According to Burley, “this argument can be derived from Avicenna’s words.” (ii) His second argument is a sort of reductio ad absurdum. “‚Secundum vocem concordare, sensu tamen discrepare.‘ Der Streit um die Deutung des Aristoteles an der Universität Ingolstadt im späten 15. und frühen 16. Jahrhundert,” in: A. FIDORA – J. FRIED – M. LUTZ-BACHMANN – L. SCHORN-SCHÜTTE (eds.), Politischer Aristotelismus und Religion im Mittelalter und Früher Neuzeit, Berlin 2007, pp. 67-87, esp. 67-71. 54. Burley clearly reacts against Ockham’s conceptualist comments on Physics, I, 1, 184a24-26 (i.e., the passage Burley comments on here): see OCKHAM, Expositio in libros Physicorum Aristotelis, I, 1, ed. V. RICHTER – G. LEIBOLD, St. Bonaventure, N.Y., 1985, p. 29, 12-18. On Burley (d. 1344 or later), see M. VITTORINI, “Life and Works,” in: A. D. CONTI (ed.), A Companion to Walter Burley. Late Medieval Logician and Metaphysician, Leiden / Boston 2013, pp. 17-47.

NATURA INTENDIT SPECIEM

359

In the same measure in which an agent produces one individual, it could also produce another individual of the same species: in the same measure in which fire a produces fire b, it could also produce fire c. Hence, if a intended determinately to produce b, the consequence would be that if a produced c in the same measure, this action would be in vain, since a would not produce what it intends (namely b). Accordingly, the assumption that nature intends an individual is false. (2) The second objection Burley presents comes from opponents who concentrate on the vague individual and seem to hold a moderately realist view on universals. They argue that nature does not intend the concrete individual as such, but only insofar as it represents its species (nature intends, for example, “Socrates insofar as he is a human being, which is the name of the species”). However, this objection is also unconvincing, for in the same manner in which nature intends Socrates qua human being, it would intend every human being qua human being, but that is impossible: since nature is unable to produce every human being qua human being through one determinate action, its actions would be in vain. Finally, Burley confronts his opponents with a dilemma: Does nature primarily intend to produce a universal or an individual? If they answer “a universal,” they are compelled to agree with his thesis. If, however, their answer is “an individual,” it would follow that just as nature intends one individual, it would also intend any other individual. In each of its actions, then, nature would intend to produce “any individual whatsoever,” but since this is unrealizable, nature would fall short of what it intended to achieve, which is impossible because according to Aristotle nature does nothing in vain. Moreover, if nature primarily intends to produce Socrates qua human being, it primarily intends to produce some human being and consequently a species.55 55. WALTER BURLEY, Super octo libros Phisicorum, Venetijs impressa per Simonem de Luere [...] 1501, fol. 8rb-va (commentary on “Unde ex uniuersalibus ad singularia oportet procedere [...] posterius autem determinant horum unumquodque” = Phys., I, 1, 184a23b14). On Burley’s realism concerning universals, see H.-U. WÖHLER, “Universals and Individuals,” in: CONTI (ed.), A Companion to Walter Burley, pp. 167-189, esp. 176 (according to Wöhler, the premise “that which nature primarily strives toward is not a singular” “is based on a concrete and immediately intelligible empirical fact”; he does not notice the Avicennian origin of this ‘fact’). See also CONTI, “Walter Burley” (esp. §§ 6 and 8), in: Stanford Encyclopedia of Philosophy: https://plato.stanford.edu/entries/burley/

360

GUY GULDENTOPS

Burley’s anti-individualist interpretation of the Avicennian dictum provoked both positive and negative reactions. Several realist philosophers are influenced by Burley’s interpretation of Avicenna. Let us consider three of them: Francis of Prato, an early-fourteenth-century Dominican; John Sharpe (or Scharp), a late-fourteenth-century German theologian active in Oxford; and Petrus Nigri, a notorious fifteenthcentury Dominican.56 In his Treatise on Universals, Francis explains how Avicenna is to be understood in detail: Univocal nature, which produces several effects that belong only to one and the same species, does not primarily intend the genus (according to the primacy of adequateness, i.e., as an adequate effect), since it does not extend itself beyond its own species. Nor does it primarily intend an individual as its adequate effect, since it produces several individuals. Rather, it primarily intends the species as its adequate effect. Or one could say that such a univocal nature primarily intends the species because it renders its effects similar to itself through specific beingness. [...] If, however, nature were an equivocal cause, which produces effects belonging to different species (as the heavens are [an equivocal cause]), then such a nature would not primarily intend the species as its adequate effect; rather it would intend the genus.57 (accessed in March 2022). It is not clear what Burley means with “in the measure.” According to John of Venice, mensura is used here, not in the sense of “power” (virtus), but rather in the sense of “portion of time” (tempus): see his lectures on Burley’s Physics commentary, in: S. CAROTI, Giovanni da Venezia, un tardoscolastico nelle Università toscane, Firenze 1986, p. 167. 56. On these scholastics, see C. RODE, “Francis of Prato,” in: ID. (ed.), A Companion to Responses to Ockham, Leiden 2016, pp. 273-302; A. CONTI, “Johannes Sharpe,” in: Stanford Encyclopedia of Philosophy: https://plato.stanford.edu/entries/sharpe (accessed in March 2022); S. T. BONINO, “La question de l’intellect agent dans le Clipeus Thomistarum (1481) de Pierre Schwarz,” in: Revista Española de Filosofía Medieval 9 (2002), pp. 163183; M. DIEMLING, “Petrus Nigri (Peter Schwarz). Fifteenth-Century Polemicist, Preacher and Hebraist,” in: E. H. FÜLLENBACH – G. MILETTO (eds.), Dominikaner und Juden / Dominicans and Jews. Personen, Konflikte und Perspektiven vom 13. bis zum 20. Jahrhundert, Berlin 2015, pp. 299-317; E. JINDRÁČEK, “The Western Reception of Aquinas in the Fifteenth Century,” in: M. LEVERING – M. PLESTED (eds.), The Oxford Handbook of the Reception of Aquinas, Oxford 2021, pp. 93-102, esp. 96. 57. FRANCIS OF PRATO, Tractatus de universalibus, in: I trattati De universalibus di Francesco da Prato e Stefano da Rieti, ed. F. AMERINI, Spoleto 2003, p. 131, 304-319 (cf. p. 80, 151-163): “Et quando Avicenna dicit quod natura [...] intendit primo speciem, dico quod natura univoca [potens vel] producens plures effectus eiusdem speciei tantum, non intendit primo genus primitate adequationis, idest tamquam effectum adequatum, cum non extendat [conieci: intendat ed.] se ultra speciem suam; nec intendit primo individuum tamquam effectum adequatum, cum producat plura individua; sed intendit primo

NATURA INTENDIT SPECIEM

361

This thesis is reminiscent of what Aquinas says in question 1 of Quodlibet VIII: the species is the most perfect aspect of the individual and as such it is the goal of nature. Francis, however, nuances this view as follows: even though nature intends to produce one single being by one single action, nature’s action taken commonly (i.e., in the sense of an action belonging to different individuals of the same species) aims at the production of several individual beings through several actions (this production of a multitude of individuals is also a real action, though its unity is not real, but only conceptual). Nevertheless, for Francis, it is undeniable that “nature, by its own activity (taken commonly), intends a universal effect, and not a singular effect.”58 In his Question on Universals, Sharpe also cites the Avicennian dictum to substantiate his realist view on universals: “Whatever nature intends per se belongs to the thing [i.e., to reality] or can belong to it. However, nature per se intends the species, not an individual. Consequently, the species belongs to the thing (thereby we distinguish thing from sign).” The major premise rests on the assumption that “the intention of nature bears upon a natural thing, not upon an artificial thing.” Given that all signs are artificial, nature intends something real and not some sort of sign. For the minor premise, Sharpe refers to Avicenna’s Physics. Furthermore, he corroborates the Avicennian argument with three additional arguments. (1) If the intention of unconscious nature were directed toward the individual (as Buridan claims),59 this intention would be as variable as the intention of a voluntary agent. This is not the case, since unconscious nature is determinately directed toward one goal. (2) By nature, irrational animals are more affected (or loved) by members of the same species speciem tamquam effectum sibi adequatum. [alio modo] Vel potest dici quod talis natura univoca intendit primo speciem, quia effectus suos assimilat sibi entitate specifica. Et quamvis illud quod intenditur a natura debeat esse verum ens reale, non tamen habetur quod debeat esse unum ens reale, cum sit suppositum quod talis natura producat plures effecus reales. Si vero natura esset causa equivoca, producens effectus diversarum specierum, sicut est celum, tunc talis natura non intenderet primo speciem tamquam effectum sibi adequatum, sed primo intenderet genus.” On Francis’s theory of universals, see C. RODE, Franciscus de Prato. Facetten seiner Philosophie im Blick auf Hervaeus Natalis und Wilhelm Ockham, Wiesbaden 2004, pp. 184-204; on Burley’s influence on Francis, see AMERINI, “14th-Century Reactions to Burley,” in: CONTI (ed.), A Companion to Walter Burley, pp. 377-409, esp. 395-397. 58. See FRANCIS OF PRATO, o.c., pp. 130, 292 - 131, 304. 59. Cf. infra, n. 63.

362

GUY GULDENTOPS

than by members of a different species. This emotional reaction, however, is not due to some individual feature that inheres in one single animal and not in another, since all animals belonging to one species are equally affected by any other member of the same species. Hence their emotional reaction must be due to a common feature. (3) Nature intends the conservation of the species more principally than it intends the conservation of an individual. However, what it intends to preserve is not a notion or a sign. Thus, it principally intends the conservation of a real species. In this context, Sharpe invokes Aristotle’s De anima for the view that “the multitude of individuals exists for the sake of the conservation of the species.” Consequently, nature intends the conservation of an individual for the sake of the conservation of the species, and since the cause (including the final cause) of a thing has that thing’s properties to a higher degree, it follows that nature intends the conservation of the species more than the conservation of an individual.60 In Petrus Nigri’s Shield of the Thomists, the Avicennian dictum is also cited as a proof of realism. The minor premise of one of his twelve extensive arguments for his realist view on universals is based on Avicenna. Something which nature per se intends to preserve is a real being; nature per se intends the conservation of the species, not the conservation of the individual; consequently, the species is a real being. The major premise is clear, since what nature intends is acquired through the operation of nature, and what is acquired through the operation of nature is not a mental being, but rather a real being. The minor premise is proved not only “by the authority of Avicenna” (Nigri cites a line from book I of Avicenna’s Metaphysics), but also with some additional arguments. (1) Nature intends the conservation of what it preserves; it does not endow the individual with perpetual being, but only preserves the nature of the species; hence it only 60. JOHN SHARPE, Quaestio super universalia, ed. A. D. CONTI, Firenze, 1990, pp. 33, 1 - 34, 35. With regard to “multitudo individuorum est propter conservationem speciei,” Conti notes “locum non inveni”; the idea is clearly based on ARISTOTLE, De an., II, 4, 415b3-7. At line 32, one should read “unumquodque propter quod [rather than quid], et illud magis”; cf. ARISTOTLE, An. post., I, 2, 72a29-30 and Metaph., II, 1, 993b24-25 (see also Auctoritates Aristotelis, 1, 41 and 35, 29, ed. HAMESSE, p. 118, 54-57 and p. 313, note). As Conti points out, Sharpe’s argument is influenced by BURLEY, Super octo libros Phisicorum, fol. 8rb (cf. n. 55); on Burley’s influence on Sharpe, see also AMERINI, “14thCentury Reactions to Burley,” pp. 395-396 (this topic, however, is not discussed).

NATURA INTENDIT SPECIEM

363

intends the conservation of the specific nature (if it had also intended the conservation of the individual, this intention would be frustrated). (2) This is confirmed by Aristotle’s De anima: for all living beings, reproducing oneself is the most natural activity, the goal of which is the conservation of the specific nature and the assimilation to divine eternity; a specific nature that has its perpetuity only through reproduction does not participate in perpetuity through the individual; consequently, nature intends to preserve the perpetuity of the species through reproduction. (3) A real feature can belong only to a real being; perpetuity is a real feature that is proper to a species preserved through reproduction; therefore, the species must be a real being.61 Not all realists, however, shared Burley’s use of the Avicennian dictum. In his Treatise on Logic, John Wyclif calls the question as to whether nature intends singulars “a quite difficult problem.” Without analyzing this problem, he briefly notes that “it seems that nature does intend singulars and also monstrous living beings and other mistakes [...] owing to a corruption and against the primary ordering of nature.” This claim remains rather obscure, since Wyclif does not explain whether (and how) nature primarily intends individuals.62 While Burley’s use of the Avicennian dictum was apparently palatable to most realist philosophers, it was criticized in nominalist (or conceptualist) circles. In his Questions on Aristotle’s Metaphysics, John Buridan severely opposes Burley’s view. His criticism can be summarized in some four points. (1) Universals are not more perpetual than individuals. (2) God does not intend to perpetuate a common 61. PETRUS NIGRI, Clypeus Thomistarum In quoscumque aduersos [...], Venetijs, per Simonem de Luere, 1504, questio 20 (“Utrum vniuersale de quo considerant scientie reales sit ens reale”), fol. 47rb (“Tertia ratio”). In this passage, Nigri might be influenced by Vicent Ferrer. Ferrer cites Avicenna to the effect that “nature intends to produce the specific nature,” but he denies that what nature intends is something one and argues that “nothing requires [us to think] that nature should tend toward that which is common in reality owing to a commonness that arises from abstraction from individuals.” (Fidora’s translation, slightly modified). See VICENT FERRER, Tractatus sollemnis de universali, in: ID., Quaestio de unitate universalis [...]. Latin text and medieval Hebrew version with Catalan and English translations, ed. A. FIDORA – M. ZONTA, in collaboration with J. BATALLA and R. D. HUGHES, Santa Coloma de Queralt 2010, p. 283, 3-12, p. 301, 5-10, and p. 353; on Ferrer’s influence on Nigri, see FIDORA, o.c., pp. 67-75. 62. JOHN WYCLIF, Tractatus de logica, ed. M. H. DZIEWICKI, London 1893, p. 171, 1-6. On Wyclif’s realism, see A. D. CONTI, “Wyclif as an Opponent of Ockham: A Case of Realist Reaction to Ockham’s Approach to Logic, Metaphysics, and Theology,” in: RODE (ed.), A Companion to Responses to Ockham, pp. 109-139, esp. 125-126.

