Aristotle’s "Metaphysics" Lambda – New Essays 9781501503443, 9781501510915

Citation preview

Aristotle’s Metaphysics Lambda – New Essays

Philosophie der Antike Veröffentlichungen der Karl und Gertrud Abel-Stiftung Herausgegeben von Wolfgang Kullmann in Verbindung mit Jochen Althoff und Georg Wöhrle Band 33

De Gruyter

Aristotle’s Metaphysics Lambda – New Essays edited by Christoph Horn Proceedings of the 13th Conference of the Karl and Gertrud-Abel Foundation Bonn, November, 28th –December 1st, 2010

De Gruyter

ISBN 978-1-5015-1091-5 e-ISBN (PDF) 978-1-5015-0344-3 e-ISBN (ePUB) 978-1-5015-0334-4 ISSN 0943-5921 Library of Congress Cataloging-in-Publication Data A CIP catalog record for this book has been applied for