364

GUY GULDENTOPS

nature; the aim of his providence is only that there always be some individuals of every species. (3) The phrase “intention of a natural agent” should be correctly used and understood. Properly speaking, the words ‘intention’ and ‘appetite’ can be applied only to cognitive beings (e.g., irrational animals and human beings); with regard to such beings, these words signify acts that are distinct from the beings that have the intention or appetite. In a further step, these words are metaphorically applied to natural agents that do not have cognition; with regard to non-cognitive beings, they “do not signify acts that are distinct from the agents having the intention or appetite.” A natural agent is “metaphorically said to intend, or to strive after, an action or a form because it is determined by its nature to produce such a form in matter that is able to receive such a form.” (4) If, then, a realist opponent claims that “fire intends to produce fire,” he must mean either that it intends to produce an individual fire (which is in accord with Buridan’s view) or that it intends to produce a universal fire. In the latter case, the intention would be in vain, since it cannot produce a universal (i.e., perpetual) fire, given that what is perpetual is neither generable nor corruptible. According to Buridan, a natural agent “intends to produce every fire that can be produced by it, since it is sufficiently determined to do so if matter is appropriately offered.”63 This, however, does not mean that natural agents envisage the species as their goal. A somewhat different but equally fierceful criticism of Burley is found in Hugolinus of Orvieto’s Physics commentary. According to this Augustinian Hermit, Burley’s “modern” opinion “is evidently false or rather a sort of fancy fable” (“quedam pulchra truffa”). While accepting the major premise (“what nature principally intends is something outside the soul”), he rejects the minor (which identifies the universal with what nature principally intends), “because nature 63. JOHN BURIDAN, In Metaphysicen Aristotelis Quęstiones argutissimæ Magistri Ioannis Buridani in vltima prælectione ab ipso recognitæ, Vænundantur Badio, 1518, VI, q. 16, fol. 51vb. The idea that the word ‘intention’ is metaphorically applied to non-cognitive agents is reminiscent of Aristotle’s observation that “the final cause is productive only in a metaphorical sense” (see De gen. et corr., I, 7, 324b14-15 and Auctoritates Aristotelis, 4, 14, ed. HAMESSE, p. 168, 59; cf. PASNAU, “Teleology in the Later Middle Ages,” pp. 96 and 100). On this passage, see J. BIARD, Science et nature. La théorie buridanienne du savoir, Paris 2012, pp. 244-245; on Burley’s influence on Buridan, see AMERINI, “14thCentury Reactions to Burley,” pp. 387-393 (Amerini does not discuss this topic).

NATURA INTENDIT SPECIEM

365

intends only to produce the singular thing or singular things.” Burley’s argument would hold only “if nature intended to generate only a single individual thing,” but that is not the case: “Rather, nature intends to produce many individual things successively, even an infinite number [of them] if the world were eternal,” since it would never stop producing individuals.64 6. The Renaissance continuation of scholastic debates Burley’s view was still discussed by Italian Renaissance scholastics. In his meta-commentary on book I of Burley’s Physics commentary, the late-fifteenth-century Carmelite John of Venice makes several critical comments on Burley’s argument. In his view, Burley’s realist argument can be turned against himself: what is primarily intended by nature is also produced by nature; however, nature does not produce anything universal, since it does not produce anything eternal; consequently, nature does not produce a species. If a realist objects that a universal is produced accidentally, one may reply that if this is true, it follows that nature does not intend a species primarily. Moreover, if nature did not intend the individual, this would imply that nature’s production of an individual would be a casual effect. This is not true, since the individual is precisely what nature primarily intends and per se produces. Besides such trifling arguments against Burley, John of Venice formulates a more fundamental criticism (reminiscent of Buridan): “One should know that in a natural agent there is, properly speaking, no intention, for it does not have an appetite or a will to produce.” The word intentio here stands “for the conformity and the inclination to produce a determinate effect.” A natural agent gets such an inclination “primarily from matter, when it is close enough to the agent so that the agent can produce something else out 64. See HUGOLINUS, In libros Physicorum, I, 2, in: W. ECKERMANN, Der Physikkommentar Hugolins von Orvieto OESA. Ein Beitrag zur Erkenntnislehre des spätmittelalterlichen Augustinismus, Berlin / New York 1972, p. 53, 28-36, p. 55, 79, and p. 69, 517-524. On Hugolinus, see e.g. D. RISERBATO, “Esistenza e verità in Ugolino da Orvieto O.E.S.A. († 1373). Verum incomplexum e significabile complexe tra semantica e ontologia,” in: Documenti e studi sulla tradizione filosofica medievale 29 (2018), pp. 457-478, and S. ROUDAUT, “Hugolinus of Orvieto and the Controversies about the Perfection of Species: The Context and Influence of His De perfectione specierum,” in: Augustiniana 69 (2019), pp. 299-331; on his conceptualism, see ECKERMANN, o.c., p. 117.

366

GUY GULDENTOPS

of it, and secondarily from the universal agent, namely God, who orders everything.” This means that the so-called intention of a natural agent does not always remain the same (as Burley claimed): after producing one thing, it will intend and produce something else. This “variation in intention” depends not only on the natural agent, but also on several other causes: the matter it uses, the universal agent (i.e., God, who determines the natural agent), and “the different influences of the stars,” which change over time. John of Venice concludes that “nature always intends what it produces,” i.e., the individual. Still, he is willing to apply a hermeneutic principle of charity to “Avicenna’s authoritative argument”: nature (i.e., the nature of a natural agent) primarily intends the species in the sense that it always intends to produce what is specifically the same (fire, for instance, intends to produce fire).65 The Avicennian dictum was also discussed at some length by Alessandro Achillini, the renowned Bolognese crypto-Averroist.66 In his commentary on book I of the Physics, he refutes the view that nature intends the species. Surprisingly, he ascribes this view to the nominales and traces it back to Albert the Great. Achillini denies that “the species is better known according to nature because the species is more intended than the genus.” Rather, nature intends every corruptible individual, since the individual is what is produced and preserved by nature. In Achillini’s view, the ‘nominalist’ claim that nature intends the species must either be rejected or be taken to mean that nature does not exclusively intend one single individual, but always intends a multitude of individuals of one species. Furthermore, Achillini attacks Albert on four points. (1) Whereas Albert argues that “nature does not intend the designated individual,” Achillini holds 65. See John of Venice’s lectures on Burley’s Physics commentary, ed. CAROTI, pp. 166169 (cf. n. 55). Agostino Nifo and Matthias Aquarius hold similar views: “nature intends the individual, not distinctly or exclusively [i.e., qua individual], but rather in a confused manner”; see EUTYCHI AUGUSTINI NYPHI [...] Metaphysicarum Disputationum Dilucidarium, VII, Caput solutionum quintum, ad 10am, Impressum Neapoli per Sigismundum Mayr, 1511, fol. 275v, and MATTHIAS AQUARIUS, Dilucidationes in XII Libros Primæ Philosophiæ Aristotelis, VII, 16, Romæ, Ex Typographia Bartholomæi Bonfadini & Titi Diani, 1584, p. 415D. One of these authors may be the source of the dictum “nature intends the vague individual” in GIOVANNI BATTISTA BERNARDO’s Seminarii Totius Philosophiæ Stoicæ Tomus Tertius, In Officina I. Stœr et F. Fabri Lugdunensis, 1605, p. 51a. 66. On Achillini (1463-1512), see D. N. HASSE, Success and Suppression. Arabic Sciences and Philosophy in the Renaissance, Cambridge (Mass.) 2016, pp. 205-206 and 303.

NATURA INTENDIT SPECIEM

367

the opposite view “on the ground that the form of the designated individual ends the process of generation and is its goal, and therefore it is what is intended.” In this respect, Albert’s distinction between particular and universal nature does not help him make his case. Particular nature obviously intends the individual, and so does universal nature, “since its action ceases in this [concrete] matter, even though its action does not cease absolutely.” Universal nature, i.e., the universal celestial cause of sublunary beings, intends the concrete individuals, not according to its primary intention, but secondarily, because the heavenly bodies do not primarily act for the sake of the sublunary world. (2) Albert’s claim that “nature intends the species” is wrong, “since there are no universals on the side of reality.” The species is nothing but “the result of the agent intellect’s act of abstracting” (here Achillini transforms Albert’s formula “opus nature est opus intelligentie” into “species [...] est opus intellectus agentis abstrahentis”).67 If ‘species’ (in Albert’s thesis) is understood as the specific “nature in the individual,” it is identical with the enmattered form, which is also something individual. (3) Albert’s view that “the individual is intended accidentally” is incorrect.68 According to Aristotle, per se causes are divided into universal and individual per se causes; individual per se causes correspond to their individual per se effects. (4) Albert maintains that “it is not true that the complete being (esse) in natural things is diversified in concrete individuals in the same manner as it is in species, since in species it is diversified by the essential principles, whereas it is multiplied in individuals only by the dispositions of matter.” For Albert, all of this also shows that “according to its principal intention nature never looks at individuals.”69 Achillini defends the opposite view: “This [concrete] human being and this horse differ [from other human beings and horses] by a [similar] essential distinction whereby [the species] human being and horse differ.”70 One may doubt whether Achillini’s criticism of Albert is 67. On Albert’s dictum, see HÖDL’s article (cf. supra, n. 11). 68. On the distinction between per se and accidental causes, see ARISTOTLE, Phys., II, 5, 196b24-28. 69. Cf. ALBERT, Phys., I, 1, 2, ed. HOSSFELD, p. 4, 74-80. 70. ALESSANDRO ACHILLINI, Physicorum interpretatio (= Dubia Physicae), in: Opera Omnia in Vnum Collecta, Venetijs apud Hieronymum Scotum, 1545, fol. 69va-b. On Achillini’s theory of universals, see H. S. MATSEN, Alessandro Achillini (1463–1512) and His Doctrine of ‘Universals’ and ‘Transcendentals’. A Study in Renaissance Ockhamism,

368

GUY GULDENTOPS

really fair, since Albert argues that universal nature intends the vague individual while particular nature intends the concrete individual. Still, the main divergence between these two philosophers concerns the dictum “nature intends the species.” While Albert defends the primacy of the species, Achillini expounds an individualist interpretation of the Avicennian topos that is comparable not only to the views of conceptualists such as Buridan and Hugolinus but also to those of Scotus and some Scotists. The extramental reality of universals is a topic of controversy, not only in the Physcis tradition, but also in commentaries on De anima. An interesting example is found in Arcangelo Mercenario’s commentary on the first book. One of the arguments for a Peripatetic realism (a position with which he disagrees) relies on the thesis that “nature intends the species before intending the individual.” While leaving Avicenna unmentioned, he refers for this thesis to the often cited parallel passage from book II of De anima: since individuals cannot be perpetuated, the species must be perpetuated; consequently nature produces the individuals for the sake of the species. Thus, it follows that nature primarily has the species as its goal. If nature first intends the species, it also first produces the species before producing the individual. Since everything that is produced exists outside the soul, natural species must have extramental existence.71 Mercenario, however, rejects this view and instead defends a conceptualist view according to which a universal exists only in the soul as “the common nature Lewisburg / London 1974, pp. 42-65, esp. 51-54 (on Achillini’s criticism of Albert; Matsen does not refer to the Avicennian dictum in this context). 71. ARCANGELO MERCENARIO, In Primum De anima, ed. L. BRIGUGLIO – E. SCAPIN, Venezia / Roma 1961, p. 41, primum fundamentum: “Natura prius intendit speciem quam individuum; ergo et prius producit speciem. Probatur antecedens: Aristoteles, secundo De anima, [...] dicit: ‘ut species rerum conservetur’. [...] Probatur consequentia: si prius [producit] intendit speciem, ergo prius etiam producit [intendit] speciem quam singulare, quam individuum. Productum autem omne habet esse extra animam; ergo etiam et species erunt extra animam.” (I have deleted the verbs “producit” and “intendit,” which seem completely superfluous and incorrect; “quam singulare, quam individuum” is a doubtful double reading.) On Mercenario (d. 1585), see the introduction to the edition, pp. IX-XI (his commentary on book I of De anima is based on his Paduan lectures of 1578-1579); on his Averroism and his influence on Leibniz, see A. BLANK, “Renaissance Aristotelianism and the Conciliatory Approach to Individuation in the Early Leibniz,” in: J.-A. NICOLÁS – N. ÖFFENBERGER (eds.), Beiträge zu Leibniz’ Rezeption der Aristotelischen Logik und Metaphysik, Hildesheim 2016, pp. 257-272, esp. 260-269.

NATURA INTENDIT SPECIEM

369

that is known and apprehended by the intellect without matter and without material characteristics.” In his reply to the abovementioned argument, he asserts that the Avicennian principle (which he again ascribes to Aristotle) is true. Indeed, nature intends to preserve the species, but it can perpetuate the species “only by means of a continuous succession of individuals,” and “therefore nature secondarily intends individuals.” However, what Mercenario denies is the inference: “If nature primarily intends the species, it primarily produces it.” This implication is invalid, since nature does not primarily intend to generate the species. Rather, nature primarily intends to generate that which preserves the species, and as this goal cannot be reached without individuals, nature intends to generate individuals before intending to generate the species.72 Unlike other conceptualists, Mercenario accepts the Avicennian topos, which he combines with the idea that nature also intends individuals. The theory of universals is not the only context in which Renaissance philosophers refer to the Avicennian dictum. A brief allusion to it is found, for instance, in Pietro Pomponazzi’s Apologia, the selfdefense he wrote after several of his philosophical positions, especially his mortalism, had come under attack. In his On the Immortality of the Soul, Pomponazzi refutes the belief that dreams can be considered a proof of the soul’s indestructibility. At best, they suggest that “the gods take care of inferior beings.”73 In his criticism of Pomponazzi’s arguments, Pietro Manna argues that the Aristotelian view that “dreams occur owing to a divine power and owing to God’s care for inferior beings” is contradicted by Averroes, who in his Metaphysics commentary claims that “God does not care for these [i.e., terrestrial] 72. MERCENARIO, o.c., p. 54: “Huic rationi respondeo: concedimus omnes quod natura primo intendat speciem [...] Nam individua non perpetuantur; sed quia natura non potest speciem conservare perpetuam nisi per continuam successionem individuorum, igitur secundario natura intendit individua. Sed consequentiam [conieci: contrariam ed.] nego: si natura prius intendit speciem, ergo prius eam producit. Sane enim hoc intelligendum est: natura non prius intendit generare speciem, sed conservans tamquam finem generationis individuorum. Sed quia conservatio et finis naturae non potest esse sine individuis, ideo intendit generare individua; intendit itaque generare individua prius quam speciem.” 73. PIETRO POMPONAZZI, Tractatus de immortalitate animae, 13-14, in: ID., Abhandlung über die Unsterblichkeit der Seele, ed. B. MOJSISCH, Hamburg 1990, pp. 158-160 and 214. On Pomponazzi (1462-1525), see HASSE, Success and Suppression, pp. 214-217 and 649.

370

GUY GULDENTOPS

things in their individuality.”74 Rebutting this objection, Pomponazzi argues that there is no contradiction at all. The fact that God and the celestial intelligences care about the sublunary world does not imply that “they know these things in their individuality.” In this context, he cites the Avicennian dictum (without mentioning Avicenna by name): “What is known in the universal is generated in the particular, just as nature intends the species and nevertheless produces only the individual.” Moreover, he paraphrases Albert’s criticism of Avicenna’s claim that the heavens have sensory souls: “Although a celestial intelligence does not know [what exists] here and now except in the universal, this knowledge suffices to produce the heavenly motion because it is individualized by the heavenly body, which exists here and now.”75 Pomponazzi’s parenthetical use of the Avicennian dictum is fairly significant: it testifies that in the early sixteenth century an acute and highly independent mind could avail himself of the topos in order to defend an Averroist (or Avicennian) view on divine providence, while at the same time criticizing a distinctive aspect of Avicenna’s cosmology. Crisostomo Javelli, a famous early-sixteenth-century Thomist and one of Pomponazzi’s opponents,76 discusses the Avicennian topos 74. POMPONAZZI, Apologia, I, 4, in: PETRI POMPONATII MANTUANI Tractatus acutissimi, vtilissimi, et mere peripatetici [...], Venetijs impressum arte et sumptibus hęredum [...] Octauiani Scoti, 1525, fol. 69ra (fourth doubt). On Manna, one of Pomponazzi’s former students, see B. NARDI, Studi su Pietro Pomponazzi, Firenze 1965, pp. 49 and 181. Manna refers to AVERROES, De divinatione per somnum, 1, in: Aristotelis Libri Reliqui ad Animalium cognitionem attinentes [...] cum Averrois [....] Paraphrasibus, Venetiis, Apud Iunctas, 1562, fols. 30vF-G, 34rA, and 34vI (based on ARISTOTLE, De divinatione per somnum, 2, 463b1215), and AVERROES, In Metaph., XII, 37 and 52, fols. 320vI-K and 338rE-F. 75. POMPONAZZI, Apologia, I, 4, fol. 69rb (“Ad 4m”). Cf. AVICENNA, Liber de philosophia prima, IX, 2, ed. VAN RIET, p. 454, 86-01; ALBERT, Metaphysica, XI, 3, 4, ed. B. GEYER, Münster i. W. 1964, pp. 538-539 (esp. p. 539, 5-22). Pomponazzi also refers to THOMAS AQUINAS, Qu. disp. de anima, 8, ad 17, ed. BAZÁN, pp. 73, 489 - 74, 513, but there Aquinas actually holds that “the higher intellectual substances can know particulars without any sensory power.” On this passage from the Apologia, see V. PERRONE COMPAGNI, “Evidentissimi avvertimenti dei numi. Sogni, vaticini, profezie in Pomponazzi,” in: Annali del Dipartimento di Filosofia (Nuova Serie) 17 (2011), pp. 21-59, esp. 33; EAD., “Métamorphoses animales et génération spontanée. Développements matérialistes après le De immortalitate animae,” in: J. BIARD – Th. GONTIER (eds.), Pietro Pomponazzi entre traditions et innovations, Amsterdam / Philadelphia 2009, pp. 65-81, esp. 66-67 (she does not refer to Avicenna). 76. On this Dominican philosopher and theologian (d. ca. 1538), see L. SPRUIT, “Javelli, Giovanni Crisostomo,” in: M. SGARBI (ed.), Encyclopedia of Renaissance Philosophy, Cham 2016: https://doi.org/10.1007/978-3-319-02848-4_734-1 (accessed in March 2022).

NATURA INTENDIT SPECIEM

371

more extensively. In his commentary on Aristotle’s Physics, he distinguishes four meanings of the phrase ‘more [known] and first known according to nature’. This phrase can be predicated of (1) things “that have a more perfect beingness” (e.g., divine, immaterial substances);77 (2) things “that are intended by nature primarily and to a higher degree”; (3) things “that can be known without anything else” (e.g., substance, which can be known without accidents, whereas the knowledge of accidents presupposes the knowledge of substance);78 and (4) “the causes of natural things” (the causes do not depend on their effects, therefore they are first known according to nature). 79 The second sense of ‘first known according to nature’ applies to “the most specific species.” By contrast, individuals are “secondarily known [according to nature], because [nature] intends the individuals only for the conservation of the species.” This instrumentality of individuals also implies that an incorruptible form that can be preserved in one single individual is not pluralized in several individuals (each angel, for instance, is unique in his species).80 In his commentary on the Metaphysics, Javelli clarifies in what sense nature intends the species. Commenting on Metaphysics, I, 1 (981a16-17), he discusses the question as to “whether every activity is related to an individual,”81 and he argues that “transitive actions, which are the Philosopher’s main focus here, are per se concerned with something individual.” Transitive actions are concerned with something universal “only through the mediation of the individual in which the universal nature 77. For this sense of priority, cf. ARISTOTLE, Cat., 12, 14b4-5. Here, Javelli refers to Metaph., II, 1, 993b9-11, where Aristotle compares our intellect to the eyes of the night-owl (or the bat): just as the night-owl is too weak to see the sunlight, so we have difficulties in knowing the immaterial substances. On this simile, see C. STEEL, Der Adler und die Nachteule. Thomas und Albert über die Möglichkeit der Metaphysik, Münster i. W. 2001. 78. For this meaning of priority, cf. ARISTOTLE, Metaph., V, 11, 1019a2-4. 79. For this meaning of priority, cf. ARISTOTLE, Cat., 12, 14b11-13. 80. DN. CHRYSOSTOMI IAVELLI CANAPICII [...] Commentarii in VIII libros Physicorum Aristotelis [...], Lugduni, Apud S. B. Honorati, 1555, pp. 12-13. 81. On the medieval scholastic background of this question, see A. DE LIBERA, “Les actions appartiennent aux sujets. Petite archéologie d’un principe leibnizien,” in: CAROTI – IMBACH – KALUZA – STABILE – STURLESE (eds.), « Ad ingenii acuitionem ». Studies in Honour of Alfonso Maierù, pp. 199-219; R. CROSS, “Accidents, Substantial Forms, and Causal Powers in the Late Thirteenth Century: Some Reflections on the Axiom ‘actiones sunt suppositorum’,” in: C. ERISMANN – A. SCHNIEWIND (eds.), Compléments de substance. Études sur les propriétés accidentelles offertes à Alain de Libera, Paris 2008, pp. 133-146.

372

GUY GULDENTOPS

has its being.” Since, then, a transitive action is an individual action (for instance, “this [particular act of] healing”), both the subject which performs the activity and the subject in which it is received must be concrete individual beings. Obviously, ‘concerned with’ (“circa,” the Latin translation of περὶ) is to be understood, “not merely objectively,” but rather “subjectively” (i.e., as signifying the subject of the activity): “for even though, objectively speaking, thinking is concerned with the universal, it [i.e., the concrete act of thinking] inheres in the intellect subjectively, inasmuch as the intellect is individual.”82 Having expounded his view on the individuality of transitive actions, Javelli disproves a series of arguments that seem to contradict it. One of these arguments is based on Avicenna’s claim that “nature intends to produce the species.” Since ‘species’ denotes a universal nature, it follows that “the production ends in a universal.” In his reply to this argument, Javelli notes that if Avicenna meant that nature intends to produce a species that exists by itself, the argument would be valid. However, on Javelli’s reading, such a ‘realist’ or ‘Platonic’ interpretation of Avicenna is incorrect. Avicenna rather means that nature intends to produce a species “through the mediation of an individual in such a manner that the individual is produced only in order to preserve the species.” Otherwise put, nature’s “reason for producing an individual is not the individual itself, but rather the specific nature in the individual.” Still, although the species is what is primarily intended in natural production, “it is not necessarily that which is primarily produced.” The backdrop to this entire question is John of Jandun’s claim that the universal is something which can primarily and per se be generated and destroyed.83 According to Javelli, Jandun was led to this view because he did not grasp that something 82. CHRISOSTOMI IAVELLI CANAPICII Quæstiones super libros Metaphysicæ Aristotelis iuxta Thomisticæ doctrinæ dogmata decisæ [...], I, 7, Venetiis apud H. Scotum, 1555, fols. 18r-20r, esp. 18v. For the idea that motions and actions inhere in something, Javelli refers to ARISTOTLE, Phys., III, 1, 200b32-33 and 3, 202a13-14. In his commentary on Metaph., IX, 8, 1050a21-b2 (= o.c., IX, 15, fols. 274v-275r), he explains that ‘transitive action’ can be taken to mean either “the real relation which an agent has with regard to that which undergoes its action” or “the thing or form caused by the agent and acquired by that which undergoes its action”; heating, for instance, is the heat insofar as the heat (i.e., the form of hotness) is caused by the heating agent (e.g., the fire) in something which is being heated (e.g., the water). 83. IOANNIS DE IANDUNO [...] Acutissimæ Quæstiones in duodecim Libros Metaphysicæ [...], I, 16, Venetijs, Apud Hieronymum Scottum, 1560, cols. 53-62.

NATURA INTENDIT SPECIEM

373

can be intended per se without being generated per se (and vice versa). Jandun mistakenly believed that everything which is intended per se is also generated per se. In reality, however, the species is what nature intends per se when generating anything, and yet nature generates the species, not per se, but rather by means of the individuals that are generated.84 A somewhat similar view is propounded by the Coimbra Jesuits, who succinctly discuss the Avicennian dictum in their commentary on book II of De generatione et corruptione. There, it is used as one of the arguments for the thesis that the generation of individuals is not per se intended by nature. Rather, nature “is per se directed to the conservation of the species, not to the multitude of corruptible individuals; for this would end in an infinite regress, which nature abhors, since it always seeks the finite.” According to the Jesuit commentators, this argument is inconclusive. All creatures have “a forceful inborn desire to perpetuate themselves.” Corruptible beings, however, “can attain perpetuity only by means of reproduction,” and consequently, “they per se intend procreation most forcefully.” Like Thomas Aquinas, the Jesuit commentators maintain that “the good of the species is more universal and therefore more desirable” than the good of the individuals. Accordingly, “by itself and owing to its primary intention, nature is directed toward the species.” In this sense, Avicenna is right. Still, nature can conserve the species only through the individuals, and therefore “it intends the multitude of individuals by itself, though not owing to its primary intention, but rather owing to its secondary, less principal intention.” Nature does not abhor the succession of individuals, which can go on infinitely. What is impossible is the actual infinity in any class and the infinite multiplication of species.85 84. JAVELLI, Quæstiones super libros Metaphysicæ, fol. 20r. 85. Commentarii Collegii Conimbricensis Societatis Iesu, In Duos Libros de Generatione & Corruptione Aristotelis Stagiritæ [...], I, 4, q. 16, art. 4, Conimbricæ, Ex Officina Antonij à Mariz [...] 1597, pp. 121-122 = Commentarii Collegii Conimbricensis, e Societate Iesu, In Duos Libros de Generatione & Corruptione Aristotelis Stagiritæ [...], Venetiis, Apud Andream Baba, 1616, cols. 179E-180E. For the idea that “nature abhors the infinite,” the commentators refer to ARISTOTLE, De gen. an., I, 1, 715b14-15 (cf. Auctoritates Aristotelis, 9, 175, ed. HAMESSE, p. 223, 97). For Avicenna’s dictum, they also refer to THOMAS AQUINAS, Summa contra gentiles, II, 89, ed. Leon. 13, Romae 1918, p. 541b49-50 (= transl. by J. F. ANDERSON, Notre Dame 1975, pp. 302-303; translation modified): “the function

374

GUY GULDENTOPS

Only a couple of years before the Coimbra commentary on De generatione et corruptione was published, the young Galileo, who was influenced by the Aristotelianism of the Jesuit Collegio Romano, used the Avicennian dictum in one of his early methodological treatises. In the questions on the previous knowledge that is required in every science (questions related to Posterior Analytics, I, 1, 71a11-17), he argues that on the whole the existence of the subject-matter of a ‘real science’ (in particular of the natural sciences and mathematics) must be known before one starts producing strict demonstrations in that science. In this context, he tackles the objection that only individuals exist, whereas science is not concerned with individuals. In his reply to this objection, Galileo points out that “the existence that accompanies a universal nature, not in the abstract, but indeterminately in some individual, is proper to the species.” In order to strengthen this claim, which is tied up to scholastic theories about the vague individual, he notes that “the existence of the species is more intended by nature than is that of individuals, because nature is perfected more by species than it is by individuals.”86 Galileo’s paraphrasing of the Avicennian topos comes close to its interpretation by late-medieval realists. 7. Piccolomini’s reconstruction of the Aristotelian priority debate: from Alexander over Avicenna to Scotus The last author to be presented in this paper is Francesco Piccolomini, a contemporary of Galileo and one of the most erudite and influential Aristotle commentators of the late sixteenth century.87 In the of the generative power is directed, not toward the perfection of the individual, but rather toward the preservation of the species.” On these Jesuit commentators, see C. S. MARINHEIRO, “Conimbricense, Collegium,” in: SGARBI (ed.), Encyclopedia of Renaissance Philosophy: https://link.springer.com/referenceworkentry/10.1007/978-3-319-02848-4_316-1 (accessed in March 2022). 86. GALILEO GALILEI, Tractatio de praecognitionibus et praecognitis, disp. 3, q. 1, cited in: W. A. WALLACE, Galileo and His Sources. The Heritage of the Collegio Romano in Galileo’s Science, Princeton 1984, p. 39 (col. b, ll. 133-144) = W. A. WALLACE, Galileo’s Logical Treatises. A Translation, with Notes and Commentary, of His Appropriated Latin Questions on Aristotle’s Posterior Analytics, Dordrecht 1992, p. 94, § 14 (translation slightly modified). According to Wallace, the Tractatio de praecognitionibus et praecognitis dates from 1589 (see Galileo’s Logical Treatises, p. 39). 87. On Piccolomini, see e.g. D. A. LINES, “Piccolomini, Francesco,” in: SGARBI (ed.), Encyclopedia of Renaissance Philosophy: https://link.springer.com/referenceworkentry/10.

NATURA INTENDIT SPECIEM

375

final dubitatio concerning the prologue of the Physics, Piccolomini discusses “whether individuals must be the first or rather the last according to nature.” The horizon against which to read this text is constituted by Renaissance interpretations of the phrase ‘first known according to nature’ (as found, for instance, in Javelli’s commentary on Physics, I, 1) and, more broadly, by Renaissance discussions of scientific method, which also draw on the first chapter of the Physics.88 However, Piccolomini’s main interlocutors in this question are not contemporaneous authors, but rather some outstanding ancient and medieval authorities. On his reading, the aporia concerning what is first according to nature (whether it is the individual or the universal) goes back to Alexander of Aphrodisias’s lost Physics commentary (known through fragmentary citations in Simplicius and Averroes). According to Piccolomini, there is something to be said on both sides of the question. On the one hand, individuals are immediately accessible to sense-perception and so seem to be “the first according to us, but the last according to nature, especially since universals are the first according to nature and the farthest removed from the senses.” This view (based on Posterior Analytics, I, 2) is corroborated by the observation that the individual, unlike the universal, is perishable and subject to change; therefore the individual does not seem to be the first according to nature.89 On the other hand, the individual does appear to be the first according to nature. Piccolomini adduces three arguments for this view. (1) As Aristotle notes in his De anima, “the 1007/978-3-319-02848-4_354-1 (accessed in March 2022); and my “Francesco Piccolomini’s Christian-Neoplatonic Reading of Aristotle’s Theory of Friendship,” in: American Catholic Philosophical Quarterly 93 (2019), pp. 551-576; on his physics, see G. CLAESSENS, “Francesco Piccolomini on Prime Matter and Extension,” in: Vivarium 50 (2012), pp. 225244. 88. See e.g. JACOPO ZABARELLA, De methodis, I, 5-6, ed. J. P. MCCASKEY, Cambridge (Mass.) 2013, pp. 26-28; cf. N. JARDINE, “Keeping Order in the School of Padua: Jacopo Zabarella and Francesco Piccolomini on the Offices of Philosophy,” in: D. A. DI LISCIA – E. KESSLER – C. METHUEN (eds.), Method and Order in Renaissance Philosophy of Nature. The Aristotle Commentary Tradition, Aldershot 1997, pp. 183-209; Th. LEINKAUF, Grundriss. Philosophie des Humanismus und der Renaissance (1350–1600), Hamburg 2017, pp. 186-211, esp. 199-203. 89. FRANCISCI PICCOLOMINEI SENENSIS [...] Librorum ad Scientiam de Natura attinentium Pars Prima [...], Venetiis, Apud Hæredes F. de Franciscis, 1600, fol. 60r. Piccolomini refers to ALEXANDER APUD SIMPLICIUM, In Aristotelis Physicorum libros quattuor priores commentaria, I, 1, ed. H. DIELS, Berolini 1882, p. 19, 5-11; he also invokes ARISTOTLE, An. post., I, 2, 72a1-5 and Phys., I, 5, 189a1-8.

376

GUY GULDENTOPS

universal is either nothing or posterior”; (2) in the category of substance, individuals are called “primary substances”; (3) “that after which nature primarily strives is the individual taken indefinitely.” Piccolomini correctly ascribes this third argument to Avicenna and Scotus, and adds that “without the mind’s activity, nothing is to be found in the universe except the singular, the only thing that by itself exists, generates, and is generated.”90 Since Aristotle (in Physics, I, 1) recommends a method that “advances from the universal to the individual,” this must mean that “the individuals are prior and better known according to nature.” Hence it seems likely “to conclude that the individuals are the first both according to us and according to nature.”91 Piccolomini, however, is keen to stress that this is “an utmost thorny issue, on which the Greek commentators express various views.” He disagrees with Philoponus’ view that “the determinate and distinct individuals are prior and better known according to nature,” not only because the universals are called “first according to nature” (in Posterior Analytics, I, 2 and Metaphysics, I, 1), but also because the determinate individual is the end of the epigenetic development of an embryo (as explained in On the Generation of Animals, II, 3).92 Likewise, Piccolomini is discontented with Alexander’s view that “the universal is an accident and therefore posterior to all individuals taken together, [...] while being prior to each single individual taken separately.” This view is doubtful, since it does not consider “the absolute nature of the common thing” (e.g., the “nature of animal” as such), but rather concentrates on the common features that are abstracted 90. PICCOLOMINI, o.c., fol. 60r. Cf. ARISTOTLE, De an., I, 1, 402b7-8; Cat., 5, 2a1114. It is unclear with which texts of Scotus Piccolomini was acquainted; perhaps he had only some second-hand knowledge of Scotus’s philosophy (there were chairs for Scotist theology and metaphysics at the University of Padua: see JARDINE, “Keeping Order in the School of Padua,” pp. 189-190). 91. PICCOLOMINI, o.c., fol. 60r. Cf. ARISTOTLE, Phys., I, 1, 184a16-25. 92. PICCOLOMINI, o.c., fol. 60r-v. Cf. PHILOPONUS, In Aristotelis Physicorum libros tres priores commentaria, ed. H. VITELLI, Berolini 1887, p. 18, 5-26; ARISTOTLE, An. post., I, 2, 72a1-5; Metaph., I, 1, 982a24-26 and 982b2; De gen. an., II, 3, 736b3-5 (on this passage, see B. HINTON, “Generation and the Unity of Form in Aristotle,” in: Apeiron 39 [2006], pp. 359-380, esp. 368, and D. HENRY, “Aristotle on Epigenesis: Two Senses of Epigenesis,” in: A. FALCON – D. LEFEBVRE [eds.], Aristotle’s Generation of Animals. A Critical Guide, Cambridge 2018, pp. 89-107, esp. 89 and 94-95).

NATURA INTENDIT SPECIEM

377

by the mind and attributed to the individuals sharing them.93 Piccolomini is even more critical of Simplicius’ comments. According to Piccolomini’s somewhat imprecise paraphrase, Simplicius holds that “since the common [universals] are faint impressions, abstracted by the mind from the individuals, they are posterior by nature, though prior according to us, and they are prior according to us in respect of our confused knowledge, for we obtain our distinct knowledge later.” Piccolomini criticizes Simplicius for several reasons. His observation that the universal is a faint notion is irrelevant, since the question is about “the absolute nature of the thing, not about the notion formed by the mind.” (If we see Socrates from a distance, we do not believe that what is arriving is a notion.) Furthermore, such faint notions are not prior to the individuals, given that we form them on the basis of our knowledge of the individuals. Finally, our distinct knowledge of what is more universal precedes our knowledge of what is less universal (as Aristotle argues in Topics, VI, 4).94 Having left the Greek commentators, Piccolomini turns his attention to Averroes. Following Alexander, the Arabic commentator claims that “the individuals are prior in being, whereas the universals are prior according to the mind.” Although this is true, it again does not answer the question whether the individual or the universal is the first by nature.95 Dissatisfied with the commentary tradition, which offers no adequate solution to this problem, Piccolomini formulates his own answer in an attempt to disentangle the Gordian knot. First, he distinguishes between two kinds of individuals: immaterial individuals (which cannot be perceived by the senses) and material individuals (which are perceived by the senses). In immaterial individuals, the purely intelligible nature of the species is not distinct from the equally intelligible nature of the individual (Piccolomini is obviously thinking of angels, each of which is a single species). In material individuals, however, the nature of the species differs from the nature of the individual. The 93. PICCOLOMINI, o.c., fol. 60v. Cf. ALEXANDER OF APHRODISIAS, Quaestiones, I, 11, ed. I. BRUNS, Berolini 1892, p. 21, 21-24, p. 22, 4-17, and p. 24, 5-19. 94. PICCOLOMINI, o.c., fol. 60v. Cf. SIMPLICIUS, In Phys., I, 1, pp. 17, 38 - 18, 23, esp. 18, 20-22; ARISTOTLE, Top., VI, 4, 141a26-33 and 141b3-10. 95. PICCOLOMINI, o.c., fol. 60v. Cf. AVERROES, In Metaph., XII, 27, fol. 311vG-H (Averroes refers to Alexander).

378

GUY GULDENTOPS

individuals are prior according to us, since we produce the universals through abstraction from the individuals we perceive by the senses. Still, it is unclear whether the individuals are prior according to nature. Here, a distinction must be made between the order of origination or generation and the order of finality or perfection.96 (1) As said above, the individuals are not prior in the order of origination (according to this order, what is universal or more common precedes the individuals). (2) The debate concerns the order of finality. The ‘Academics’ (i.e., Platonists and perhaps also Burleian realists) assert that “the first according to nature’s design is the species, which is eternal, excludes accidental properties, and corresponds to the idea.” Avicenna and Scotus, however, hold that “nature primarily desires the individual, taken indefinitely, which by itself exists and generates a similar [being].” This latter view relies on the Aristotelian observation that the individual is the last that is generated and the first that is intended.97 According to Piccolomini, the conflict between the ‘Platonic’ and the ‘Avicennian’ or ‘Scotist’ positions can be solved if we recall Alexander’s theory of definition. On Alexander’s view, a definition pinpoints on the one hand “universals which exist in individuals” and on the other “individuals in respect of what is common to them.” In this way, Alexander indicates that “the nature of the species is identical to the nature of the individuals, taken indefinitely, and is only notionally distinct from it” (a human being indeterminately includes the aspects of the individual which are not included in the nature of the species ‘human being’).98 Hence Piccolomini arrives at a balanced conclusion in which he foregrounds the pedigree of his own solution to the aporia: Avicenna and Scotus apparently formulate a quite correct view, since nature primarily aims at the human essence joined together with the individual features with which [individuals] exist and are active, and these [i.e., the individuals] are last in the order of generation. Finally, 96. On these two orders, see ARISTOTLE, Metaph., IX, 8, 1050a2-10 and THOMAS AQUINAS, Summa theologiae, I, 85, 3, ad 1 and I-II, 62, 4, resp.; cf. W. GORIS, Absolute Beginners. Der mittelalterliche Beitrag zu einem Ausgang vom Unbedingten, Leiden / Boston 2007, p. 32, n. 50. 97. Cf. supra, n. 19 and 92. 98. PICCOLOMINI, o.c., fol. 60v. Cf. ALEXANDER OF APHRODISIAS, Quaestiones, I, 3, ed. BRUNS, esp. p. 7, 27-28; on this text, see RASHED, Essentialisme, pp. 257-260.

NATURA INTENDIT SPECIEM

379

if [...] one compares the individuals taken indefinitely with the definite and distinct ones, they [i.e., the indefinite individuals] are the first according to nature in the order of design, since nature first desires them, as I said; for us, however, before the multitude [of individuals] appears, some definite individual is better known and through it we make up our judgment about all other individuals.99

Piccolomini’s syncretist reconstruction of the debate on what is first according to nature is remarkable, not only because here he deviates from the Neoplatonic commentators whom he follows in many other contexts, but also because he presents Alexander as the initiator of the debate and as the intermediary who bridges the gap between the ‘Platonic’ and the ‘Avicennian-Scotist’ viewpoints.100 8. Conclusion The foregoing survey of the late-medieval and Renaissance reception of the Avicennian topos is undoubtedly incomplete. Still, it makes clear that the dictum “nature intends the species” was cited in various contexts (not only in natural philosophy and metaphysics, but also in logic and theology) and interpreted in divergent ways. These manifold interpretations can be represented in the following table101:

99. PICCOLOMINI, o.c., fols. 60v-61r: “Et propterea rectiùs uidentur loqui Auicenna et Scotus, quia natura primo expetit humanam essentiam cum singularibus conditionibus iunctam, cum quibus subsistunt operantur, et hæc sunt generatione postrema. Si demùm [...] singularia indefinite sumpta cum definitis et distinctis comparentur, ea sunt naturæ prima secundum institutum, cum natura primò ea expetat [conieci: expectat ed.], vt dixi; nobis verò, antequam apparuerit multitudo, definitum aliquod singulare notius est, et per id iudicantur cætera [...].” 100. On Alexander’s influence on Avicenna, see A. BERTOLACCI, “‘The Excellent among the Earlier Scholars’: Alexander of Aphrodisias in Avicenna’s Metaphysics,” in: P. B. ROSSI – M. DI GIOVANNI – A. ROBIGLIO (eds.), Alexander of Aphrodisias in the Middle Ages and the Renaissance, Turnhout 2021, pp. 33-58; on Alexander as one of the most crucial sources of Avicenna’s theory of the universal and its relation to individuals, see A. DE LIBERA, L’Art des généralités. Théories de l’abstraction, Paris 1999, pp. 499-529 and 559-571; on the genealogy Alexander – Avicenna – Scotus, see ID., o.c., p. 607, and T. BATES, Duns Scotus and the Problem of Universals, London / New York 2010, pp. 59-60 and 151 (n. 2). 101. In this table, the question mark after an author’s name indicates that it is uncertain whether he explicitly held the view I ascribe to him.

380

GUY GULDENTOPS

Nature Intends

Universal Nature Particular Nature Intends Intends

Species

Albert Aquinas Quidort Duns Scotus Burley Baconthorpe Francis of Prato Sharpe Tostatus Nigri John of Venice Pomponazzi Javelli Mercenario Galileo Conimbricenses Piccolomini

Peter of Spain Albert Roger Bacon Quidort

Individuals

William of La Mare (?) Duns Scotus Bassoles Peter of Poznań Buridan Wyclif Hugolinus John of Venice Achillini Mercenario Conimbricenses

Roger Bacon Achillini

Vague Individuals

Quidort Duns Scotus Buridan (?) Baconthorpe Nifo Aquarius Piccolomini Bernardo

Albert Quidort

Designated Individuals

Duns Scotus (?) Baconthorpe Bassoles (?) Peter of Poznań (?)

Achillini

Peter of Spain

Albert Quidort Achillini

NATURA INTENDIT SPECIEM

381

As this schematic classification of the different positions shows, the idea that nature intends the species is regularly combined with the idea that nature intends the vague individual and sometimes with the claim that particular nature aims at the individual. However, the idea that nature (or universal nature) is oriented toward the concrete individual is not common at all, and the most important medieval proponents of this view (Roger Bacon, Duns Scotus, and Baconthorpe) do not deny that nature intends the species. Furthermore, it is noteworthy that the idea that nature intends the individual (or the vague individual), though initially defended in theological contexts, becomes a philosophical commonplace propounded by late-medieval nominalists against realist theories of universals. In Renaissance Aristotelianism, authors as various as Achillini (who polemicizes against Albert the Great’s focus on the species) and Piccolomini (whose reading of Avicenna is influenced both by Alexander’s anti-Platonic view on individuals as the object of definitions and by Scotus’s anti-Thomist view on individuals as the object of nature’s intention) indirectly echo Philoponus’ proto-individualist observation that “nature has as its goal to produce individuals.”102 However, that this view has never become predominant in the Renaissance is evidenced not only by Pomponazzi, Javelli, and Galileo, but above all by the Coimbra Jesuits, who in line with Aquinas emphasize that individuals result only from the secondary intention of nature. 102. PHILOPONUS, In Phys., I, 1, ed. VITELLI, pp. 13, 17 - 14, 5, esp. 13, 30 (= On Aristotle. Physics 1.1-3, transl. by C. OSBORNE, London 2006, p. 35; translation modified). In this sentence, ‘goal’ renders σκοπός, whose Latin equivalent is intentio: see Physicorum, hoc est de Naturali Auscultatione Primi Quatuor Aristotelis Libri, cum Ioannis Grammatici cognomento PHILOPONI eruditissimis commentariis, [...] translatis, GUILELMO DOROTHEO Veneto Theologo Interprete, Venetijs, Apud Octauianum Scottum, 1554, fol. 4ra (ll. 56-57) = Aristotelis Physicorum Libri Quatuor, cum Ioannis Grammatici cognomento PHILOPONI Commentarijs, quos [...] restituit IOANNES BAPTISTA RASSARIUS, Venetiis, Apud Hæredem Hieronymi Scoti, 1581, p. 5b (ll. 5-6): “[...] ipsam naturam intentionem habere producendi singularia [...]” (on this meaning of σκοπός, cf. H. G. LIDDELL – R. SCOTT – H. S. JONES – R. MCKENZIE, A Greek-English Lexicon, Oxford 1996, s.v., II.2, p. 1614, and H. HOLTKAMP, Zum Begriff „intentio“ in der scholastischen Philosophie. Eine bedeutungsgeschichtliche Untersuchung, Bonn 1926, pp. 28-29). On Philoponus’ view, see FRITSCHE, Methode und Beweisziel, pp. 326-327, n. 376. On Dorotheus, a humanist Augustinian, see F. BOSSIER, Simplicius. Commentaire sur le Traité du Ciel d’Aristote. Traduction de Guillaume de Moerbeke, Vol. I, Leuven 2004, pp. CXX-CXXII; on Rasario, a Paduan Grecist and historian of ancient philosophy, see M. T. GIRARDI, Il sapere e le lettere in Bernardino Tomitano, Milano 1995, p. 187.

382

GUY GULDENTOPS

Even in the eighteenth and nineteenth centuries, the idea that “nature (only) intends the species” is still current in the writings of (neo-)scholastic conservatives.103 Unsurprisingly, the traditional topos also spreads outside the boundaries of scholasticism and enters the modern republic of letters. Let us cite just one witty example, taken from Voltaire’s anti-physiocratic novel L’Homme aux quarante écus, where a geometrician tries to console the poor protagonist with the following naturalistic (or perhaps even cynical) argument: La nature se soucie fort peu des individus. Il y a d’autres insectes qui ne vivent qu’un jour, mais dont l’espèce dure à jamais. La nature est comme ces grands princes qui comptent pour rien la perte de quatre cents mille hommes, pourvu qu’ils viennent à bout de leurs augustes desseins.104

103. See e.g. BERNARDUS OBERHAUSER, Coronis Biennii Philosophici, PeripateticoThomistici [...] sive Physica Particularis [...], Salisburgi, Typis Joan. J. Mayr, 1729, fol. (f)2r, § 179: “Natura solùm intendit speciem, et non individuum” (Oberhauser actually incorporates Pseudo-Bede’s Axiomata philosophica in his Coronis; like Pseudo-Bede, he interprets natura as “common nature, e.g., the heavens”); JOANNES HIDALGO ASTIGIENSIS, Cursus Philosophicus ad Mentem B. Ægidii Col. Rom., Tomus I, Cordubæ, per Petrum Arias, 1736, p. 325, §§ 213-214; NUNTIO SIGNORIELLO, Lexicon Peripateticum PhilosophicoTheologicum [...], Altera editio, Neapoli, Apud Officinam Bibliothecae Catholicae Scriptorum, 1872, p. 187 (he briefly discusses “Intentio naturae est ad speciem” among the so-called effata, i.e., the axioms of Peripatetic philosophy). On Oberhauser (1694-1739), a Benedictine Thomist at the University of Salzburg, see B. JANSEN, “Quellenbeiträge zur Philosophie im Benediktinerorden des 16./17. Jahrhunderts,” in: Zeitschrift für katholische Theologie 60 (1936), pp. 55-98, esp. 81; E. J. BAUER, Thomistische Metaphysik an der alten Benediktineruniversität Salzburg. Darstellung und Interpretation einer philosophischen Schule des 17./18. Jahrhunderts, Innsbruck 1996, pp. 23, 749, and 846 (index). On Hidalgo (d. 1768), an Augustinian Hermit active in Sevilla, see G. DÍAZ DÍAZ, Hombres y documentos de la filosofía española, Vol. IV, Madrid 1991, pp. 64-65; F. SÁNCHEZ-BLANCO, “Agustinismo político y moral en la España del siglo XVIII,” in: M. MESTRE ZARAGOZÁ – J. PÉREZ MAGALLÓN – Ph. RABATÉ (eds.), Augustin en Espagne (XVIe-XVIIIe siècles), Toulouse 2015, pp. 289-309, esp. 294. On Signoriello (1820-1889), see A. PELZER, “Les initiateurs du néothomisme contemporain,” in: Revue Néo-Scolastique de Philosophie 18 (1911), pp. 230-254, esp. 250-251 (he praises Signoriello’s Lexicon as “un des meilleurs travaux dans un domaine où il reste beaucoup à faire”); Th. MARSCHLER, “Nineteenth-Century Catholic Reception of Aquinas,” in: LEVERING – PLESTED (eds.), The Oxford Handbook of the Reception of Aquinas, pp. 359-374, esp. 361. 104. Œuvres complètes de Voltaire, Tome 45e, À Gotha, Chez Ch.-G. Ettinger, 1787, p. 1. On this Bildungsroman, see M. KASTEN, “Labouring in Reason’s Vineyard: Voltaire and the Allegory of Enlightenment,” in: W. OTTEN – A. VANDERJAGT – H. DE VRIES (eds.), How the West Was Won. Essays on Literary Imagination, the Canon, and the Christian Middle Ages for Burcht Pranger, Leiden 2010, pp. 101-116.

NATURA INTENDIT SPECIEM

383

While this evocative description of nature is devoid of the metaphysical optimism that pervades scholastic views on nature,105 it nonetheless does retain the gist of the Avicennian adage. In his essay on Das Wesen des Christenthums, Ludwig Feuerbach remarks: “Human beings pass away, but humankind remains, says a pagan philosopher. [...] Christianity, on the contrary, discarded the species, and had only the individual in its eye and mind.”106 Such a profound antithesis between pagan collectivism and Christian individualism has often been reiterated and it still survives in historiographical narratives that depict high-medieval culture as the epoch which ‘discovered’ the individual.107 However, the reception history of the Avicennian dictum “nature intends the species” must lead us to debunk the myth that scholasticism decisively established the primacy of the concrete individual. At least in the areas of natural philosophy and metaphysics, most scholastics seem to have clung firmly to the originally pagan prioritization of the species. Even in the Renaissance (a period frequently associated with individualism),108 the idea that nature intends the individual does not supersede the species-centered interpretation of the Avicennian dictum. 105. Cf. the article “Bien (Tout est –)” in VOLTAIRE, Dictionnaire philosophique, ed. R. POMEAU, Paris 1964, p. 71, where he applies the simile of the Machiavellian despot to “l’auteur de toute la nature” as portrayed by Lord Shaftesbury and other British optimists. On this article, see M. HELLWIG, Alles ist gut. Untersuchungen zur Geschichte einer Theodizee-Formel im 18. Jahrhundert in Deutschland, England und Frankreich, Würzburg 2008, p. 152. 106. L. FEUERBACH, Das Wesen des Christentums, Nachwort von K. LÖWITH, Stuttgart 2011 (1849), pp. 237-238 = The Essence of Christianity, transl. by G. ELIOT, Cambridge 2012 (1854), p. 150 (translation modified). Although Feuerbach does not cite the Avicennian dictum in this context, it is not unlikely that he alludes to it. He may also have been thinking of Cesare Cremonini, who in his Dialectica notes: “If all human beings were destroyed at once, humanity would necessarily be destroyed, but since they pass away successively and given that, after some human beings have passed away, there still remain others, humankind is preserved in eternity” (Dialectica, Venetiis, Apud Guerilios, 1663, p. 97). 107. See e.g. H. DERSCHKA, Individuum und Persönlichkeit im Hochmittelalter, Stuttgart 2014, pp. 9-23. 108. See e.g. E. CASSIRER, Individuum und Kosmos in der Philosophie der Renaissance, Hamburg 2013 (1927), pp. 147-153; M. P. HEATH, The Christian Roots of Individualism, Cham 2019 (esp. chs. 6-8); for more nuanced re-appraisals, see J. J. MARTIN, Myths of Renaissance Individualism, Basingstoke / New York 2004 (esp. chs. 1 and 7), and D. BIOW, On the Importance of Being an Individual in Renaissance Italy. Men, Their Professions, and Their Beards, Philadelphia 2015, esp. pp. 226-228.

384

GUY GULDENTOPS

However this may be, it is clear that the Avicennian topos functioned as a building block of philosophical and theological discourse over many centuries. Nowadays, most philosophers (and perhaps even most historians of medieval thought) are unaware of this topos and its convoluted history. Even though the modern disenchantment of the world has radically undermined the belief in nature’s teleological structure and a fortiori in its intentionality, and though evolutionary biology has problematized the very notion of species,109 quite a few of us would still be inclined to agree with Arthur Schopenhauer’s thesis that nature is “as careful for the maintenance of the species as it is indifferent to the destruction of the individuals.”110 This picture need not necessarily plunge us into a deep and gloomy pessimism. On the contrary, it shows that the demise of individuals, while thwarting and terminating their singularized nature, emanates from “the law of universal nature”: “Even if death is not what is intended with respect to the particular nature that is in Zayd, it is in certain ways what is intended with respect to the universal nature.”111 Already in Antiquity, taking up the quasi-cosmic standpoint of the natural sciences (including psychology and cosmology) was valued as a vital step toward human greatness.112 Insofar as such a scientific viewpoint helps us 109. See e.g. F. E. ZACHOS, Species Concepts in Biology. Historical Development, Theoretical Foundations and Practical Relevance, Springer (Switzerland) 2016, esp. pp. 3-16 and 175-179. 110. A. SCHOPENHAUER, Die Welt als Wille und Vorstellung, Ergänzungen zum vierten Buch, § 41 (“Ueber den Tod und sein Verhältniß zur Unzerstörbarkeit unsers Wesens an sich”), in: Sämtliche Werke, Dritter Band, Wiesbaden 1949, p. 554 = The World as Will and Idea, Translated from the German by R. B. HALDANE and J. KEMP, Vol. III, Containing Supplements [...], London 1886, p. 277. On this passage, see J. LACHS, “The Insignificance of Individuals,” in: Transactions of the Charles S. Peirce Society 38 (2002), pp. 79-93, esp. 82-83. 111. See AVICENNA, Liber primus naturalium, I, 7, ed. VAN RIET, p. 68, 71-74 (cf. supra, n. 36) = MCGINNIS, The Physics of The Healing [...] Books I & II, p. 53. The Latin lex renders the Arabic maqṣūd, which corresponds to the English ‘intended [goal]’. On Bacon’s use of this passage, see JANSSENS, “The Reception of Avicenna’s Physics in the Latin Middle Ages,” p. 60; M. PEREIRA, “Roger Bacon on Nature,” in: N. POLLONI – Y. KEDAR (eds.), The Philosophy and Science of Roger Bacon, London / New York 2021, pp. 17-35, esp. 27-28. 112. See R. BRAGUE, La sagesse du monde. Histoire de l’expérience humaine de l’univers, Paris 1999, pp. 139-140; P. HADOT, Le voile d’Isis. Essai sur l’histoire de l’idée de Nature, Paris 2004, p. 195; G. D. WILLIAMS, The Cosmic Viewpoint. A Study of Seneca’s Natural Questions, Oxford 2012, pp. 17-53; M. RASHED, L’héritage aristotélicien. Textes inédits de l’Antiquité, Paris 2016, pp. 450-451 and 465-467. Cf. AVICENNA, Metaphysics, X, 5, transl.

NATURA INTENDIT SPECIEM

385

overcome an individualist and self-absorbed perspective on the world, it may sometimes enable us to experience something analogous to what Avicenna calls “intelligible beauty and enduring joy.”113 Guy GULDENTOPS Thomas-Institut – Philosophische Fakultät Universität zu Köln [email protected]

MARMURA, p. 378: “Whoever combines theoretical wisdom with justice is indeed happy”; Avicenna, however, does not explicitly dwell on the moral utility of natural philosophy (I owe this reference to Amos Bertolacci). 113. AVICENNA, Liber de anima seu Sextus de naturalibus, V, 6, ed. S. VAN RIET, Louvain / Leiden 1968, p. 150, 71-73. This passage is quoted or paraphrased by several Latin scholastics: see D. N. HASSE, Avicenna’s De Anima in the Latin West. The Formation of a Peripatetic Philosophy of the Soul, 1160–1300, London / Turin 2000, p. 308 (§ p); to the passages mentioned there, one should add HENRY BATE, Speculum divinorum, III, 12, ed. E. VAN DE VYVER, Louvain 1967, p. 202, 37-40, and XVI, 15, ed. G. GULDENTOPS, Leuven 2002, p. 336, 94-97; for much later quotations, see e.g. BERNARDINUS DE BUSTO (O.F.M. Obs.), “De nobilitate animæ sermo decimus,” in: ID., Rosarium sermonum [...]. Tomus secundus, Brixiæ, Apud P. M. Marchettum, 1588, p. 285a; GIOVANNI BATTISTA BERNARDO, Seminarii Totius Philosophiæ Stoicæ Tomus Tertius, p. 11b; FILIPPO FABRI (O.F.M. Conv.), Expositiones et Disputationes in XII Libros Aristotelis Metaphysicorum, Venetiis, Typis M. Ginammi, 1637, p. 39a; ANGELO MARIA MARCHESINI (O.F.M. Cap.), La Consulta d’infedeli fedele, Vicenza, Per G. Berno, 1684, pp. 95-96. – I am grateful to the referees for their comments; I thank Cal Ledsham (Melbourne) for polishing my English.

A NOTE ON THE SOLID CIRCLE: GENTILE, ROSMINI, CONTARINI AND THE HISTORY OF PHILOSOPHY* Andrea Aldo ROBIGLIO Nun sind umschlossen/ Im engsten Ringe/ Im stillsten Herzen/ Weltweite Dinge [Wilhelm Raabe] Abstract This note deals with the formula of Antonio Rosmini’s Logic that Giovanni Gentile had focused on and aims to further explore its hitherto unnoticed sources: in particular, the origin of Rosmini’s expression ‘solid circle’. In the travels of a geometric metaphor, from a Renaissance reader of Avicenna, Gasparo Contarini (1483-1542), to a twentieth-century neo-idealist, Giovanni Gentile (1875-1944), passing through the thought of Antonio Rosmini (17971855), one may uncover distinct ways of conceptualizing the relation between philosophy and its history.

The late Frederick Copleston once conceded that “if by the history of philosophy we understand the actual historical development of philosophy in time, not only is it reasonable to regard philosophy and its history as one and the same reality but also it is difficult to see how we could justifiably avoid doing so.”1 The author continues and * This note aims to celebrate a colleague and friend, and it does so in an incidental way, as the by-product of my research and teaching consecrated to Gasparo Contarini in the years 2012-2016. The point analyzed here was an aspect of an unpublished paper, “‛Solid Circle’ and Transcendental Method: From Giovanni Gentile to Gasparo Contarini,” presented at the workshop “Philosophy at the Court of Charles V: Cardinal Gasparo Contarini, the immortality of the soul and the preeminence of ‘political life’,” Leuven: May 21, 2013. That meeting also represented one of the first activities of the then newly established Philosophical Review Club. The Club has provided an engaging, convivial venue for scholarly discussion at the Hoger Instituut voor Wijsbegeerte, of which Jules Janssens has been a regular and enthusiastic participant since its beginnings. The current contribution remains very limited in its scope, as it does not aim to take into account the current scholarly debate on Rosmini’s philosophy, nor does it explore Rosmini’s debt

388

ANDREA ALDO ROBIGLIO

attempts to raise a “possible line of objection” to the quoted contention. However, even those who have decidedly rejected any identity of philosophy and its history, in fact, seem ready to admit that there must exist some circular implication of the two notions, though variance may arise concerning the nature of such circularity – e.g., whether its character is vicious or virtuous.2 1. Giovanni Gentile’s theory of the mind as pure act The names of two Italian thinkers, Benedetto Croce (1866-1952) and Giovanni Gentile (1875-1944), should not be neglected if one seeks a sophisticated philosophical analysis of how the “history of philosophy” ought to be conceptualized. A major intellectual figure in Italy during the first decades of the twentieth century, the name of Gentile is less known than Croce’s. Academic, politician, and editor, Gentile arguably became the most relevant intellectual figure of Italian Fascism. Gentile had met Benedetto Croce while he was still a young researcher and, from 1903 to 1922, they edited the innovative periodical, La Critica. Despite being his junior, Gentile influenced Croce’s engagement with systematic philosophy.3 Gentile’s ‘actual idealism’ shows the deep influence of German Idealism (and, more specifically, that of Johann Gottlieb Fichte) and is the subject of Teoria generale dello spirito come atto puro, published in 1916 and dedicated to his friend.4 toward Renaissance and Early Modern ontologies. I should like to thank: Dr. Brian Garcia, for the stylistic revision of this Note; the Editors, for letting me participate in this expedient Festschrift; and the anonymous reviewers, for their observations. 1. F. COPLESTON, “Philosophy and Its History,” in: Philosophy 67 (1992), No. 261, pp. 357-365, at p. 361. 2. Relevant for the debate on the relation between philosophy and its history, see R. PETERS, History as Thought and Action: The Philosophies of Croce, Gentile, de Ruggiero and Collingwood, Exeter 2013, and M. R. ANTOGNAZZA, “The Benefit to Philosophy of the Study of Its History”, in: British Journal for the History of Philosophy 23 (2015), pp. 161-184. 3. With Gentile’s embrace of Fascism, after 1924, the philosophical disagreements crystallized into a lasting rupture between the two men. See G. TURI, “Giovanni Gentile: Oblivion, Remembrance, and Criticism,” in: The Journal of Modern History 70 (1998), pp. 913-933. On Gentile as a specialist of both medieval and Renaissance thought, see A. SCAZZOLA, Giovanni Gentile e il Rinascimento, Naples 2002; R. RUBINI, The Other Renaissance: Italian Humanists between Hegel and Heidegger, Chicago 2014, pp. 61-105. 4. See G. GENTILE, Teoria generale dello spirito come atto puro, in: ID., Opere filosofiche, ed. E. GARIN, Milan 1991, pp. 453-682. For the English translation, I follow H. Wildon Carr’s translation: G. GENTILE, The Theory of Mind as Pure Act, London 1922.

A NOTE ON THE SOLID CIRCLE

389

Croce promptly read the tract and was impressed by its coherence and argument, though he did notice a point of possible dissent. The relation between history and freedom, i.e., between history and the spirit that makes it, should never fail to do complete justice to the “concrete,” viz., the particular, individual reality, “which is history,” the only actual subject of any possible intellectual history.5 Croce, by thus writing, alluded to the risk that by taking the “pure” spiritual act in an absolute way, the philosopher might overlook the polarities and distinctions, that rebellious diversity which makes the life of the mind. Croce’s letter refers to Ch. XIV, § 6 (“History of art as history of philosophy”), to a page where he could find his name quoted together with that of Francesco De Sanctis.6 The problem he touches upon, however, had been expounded by Gentile in the previous chapter, Ch. XIII, devoted to “The historical antinomy and eternal history”.7 Let us sketch it in brief. Gentile moves from an antinomy between two statements which, according to him, express self-evident truths. The first statement is as follows: “Mind is history, because it is dialectical development.” The second principle affirms instead: “Mind is not history, because it is an eternal act.” The contradiction must be overcome, since (§ 3): “when the mind is historicized it is changed into a natural entity” – it neither is nor becomes any more, and it now may present itself as an object of criticism. However, “when the mind’s spiritual value only is kept in view, it is withdrawn from history and stands before us in eternal ideality.” Such antinomy turns into a contradiction as soon as we are no longer Platonists – i.e., “when the 5. B. CROCE, Lettere a Giovanni Gentile, ed. A. CROCE, Milan 1981, pp. 517-518. 6. See GENTILE, Teoria generale, p. 639; ID., The Theory of Mind, p. 226. 7. See GENTILE, Teoria generale, pp. 621-634; ID., The Theory of Mind, pp. 202-219; this chapter concludes with a footnote by Gentile, at p. 217, where one can read the following appreciation: “This important consideration must, I believe, have escaped the notice of my friend, Benedetto Croce, when he disputed the difference here expounded between history of art and history of philosophy, denying that ‘men have been exercised over one unique philosophical problem, the successive and ever less inadequate solutions of which form one single line of progress’. It is true that men have worked at problems which are always different; but man, mind, the spirit which is actually working in the history of philosophy, works at one problem which is its own and unique. System is not to be sought in a history in itself, which has no existence, but in that real history of which Croce himself speaks with such insight in the book referred to (p. 5), the history, as he says, ‘which is really thought in the act which thinks it’ [che realmente si pensa nel fatto che si pensa]: in the historian’s mind who writes it. The historian can only determine his object as the development of his own concept, that is, of himself.”

390

ANDREA ALDO ROBIGLIO

transcendence is demolished, and we conceive spiritual reality in its eternity no longer as something fixed but as process in act, then we no longer have history outside the eternal, nor the eternal outside history.”8 Soon thereafter (in § 5) Gentile overcomes the contradiction and solves it “as all antinomies are solved, by bringing the spiritual reality, value and history, from abstract thought to the concrete.” There must be identity, Gentile argues, between the subject who actually knows and the spiritual reality which is known. However, is not this simply a trick, an empty formula – viz., the vicious circle of subject and object? The concept of process or development is the crucial element in Gentile’s argument. First, two modes of conceiving history must be analyzed: history as facts, on the one hand, and history as “the actual historical development” of thought, as acts, on the other. Secondly, the first mode must be dismissed as mere “chronicler’s history,” history “hypostatized” and deprived of its dialectic (§ 7), while the second mode must be taken as genuine “history”. Thirdly, the spatial metaphor of the line must also be rejected, since history as act can only be described by an idea of a linear progression in a misleading fashion (§ 8). Finally, “the antinomy is solved in the concept of the process of the unity, which in being multiplied remains one” (§ 9). In the following three paragraphs, Gentile explains the appropriate metaphor to speak of any actual history and, quintessentially, of the history of philosophy: the image of a line no longer suffices, but rather a circle is offered. Paragraph 10 opens as follows: The clearest confirmation we can give of the doctrine of the identity of the ideal and eternal history with the history which is developed in time, is that which in recent years has been formulated in Italy in the theory that philosophy and history of philosophy are a circle.9

While indulging in Hegel as an historian of philosophy (§ 11), Gentile addresses the worry we already introduced above. Is not the circle of history and philosophy a vicious one? The last part of the paragraph deserves to be read, together with § 12, in full: […] Well, then, what will each include in his own history as matter which belongs to the history of philosophy? A choice of material is 8. See GENTILE, Teoria generale, p. 623; ID., The Theory of Mind, p. 205. 9. See GENTILE, Teoria generale, p. 627; ID., The Theory of Mind, p. 209 (my italics).

A NOTE ON THE SOLID CIRCLE

391

inevitable; and a choice requires a criterion. And the criterion in this case can only be a notion of the philosophy. Again, every history disposes its own materials in an order and in a certain perspective, and each disposes its materials so that what is more important stands out in relief in the foreground and what is less important and more remote is pushed into the background, and all this implies a choice among various possible orders and dispositions. And this choice of order and disposition of the already chosen material itself also requires its criterion, and therefore the intervention of a mode of understanding and judging matter which here can only be a philosophy. So we reach the conclusion that the history of philosophy which must precede philosophy presupposes philosophy. And we have the circle. § 12 [Identity and a Solid Circle] For a long while it was suspected that it was a vicious circle. Were it so, it would be necessary to choose between a history of philosophy and a solid which presupposed no philosophy but was itself presupposed by it, and a history of philosophy not presupposed by philosophy nor a basis of it. But more accurate reflection has shown that the circle is not logically vicious, it is rather one of those which Rosmini in his Logica calls solid circles (circoli solidi), that is, circles which are unbreakable, or regresses (regressi), as they were termed by Jacopo Zabarella, the Italian sixteenth century philosopher. (N) And it has been shown that no philosophy is conceivable which is not based on the history of philosophy, and no history of philosophy which does not lean on philosophy, since philosophy and its history are together one as process of mind. In the process of mind it is possible to distinguish empirically an historical from a systematic treatment of philosophy and to think of either of the two terms as presupposing the other, since speculatively the one is itself the other, although in a different form, as the various stages of the spiritual process considered abstractly are always different.10

One last passage is required for interpreting Gentile’s text. It is a long, technical note (N) in which the author, after having referred to Antonio Rosmini and to his notion of the “solid circle”, wants to 10. See GENTILE, Teoria generale, pp. 629-630; ID., The Theory of Mind, pp. 211-213. Twenty years later, Gentile further explained his overarching conception of “history as act” in an article he submitted to a Festschrift presented to Ernst Cassirer on his sixtieth birthday; cf. G. GENTILE, “The Transcending of Time in History”, in: Philosophy and History: Essays presented to Ernst Cassirer, ed. R. KLIBANSKY – H. J. PATON, Oxford 1936, pp. 91-105. Gentile’s “extreme subjectivism” also played a formative role in the theory of history elaborated by the late Hayden White (1928-2018), as Carlo Ginzburg notes; see C. GINZBURG, “Just One Witness: The Extermination of the Jews and the Principle of Reality [for Primo Levi]”, in ID., Threads and Traces: True, False, Fictive, Berkeley / Los Angeles 2012, pp. 165-179, 293-298, at p. 172.

392

ANDREA ALDO ROBIGLIO

stress his own interpretation and to show how it departs from Rosmini’s Logica. Let’s read the footnote N, for the sake of completeness: Who wrote with insight: “Sicut rerum omnium, quae in universo sunt, admirabilis est colligatio et nexus et ordo, ita in scientiis contingere necesse fuit, ut colligatae essent, et mutuum sibi auxilium praestarent” (De Regressu, in Rosmini, Logica, p. 274 n.). Rosmini, who founds his concept of the solid circle on the “synthesizing character of nature” (sintesismo della natura), formulates it thus: “The mind cannot know any particular thing except by means of a virtual cognition of the whole”, the mind therefore has “to pass to the actual cognition of the particular by means of its virtual acquaintance (notizia) with the whole; and to return from actual particular cognition to the actual acquaintance, that is with some degree of actuality of the whole itself” (p. 274). But the untenable distinction between the virtual and the actual prevents Rosmini from perceiving the deeper meaning of the circle. He will not let us take it as the Identity of the two terms bound together by the circle. It was impossible, indeed, for him to give up the distinction of actual and virtual, because he had not attained to the concept of process in which distinction is generated from within identity itself. Hegel, on this matter more exact than Rosmini, had said: “Philosophy forms a circle. It has an initial or immediate point – for it must begin somewhere – a point which is not demonstrated and is not a result. But the starting point of philosophy is immediately relative, for it must appear at another end-point as a result. Philosophy is a sequence which is not suspended in mid-air; it does not begin immediately, but is rounded off within itself” (Grundl. der Phil. des Rechts, § 2, Zusatz).11

While the figure of the circle is Hegelian, for the use of the syntagma “solid circle” – which finds no corresponding expression in Hegel’s writings – an Italian philosopher is credited. 2. Antonio Rosmini, the ‘solid circle’, and Theosophy Antonio Rosmini Serbati (1797-1855), philosopher, priest, religious reformer, and founder of the Institute of Charity, was born in Rovereto, then part of the Austro-Hungarian Empire.12 He left a 11. The passage from Hegel corresponds to the well-known addition G to § 2 in Hegel’s Grundlinien der Philosophie des Rechts. Hugh Barr Nisbet’s translation replaces here Carr’s translation from the Italian. See G. W. F. HEGEL, Elements of the Philosophy of Right, ed. A.W. WOOD, Cambridge 1991, p. 26. 12. Cf. D. CLEARY, Antonio Rosmini: Introduction to his Life and Teaching, Durham 1992.

A NOTE ON THE SOLID CIRCLE

393

considerable series of writings and a vast correspondence, which display an impressive breadth of learning and comprehend theology, spirituality, psychology, politics, ecclesiology, and – above all – an ambitious philosophical system. While acquaintance with Rosmini’s thought is widespread in Italy,13 it is generally scant elsewhere.14 And this despite the fact that some of his writings had been available in English already at the end of the nineteenth century and an illustrious thinker such as William James had paid attention to them.15 In a 13. See, for instance, P. DE LUCIA, “Antropologia e metafisica negli studi rosminiani degli ultimi cinquant’anni,” in: Rivista di Filosofia Neo-Scolastica 90 (1998), pp. 215-231. The journal Rivista rosminiana di filosofia e di cultura, currently published by the Institute of Charity, which started its course in 1906, is arguably the official organ of the scholarly activities related to Rosmini’s legacy; see http://www.rosmini.it. For the purpose of the present note, see also M. DONÀ, L’uno e i molti. Rosmini-Hegel un dialogo filosofico, Rome 2001, and especially F. PERCIVALE, L’ascesa naturale a Dio nella Filosofia di Rosmini, Rome 2000. 14. Ultimately, one should read the remarkable volume, From Kant to Croce: Modern Philosophy in Italy, 1800-1950, ed. B. COPENHAVER – R. COPENHAVER, Toronto 2012. Significant also is that the online Stanford Encyclopedia of Philosophy includes an entry on Rosmini, authored by Denis Cleary in 2001. The project of translating Rosmini’s work in English progresses as well, cf. R. POZZO, “The Philosophical Works of Antonio Rosmini in English Translation,” in: American Catholic Philosophical Quarterly 73 (1999), pp. 609637. Both Rosmini’s ethics and his political philosophy have been discussed in recent years: see N. P. SWARTZ, “Rosmini: Thomist or Neo-Thomist? The Debate Continues,” in: Journal of Politics and Law 2 (2009), pp. 120-131; ID., “Rosmini on Individual Rights: The Soul (Reason) as Forerunner of Individual Rights in Human Society,” in: Journal of Politics and Law 2 (2009), pp. 100-106. Translations of Rosmini’s works are also available in German: for a survey of his reception, see M. KRIENKE, “Antonio Rosmini in Deutschland,” in: M. DOSSI, Antonio Rosmini. Ein philosophisches Profil, Stuttgart 2003, pp. 9-39. The acquaintance of the francophone readers with Rosmini, on the other hand, looks meager: few significant analyses of his thought are, to my knowledge, those by Léonce Paquet, Régis Jolivet, and François Évain. 15. See James’s review of The philosophical system of Antonio Rosmini Serbati, translated by Th. Davidson (London 1882), in: W. JAMES, Essays, Comments, and Reviews, Cambridge MA 1987, pp. 379-383; and James’s review of Rosmini’s Psychology (vol. ii, London 1885), in ID., Essays, Comments and Reviews, pp. 396-398. Elsewhere, sketching the philosophical profile of Thomas Davidson, James wrote: “There are men whose attitude is always that of seeking for truth, and men who on the contrary always believe that they have the root of it already in them. Davidson was of the latter class. Like his countrymen, Carlyle and Ruskin, he felt himself to be in the possession of something, whether articulate or as yet articulated by himself, that authorized him (and authorized him with uncommon openness and frequency) to condemn the errors of others. I think that to the last he never fully extricated this philosophy. It was a tendency, a faith in a direction, which gave him an active persuasion that other directions were false, but of which the central insight never got fully formulated, but remained in a state which Frederic Myers would have called subliminal. He varied to a certain extent his watchwords and his heroes. When I first knew him all was Aristotle. Later all was Rosmini. Later still Rosmini seemed forgotten.

394

ANDREA ALDO ROBIGLIO

short review of the English translation of Rosmini’s Psychology published in Science in 1886, James distraughtly commented: When one thinks of the mere quantity of labor which Rosmini accomplished in his not long life, one cannot refuse to him the title of being one of the very small number of intellectual giants of the world. He is of the race of the Aristotles, the St. Thomases, the Leibnitzes, the Kants and the Hegels. The mere cogitative energy Rosmini may show a path down from his categories to the practical details of life. It were sad that such a strenuous and in many ways such exquisite thinking as his should be among the mere superfluities of human history.16

This note deals with the formula of Rosmini’s Logic that Gentile had focused on and aims to further explore its hitherto unnoticed sources: in particular, the origin of Rosmini’s expression regarding the “solid circle.” The Logica libri tre (1850-51)17 constitutes a bridge between Rosmini’s extensive Psicologia libri dieci (1843-46)18 and his posthumous Teosofia. The accomplished expression of Rosmini’s metaphysical thought – as Enrico Berti cogently stressed – is the posthumously published Theosophy, where the thinker develops his entirely original ontology and natural philosophy.19 In that work, the author disproves He knew so many writers that he grew fond of very various ones and had a strange tolerance for systematizers and dogmatizers whom, as the consistent individualist that he was, he should have disliked. Hegel, it is true, he detested; but he always spoke with reverence of Kant” (W. JAMES, “Thomas Davidson: Individualist,” in: ID., Essays, Comments, and Reviews, pp. 86-97, at pp. 87-88). Cf. J. L. BLAU, [Review of] Rosmini, Domodossola, and Thomas Davidson, in: Journal of the History of Ideas 18 (1957), pp. 522-525. 16. ID., Essays, Comments and Reviews, pp. 396-397. 17. A. ROSMINI, Logica libri tre, ed. V. SALA, Stresa / Rome 1984. The first edition dates back to 1853, in Turin, from the press of “Cugini Pomba.” 18. A. ROSMINI, Psicologia libri dieci, t. I, ed. V. SALA, Stresa / Rome 1988. The author considered his Psychology as the continuation of the Antropologia in servigio della scienza morale (1838). 19. See, e.g., E. BERTI, La metafisica di Platone e di Aristotele nell’interpretazione di A. Rosmini, Rome 1978, pp. 126-127. The word ‘teosofia’ is taken by Rosmini in its “meaning of ‘wisdom about God’ in so far as God is the supreme Being and the apex of philosophical speculation” (Cleary). Paolo Gomarasca observes: “The entire Theosophy is built on the awareness deriving from both Psychology and Logic: each and every act of understanding originates itself as an application of the primum notum […] to what determinates the bodily fundamental feeling and, in a further step, to the objects provided by the perception to the mind. This happens in the full respect of the formal rules which Logic establishes for each use of reasoning […] always implying the immediate vision of truth” (P. GOMARASCA, Rosmini e la forma morale dell’essere. La ‘poeisi’ del bene come destino della metafísica, Milan 1998, p. 64).

A NOTE ON THE SOLID CIRCLE

395

the critiques put forwards by Vincenzo Gioberti, according to whom Rosmini’s conception of being had been “psychologistic”.20 More extensively and explicitly than he had previously done, Rosmini in the Theosophy argues for the objective character of the “idea of being,” which includes all the possibilities of being. At first blush this has a Kantian ring to it, even if the categories of “subject” and “object” employed by Rosmini are conceptualized in a different manner than those of Kant. Already in the “Introduction” to the first volume of his Psychology, the reader could find one of Rosmini’s characteristic epistemological insights, which he presents as the middle path between the opposing one-sidedness of both Empiricism and Idealism: Merely to feel is not to observe. Observation implies an act of the mind that makes a feeling its object and concludes in a judgment. This act of the mind, either a judgment or an act of reasoning, is in the last analysis the application of ideal being to the feeling on which the mind fixes its attention. Every reasoning, therefore, necessarily includes two elements: 1) ideal being and 2) the feeling to which ideal being is applied. The information obtained by way of reasoning about one of these two things cannot be had without information about the other. The two pieces of information [notizie] are therefore posited in us contemporaneously. This is what I call synthesism.21

Without the idea of being, perception would be totally blind and senseless; conversely, the idea of being would never become the object of knowledge without the participation of feeling.22 Metaphysics as a 20. Vincenzo Gioberti, an Italian statesman and philosopher, was born in Turin in 1801 and died in Paris at age 51. The work in which he criticized Rosmini’s metaphysics was first published in 1843 and again three years later as a revised second edition, in Switzerland, near Lugano: V. GIOBERTI, Degli errori filosofici di Antonio Rosmini, Capolago 1946. Cf. From Kant to Croce, Part I, and P. DE LUCIA, La ragione nei limiti della pura rivelazione. Vincenzo Gioberti e la filosofia positiva, Rome 2012. 21. ROSMINI, Psicologia, Book I, Introduction, Ch. 8, n° 34, p. 47; the English translation is taken, with very slight adjustments, from the 1884 edition published in London. 22. Relevant here is the footnote 22, related to the paragraph quoted above: “It will be objected that the intuition of being is a cognition which has no need of reasoning. This is true, but no science is formed by this alone. Every science is a witness to our consciousness [deposizione della nostra coscienza]. Even the intuition of being does not enter the field of science unless we carry out an intellective act by which we tell ourselves that we possess the intuition of being. But this is impossible unless we reflect upon what is present in our spirit; telling ourselves that such is the case already means that we pronounce some judgment. But we cannot demonstrate the certainty of this judgment without making a further

396

ANDREA ALDO ROBIGLIO

science, therefore, is only possible insofar as the intellectual perception of any being is given as actual knowledge, i.e., instantiated in the concrete synthesis of a material element (“feeling”) and a formal one. We may employ here the terminology of Giovanni Gentile which carries with it Kantian undertones: the material element is the “concrete of the abstract,” while the formal element is the “abstract of the concrete.”23 The formal element is the ideal being, of which the concrete is the “term”, at any instantiation. Actual perception terminates the idea of being to the feeling, i.e., the former is directly applied to the latter, while the object of perception terminates the actual act of understanding – which still includes feeling, though only indirectly. This explains why “the knowable essence is positive when attained by way of perception.”24 Perception enables the human mind to know the substance of the being positively. Consequently, the substance known positively may become the object of the sciences.25 In each and every perception, as well as in each and every intellection, the idea of being is implied and given as applied or instantiated. Metaphysics as the science of being presents this unavoidable circularity: knowing the real substances presupposes an innate idea of unlimited being; and, setting out from the knowledge of substances and their limited perfections, metaphysics presupposes that the human mind can grasp the notion of the most perfect and unlimited being. The being that is the highest and proper object of metaphysics is also, in a way, a required presupposition of it – yet not in one and the same way. Rosmini carefully assesses such a circularity by introducing the distinction between “virtual” and “actual” unlimited being – viz., between virtual and actual totality. Human knowledge of particulars reflection upon it, that is, an act of reasoning which takes the form: ‘We intuit being. But being is that which is. Therefore we intuit that which is. But that which is, is the truth. Therefore we intuit the truth. Therefore the intuition of being cannot mislead us; being which we intuit cannot be an appearance. If it were, this appearance would be the truth, which is a contradiction’. We need some reasoning, therefore, to make the very intuition of being an object of science” (ROSMINI, Psicologia, p. 47). 23. A recent and accurate illustration of this expression is found in P. BETTINESCHI, Critica della prassi assoluta. Analisi dell’idealismo gentiliano, Naples 2011, pp. 41-48. See § 1, above. 24. ROSMINI, Psicologia, n° 108, pp. 83-84, with reference to nn. 19-21, pp. 38-41. 25. Cf. ROSMINI, Psicologia, Book I, Ch. 9 (“The principle of psychology”), nn. 107114, pp. 83-86.

A NOTE ON THE SOLID CIRCLE

397

relies on the givenness of virtual totality, which furnishes its epistemological foundation; still, the ontological foundation consists in the actual totality, from which the virtual totality presents only a promise. Since only the relations among particular objects can make the aspiration to any scientific or systematic knowledge possible and, as such, entail the virtual notion of totality – viz., the innate idea of being – the latter has to be taken seriously: it can hardly be a merely subjective concept. In his Logic, Rosmini refers to a form of dialectic like the one rapidly sketched out here and calls it a “solid circle”.26 The description of such a peculiar syntagma, in Rosmini’s own words, is the following: In the case of a vicious circle, the line returns to the very same point from which it departed, while in the solid circle, its line may return to another, higher point, having turned around a cylinder like a spiral.27

The method of the solid circle is arguably in agreement with the insight of “synthesism” and, as it has been noticed, is constantly at work in the fundamental argument of Rosmini’s Theosophy.28 If we now apply Rosmini’s solid circle to the problem of the history of philosophy, we should affirm that the history of philosophy acquires its philosophical legitimacy insofar it realizes the passage from the 26. ROSMINI Logica, II, nn. 701-704, pp. 281-283. 27. ROSMINI, Logica, n° 741 n. 43, p. 302 ; the full footnote reads: “Chiamiamo questo sofisma circolo lineare, per distinguerlo dal circolo solido, che è il quinto modo interiore d’argomentare. La convenienza di tale denominazione appare da questo, che nel circolo lineare, la linea rientra precisamente a quello stesso punto ond’era partita, laddove nel circolo solido può rientrare in altro punto girando a guisa di spirale sopra un cilindro onde si può chiamare anche cilindrica questa maniera d’argomentare. Col circolo lineare si prova lo stesso per lo stesso, onde in questo sofisma: 1° come le due premesse provano la conclusione, così si pretende, che entrambi le premesse sieno provate dalla conclusione; 2° collo stesso mezzo termine con quale si formano le premesse (ARIST., Poster. Anal. II, t. 4, 6; AVERROÈ, In Poster. II, Comment. 4, 8), che provano la conclusione, si pretende provare le premesse; 3° la stessa identica cognizione che s’è assunta per provare la conclusione, si vuol provare assumendo per data la conclusione. All’incontro nel circolo solido una sola delle premesse, cioé la minore, riman provata argomentando dalla conclusione, il termine medio cangia ne’ due sillogismi, e la conclusione, che si assume per provare quella che era minore, non è identica, ma ampliata dalla considerazione della mente (che i Logici chiamano negotiatio intellectus, o mentale causae examen), per modo che alla nuda conclusione ricevuta del primo sillogismo s’aggiunge un elemento nuovo, su cui si fonda il secondo sillogismo. Vedi ZABARELLA, De regressu, cap. 8.” 28. See PERCIVALE, L’ascesa naturale a Dio, p. 98 (on the “solid circle” and its meaning for Rosmini, see footnote 131).

398

ANDREA ALDO ROBIGLIO

virtual knowledge of the whole to the actual knowledge of its concrete instantiations. What for Giovanni Gentile is an identity, in Rosmini assumes the appearances of a polarity in perpetual movement, like the circular ascent in a “cylinder” – which, besides the sphere and the solid torus, in Early Modern philosophy could also be referred to as circulus solidus; this solid circle is obtained not by the rotation of a line around a fixed point, but rather by the rotation of a (regular) surface around a line. Let us reconsider the Logic, which Gentile engaged with. There Rosmini explicitly refers to the Renaissance philosophical School of Padua and to the theory of regressus that Jacopo Zabarella (1533-1589) had notoriously taught there.29 However, the use of the syntagma “solid circle” remains partly unexplained – especially as its literal equivalence with “regress” is not obvious. Some writers educated at Padua might have displayed a certain proclivity to use this formula, but when they did so, it was not in the logico-philosophical acceptation, since they were mostly referring to the metaphorical grasp of supernatural notions such as the Holy Trinity or the intact uterus of the Virgin Mary (which may be described as circulus ex omni parte integer). This is certainly the case with Girolamo Panigarola (1548-1594), former disciple of the Paduan physician Bernardino Tomitano (1517-1576)30 and later bishop of Asti. In one 29. See JACOPO ZABARELLA, On Methods, vol. 2 – On Regressus, ed. J. P. MCCASKEY, Cambridge MA 2013, pp. 400-406. In Zabarella’s tract, however, one does not find the terminology of circulus solidus, even though the actual cognition of the particular through the virtual cognition of the universal is indeed discussed: “Naturalis est nobis haec progrediendi ratio a confusa ad distinctam eiusdem rei cognitionem” (p. 406). See the opening page of De regressu (p. 356): “Considerans Aristoteles in primo libro Posteriorum Analyticorum circularem demonstrationem, qua prisci quidam philosophi omnia sciri posse asseverabant, eam omnino refutavit et inutilem esse efficaciter ostendit. […] Verum in illa confutatione ita locutus est Aristoteles, ut non omnem circularem ostensionem reicisse videatur, sed aliquam admisisse, quam complures interpretes recipientes et a circulo illo seiungere volentes ‘regressum’ appellarunt”. See J. SOUTH, “Zabarella, Prime Matter, and the Theory of Regressus,” in: Marquette Graduate Faculty Philosophy Journal 26 (2005), pp. 79-98; and especially W. A. WALLACE, “Circularity and the Paduan Regressus: From Pietro d’Abano to Galileo Galilei,” in: Vivarium 33 (1995), pp. 76-97. 30. See F. PANIGAROLA, Vita scritta da lui medesimo, ed. F. GIUNTA, Bologna 2008, pp. 70-71: at Padua “a leggerli due lezioni ogni giorno: […] e l’altra di logica messer Bernardino Tomitano, uomo famoso de’ suoi tempi e nella logica principalmente tenuto singolare.” The name of Girolamo as a religious was ‘Francesco’.

A NOTE ON THE SOLID CIRCLE

399

of his sermons, Panigarola quotes the Liber XXIV Philosophorum and writes: God himself is a circle, as Hermes Trismegistus says, but at the same time He is center and circumference together. ‘Circulus ex omni parte sibi est congruus et aequalis’ – writes Saint Jerome On Job – ‘Qualiter Pater, talis Filius, talis Spiritus Sanctus’, says Athanasius in the Symbol. ‘Circulus per semetipsum solidus’ (Saint Jerome) […] ‘Circulus qui sine ullo fine in semetipsum revertitur’.31

It is when we turn our attention to the mature Teosofia,32 however, that we can find a useful indication given by the author himself. For those who wish to have an outline of Rosmini’s metaphysical project, as we said above, reading his “Preface” may be illuminating. The indication we are looking for is plainly encountered in the very opening: In the midst of his civic duties Gaspare Contarini, one of those great, outstanding Italians forgotten by their countrymen, found time and repose to write seven books Della prima Filosofia, and concluded this work by exhorting others to complete what he had generously begun. (1) His highly authoritative words were borne away on the wind but, please God, the wisdom of our fellow-countrymen will now at last come to grips with the outline which this Cardinal of Venice conceived and published three centuries ago as a prologue to development in his own time and in the future. […] This book will deal more distinctly 31. Prediche di Monsignore Francesco Panigarola Vescovo di Asti, Predica I, in: Biblioteca classica dei Sacri oratori, greci, latini, italiani, francesi, antichi e recenti, t. IX, Venice 1839, pp. 515-535, at p. 519 (my italics). “[…] è un circolo l’istesso Dio, diceva Mercurio Trismegisto, ma che nell’istesso istante ed è centro e circoferenza insieme. ‘Circulus ex omni parte sibi est congruus et aequalis’, dice san Girolamo sopra Giobbe. ‘Qualiter Pater, talis Filius, talis Spiritus Sanctus’, dice di Dio Atanasio nel Simbolo. ‘Circulus per semetipsum solidus’, San Gerolamo. […] ‘Circulus qui sine ullo fine in semetipsum revertitur’ […] ecco il circolo, e però efficiente di tutte le cose, e ultimo fine di tutte le cose”. The author of the passage which Panigarola refers to is not Jerome, as it seems, but his disciple Philipp the Presbyter; see PHILIPPUS PRESBYTER, In historiam Iob commentariorum libri tres, Basileae 1527, lib. III, Ch. 38, pp. 167-184, at p. 168: “Nam circulus ex omni parte sibi est congruus et aequalis totus, qui per semetipsum solidus vadit et integer, oculorumque aciem per suum infinitum ambitum, quo sine ullo fine super semetipsum currit et inexplicabiliter ducitur”. Cf. K. B. STEINHAUSER, “Job in Patristic Commentaries and Theological Works,” in: A Companion to Job in the Middle Ages, ed. F. T. HARKINS – A. CANTY, Leiden 2017, pp. 34-70, at pp. 43-51. See also Il libro dei ventiquattro filosofi, ed. P. LUCENTINI, Milan 1999. 32. The refence edition is A. ROSMINI, Teosofia, ed. M. A. RASCHINI – P. P. OTTONELLO, t. I, Stresa / Rome 1998, pp. 43-44; here I use ID., Teosofia, ed. S. F. TADINI, Milan 2011, pp. 245-264. The English translation available is ID., Theosophy, vol. 1, The Problem of Ontology. Being-as-One, transl. D. CLEARY – T. WATSON, Durham 2007, pp. 1-25.

400

ANDREA ALDO ROBIGLIO

with subjects attributed to First Philosophy by both Contarini and Aristotle, and by me to what I would prefer to call Theosophy.33

To Rosmini’s manifold agenda the profile of the wise religious reformer, diplomat, and later cardinal, Gasparo Contarini (14831542) would have without doubt appeared seducing.34 In the opening of his philosophical magnum opus, however, Rosmini attests to his familiarity with Contarini the philosopher and with 33. ROSMINI, Theosophy, “Preface”, § 1, pp. 1-2. The patriotic coloration of this incipit goes hardly unnoticed. After the mention of the name of “this great man” and his Primae philosophiae compendium, Rosmini adds a footnote (1) that reads as follows: “He concludes the work with the following words: ‘People more learned than myself and less involved in civil affairs can perfect the matter. My intention was to show why philosophy (or wisdom, if you prefer), by far and away the principal discipline, should be treated first. I myself feel unequal to the task, which requires more time than I have available in the midst of so many raging wars, public duties and travels entailed by my office as legate’.” The explicit of Contarini’s Compendium, from which Rosmini quotes and translates reads: “Nobis haec delibasse sufficiat ex fontibus primae philosophiae, ac de toto entis genere, ac de non ente quaedam dixisse, quae fortasse et Pythagorici et aliorum priscorum plerique sensisse videntur, praeterquam quod naturam quandam posuerunt non entis, quae nulla est: ac secuti ipsi quoque videntur intellectum nostrum, qui non ens, caeteraque illi proxima apprehendit modo altiori, quam is sit quo re vera sunt. Alii doctiores ac minus occupati civilibus negociis rem perficiant: Nostrum namque institutum fuit, ut potius aliis praescriberemus quonam pacto disciplinarum omnium facile princeps philosophia, scilicet haec prima, seu sapientiam appellare malueris, tractari debeat; quam ut nostro labore id fieret, cui impares nos esse sentimus; maiusque ocium poscit, quam nobis inter tot bellorum procellas, atque publicas occupationes, hinc indeque occursantes in hoc legationis munere, quo fungimur, praestari possit”. The author of the Theosophy was arguably using the Aldine edition of 1578, which reproduces the 1571 text printed in Paris; the text of the Primae philosophiae compendium presents no significant textual variants. See The Aldine Press: Catalogue of the Ahmanson-Murphy Collection of Books by or Relating to the Press in the Library of the University of California, Los Angeles Incorporating Works Recorded Elsewhere, ed. P. NAIDITCH – N. BARKER – S. A. KAPLAN, Berkeley 2001, n° 908, p. 422b. On Contarini, see L. BURZELLI, “Contarini, Gasparo,” in: M. SGARBI (ed.), Encyclopedia of Renaissance Philosophy, Cham 2019 (with up-to-date bibliography), and P. B. ROSSI, ‘Sempre alla pietà et buoni costumi ha exortato le genti’: Aristotle in the milieu of Cardinal Contarini († 1542), in: L. BIANCHI (ed.), Christian Readings of Aristotle from the Middle Ages to the Renaissance, Turnhout 2011, pp. 317-389. 34. Cf. G. DE ROSA, “Una rilettura delle ‘Cinque piaghe’ di Antonio Rosmini”, in: ID., Tempo religioso e tempo storico. Saggi e note di storia sociale e religiosa dal Medioevo all’età contemporánea, Rome 1998, pp. 229-258, at p. 237. See also P. MARANGON, Le fonti delle ‘Cinque piaghe’, in: M. MARCOCCHI – F. DE GIORGI (eds.), Il ‘Gran Disegno’ di Rosmini. Origine, fortuna e profezia delle ‘Cinque piaghe della Santa Chiesa’, Milan 1999, pp. 25-54, esp. note 38 on pp. 34-35. Concerning the reputation of Contarini “as a representative of the riper Italian piety, reflecting the morning glory of intellectual freedom”, the reader of this Note may also be interested to know that Friedrich Nietzsche invokes him in Human, All Too Human, Part II, Aphorism 226 (“The Tragi-Comedy of Regensburg”).

A NOTE ON THE SOLID CIRCLE

401

a neglected metaphysical treatise written by the Venetian nobleman, himself a former pupil of Pietro Pomponazzi at the University of Padua – i.e., the same school Rosmini would attend three hundred years later to complete his education in theology. Moreover, Rosmini proudly presents himself as the successor of Contarini in the metaphysical analysis. At the very beginning of his most ambitious work, he reverently points to the thinker who had offered him a distinct inspiration. 3. Gasparo Contarini: A reader of Avicenna and the “solid circle” Contarini’s Compendium of First Philosophy is a philosophical gem. It is composed of seven books, organized according to the doctrine of the transcendentals and a cosmology forged in a theory of the intellect. While the first book deals with “the subject-matter and the epistemological status of first philosophy,”35 books two and three concern “being”, “one”, “true”, and “good”; book four focuses on the “first principles” of being; book five bears upon causality and introduces a sophisticated, though elliptic, doctrine of the intellect which then encompasses the problems of superlunary (book six) and sublunary ontology (book seven).36 It was – somehow unsurprisingly – a reader of Rosmini who drew attention to Contarini’s neglected treatise. Father Carlo Giacon 35. B. BARTOCCI, “Plato’s Parmenides as Serious Game: Contarini and the Renaissance Reception of Proclus,” in: D. CALMA (ed.), Reading Proclus and the ‘Book of Causes’, vol. I, Western Scholarly Networks and Debates, Leiden 2019, pp. 466-481, at p. 473. 36. For the text, I have used Gasparis Contarenis cardinalis Opera, Paris 1571, pp. 96-177, which I checked against two of the nine extant manuscripts, i.e., MS Berlin, Preußischer Kulturbesitz, Lat. fol. 392, ff. 1-83 (hereafter B), and MS Paris, Bibliothèque nationale de France, Lat. 6679, ff. 4v-56r (hereafter P). The critical edition of Contarini’s Opera philosophica selecta, in collaboration with Pietro B. Rossi and Luca Burzelli, is in preparation. The most detailed analysis of this treatise is found in L. BURZELLI, Gasparo Contarini. Per una filosofia dell’unità, Tesi di Perfezionamento, Scuola Normale Superiore, A.A. 20192020, pp. 23-91; at p. 88 Burzelli summarizes: “In his synthesis of the metaphysical thought, both ancient and medieval, one has to give credit to Contarini for attempting to reconcile distinct philosophical traditions: the Aristotelian, centered on substance; the Avicennian, connected to the question of causality; the Neoplatonic, absorbed in the relation of the One with Being” (my translation). See also, relevant for this Note, L. BURZELLI, “Unum indivisum. Il problema dell’Uno da Pico a Contarini,” in: ‘Fructibus construere folia’. Omaggio a Vittoria Perrone Compagni, ed. G. GARELLI – A. RODOLFI, Florence 2020, pp. 161-175.

402

ANDREA ALDO ROBIGLIO

(1900-1984) presented, in 1958, a brilliant paper on the “Avicennian Aristotelianism” of the Venetian thinker.37 While Giacon’s article fell short in explaining the precise ways in which Avicenna might have colored the peculiar brand of Contarini’s Aristotelianism, his capital reference to Ibn Sīnā certainly deserves mention here. Gasparo Contarini, it turns out, was well-read in the Avicenna Latinus. One finds evidence in several of his writings, as Luca Burzelli contends: A quick glance at Contarini’s objections is enough for us to notice that, within an argument that is clearly Aristotelian in structure and, in many aspects, indebted to Aquinas, some characteristic notions of Avicenna’s philosophy come to the surface. That Contarini was well-acquainted with Avicennian thought is a fact demonstrated by numerous textual occurrences, both in the Compendium and in the De elementis.38

To limit oneself to the Compendium, the reader finds thirteen explicit mentions of the name “Avicenna”, whom Contarini praises as a most excellent thinker (praestantissimus philosophus) and “second best” after the ancient Greek masters, Parmenides, Plato, and Aristotle (post priscos excellens autor).39 37. Cf. C. GIACON, “L’aristotelismo avicennistico di Gasparo Contarini,” in: Atti del XII Congresso internazionale di filosofia (Venezia, 12-18 settembre 1958), vol. 9, Florence 1960, pp. 109-119. In the same year of the publication of the Proceedings, Giacon authored a monograph based on the courses he had taught in the late 1950s, viz., C. GIACON, L’oggettività in Antonio Rosmini, Milan 1960, where the name of Contarini did not meet the eye. Before Giacon’s conference paper, one should go back more than seventy years to find some other analysis of the Compendium; cf. F. DITTRICH, Gasparo Contarini 14841542: Eine Monographie, Braniewo 1885, pp. 253-259. 38. See L. BURZELLI, “Aspetti della tradizione aristotelica nel De immortalitate animae: Gasparo Contarini lettore di Avicenna,” in: Rinascimento 59 (2019), pp. 365-390, at p. 368 (my translation). 39. For a guidance to grasp the semantic intricacies of the attribute “ancient” (priscus) and of its uses in Fifteenth century Italian Philosophy, see T. LEINKAUF, “Prisca scientia versus prisca sapientia. Zwei Modelle des Umgangs mit der Tradition am Beispiel des Rückgriffs auf die Vorsokratik im Kontext der frühneuzeitlichen Debatte und der Ausbildung des Kontinuitätsmodell der prisca sapientia, bzw. philosophia perennis,” in: Mediterranea. International journal on the transfer of knowledge 2 (2017), pp. 121-143. In the Compendium, at any rate, the occurrences of the name “Avicenna” distribute as follows: 1 in book two, 2 in book three, 1 in book four, 6 in book five, 2 in book six, and 1 in book seven. The repeated references to Avicenna in book five (where the name of Averroes occurs 4 times as well) is particularly relevant for Contarini’s analysis of the cosmological function of the intellect – an evaluation of that aspect, however, goes beyond the scope of the present note and will be presented elsewhere. Standard reference to Avicenna is that in the second book, used for explaining the distinction between metaphysical and numerical “unity”: CONTARINI, Opera, p. 112; B, f. 19r; P, f. 14r. See BURZELLI, “Unum indivisum,” p. 173 n. 40. The most cited

A NOTE ON THE SOLID CIRCLE

403

The realm of being explored at the end of the Compendium concerns quae infra Lunam agitantur: that is to say, the sensible realm of duality and dialectic, where “there are not simple substances” (nulla […] substantia simplex est), apart from one single exception represented by the form of the human being (intellectus inquam, immaterialis et immortalis).40 It is thanks to the act of intelligence that, despite the opacity of the sensible realities, the great chain of being may unfold, without the risk of dissipation and irremediable loss, from the first being to the abyss of non-being (itaque perfectus est universi ac totius entis ordo, in quo nihil relictum videri potest a primo entium usque ad extremam non entis abyssum).41 The line of degradation from being toward non-being, however, would be dangerously dispersible if a counter-movement, an ἐπιστροφή, would not take place: Verum quoniam [B 81v] omnis substantia secundum id quod est, emanat a primo ente, ac versus non ens tendit; nimirum necesse est si unumquodque pro suo captu perfectum futurum sit, ut etiam unicuique a natura inditae sint vires, quibus rursus convertatur ad ens, a quo est. Hac namque ratione efficitur circulus solidus atque perfectus. Nam processus cuiusque rei ab ente est, veluti linea quaedam recta, quae imperfecta existit sui natura, ideoque si perfecta futura sit, necesse est ut in circulum flectatur. Unumquodque igitur vires quasdam proprias habet, quibus operationem pro sui natura promit, qua maxime Deum aemulatur atque attingit.42

Contarini here introduces the formula “solid circle” which has a distinct cosmological denotation, aside from any gnoseological intention as in Rosmini’s use of it. authorities remain Plato and Aristotle (54 and 51 mentions respectively); Parmenides follows at safe distance: the master of Elea is cited 16 times. 40. CONTARINI, Opera, pp. 169-170; B, f. 79v; P, f. 51v: “Inter substantias inferiores homo supremum locum obtinet, qui suprema sui parte etiam caelestes mentes quadam ratione attingit; huic omnia inferiora deserviunt, ac ut ita /170/ dicam, obtemperant. Eius forma, intellectus inquam, immaterialis est ac immortalis, ut opusculis duobus quibus respondimus excellentissimo huius aetatis philosopho Petro Pomponatio praeceptori nostro, ostendimus.” Contarini refers to his two books On the Immortality of the Soul; see G. CONTARINI, De immortalitate animae/On the immortality of the Soul, ed. P. R. BLUM et al., Nordhausen 2020; L. BURZELLI, “A New Edition of Contarini’s De immortalitate animae,” in: Mediterranea. International journal on the transfer of knowledge 6 (2021), pp. 235242. 41. Ibid., p. 174; B, f. 81r; P, f. 54v. 42. Ibid.; B, f. 81v; P, f. 54v (my italics).

404

ANDREA ALDO ROBIGLIO

However, the energy (virtus) which operates in each sublunary being, as far as its nature is concerned, continues to be attracted by the simple substances. Only additional hindrances may prevent the line to bend toward the first being and, by bowing so, imitate the perfect circle of the divine Being-Intellect. In the next passage, Contarini seems to distinguish – also here, under the Moon – two circles: one of them is closer to the immobile center of being while the other is distant from such a center and runs the risk to go off on a tangent of annihilation.43 However, the circle is as much “solid” as it escapes the fall into nothingness. The non entis abyssus may be avoided by the spark of simplicity which persists even in the realm of dissimilitude and particularization. The human soul, viz. the immortal intellect, is such a spark – like the blowtorch that welds the beginning and the end together. It is possible, thereupon, that one could read into Contarini and trace a promise of “transcendental method” back to him (e.g., the role of intellectual judgment in holding reality together).44 43. Ibid.; B, f. 81v; P, f. 55r: “Nonnullae vero perfectius hoc adsequuntur ac interiorem quendam circulum [P 55r] peragunt valde propinquum entis centro immobili, quaedam vero exteriorem ac longius a centro distantem circulum; imperfectissima vero reflexam quandam lineam aemulantur, quae ad perfectionem circuli minime pervenit. […] Nam cum substantia a primo ente procedat, non videtur commode reflexio fieri posse nisi in substantia sit virtus, per quam promatur operatio. Operatione autem, quae est postremus actus, attingit unumquodque entium primum omnium principium, a quo est, prout est uniuscuiusque earum captus; inferiora vero entia [B 82r] ob imperfectionem multam adiunctam eorum naturae, saepenumero deficiunt ab operatione ac functione ea quae alioqui sibi natura deberetur, constant enim ex materia ac propterea degenerant.” (My italics.) 44. For the oblique implication of the transcendental method, on the one hand, and the “philosophical” history of philosophy, on the other hand, see B. LONERGAN, Method in Theology, ed. R. M. DORAN – J. D. DADOVSKY, Toronto 2017, Ch. 1, § 3-4. Cf. G. GUGLIELMI, “Lonergan’s Theory of Historiography and His Metaphysical Presuppositions,” in: Critical Hermeneutics 3 (2019), Issue 2, pp. 195-222. An interesting aside to my investigation could touch on Rosmini’s knowledge of Avicenna: a topic on which I have been unable to find any literature available. In his Aristotele esposto ed esaminato (1853), besides the intensely explored commentaries of Averroes, there is little evidence that Rosmini conducted any first-hand reading of Avicenna: cf. A. ROSMINI, Aristotele esposto ed esaminato, ed. G. MESSINA, Stresa / Rome 1995, n° 37, pp. 72-73 and especially n° 109, note 195, p. 160. When we open the Teosofia, a few references seem to attest to Rosmini’s encounter with the Avicenna Latinus; any conclusive evidence, however, is missing. Those references could easily derive from Aquinas, Albert, or other Modern authors. For instance, a passage like the following, “[…] that which is first conceived by the intellect as most known and to which all concepts are reduced is ens, as Avicenna says at the beginning of the Metaphysics” (Teosofia, pp. 287-288 – Theosophy, Part I, Ch. 5,

A NOTE ON THE SOLID CIRCLE

405

In the travels of a geometric metaphor, from Gasparo Contarini to Giovanni Gentile passing through Antonio Rosmini, one may find a fresh motivation for a philosophical research of some breadth.45 The history of philosophy, in this sense, would not disjoin from the actual act of philosophizing. Andrea Aldo ROBIGLIO De Wulf Mansion Centrum, Hoger Instituut Wijsbegeerte – KU Leuven Kardinaal Mercierplein 2 B-3000 Leuven [email protected]

n° 58, p. 55) translates a sentence from Aquinas’s first question De veritate, art. 1: “ut Avicenna dicit in principio suae Metaphysicae”. 45. A model for this type of approach, which could be called the micro-history of philosophy, is encountered in the fascinating research of Dietrich Mahnke devoted to the geometric metaphor of the “infinite sphere”: D. MAHNKE, Unendliche Sphäre und Allmittelpunkt. Beiträge zur Genealogie der Mathematischen Mystik, Halle (Saale) 1937. The author suggestively indicates the inspirational use of projective geometry for the metaphysical analysis, as in the case of Avicenna’s “Sphärenemanationslehre […] mit plotinischen und aristotelischen Begriffen ausgebaut” (p. 208). This approach does not yet exclude the “systematic” dimension of philosophy, since the latter remains the condition of the possibility of the former. We owe to a distinguished historian of philosophy, the late Vittorio Mathieu (1923-2020), a sophisticated and witty treatise of ontology written in the spirit of Mahnke: V. MATHIEU, Trattato di ontologia. Essere e spazio, Milan 2019. In the imposing trilogy by P. SLOTERDIJK, Sphären, Berlin 2004, on the contrary, one could hardly find anything useful for the present investigation – these are rather “things that here are best unsaid.”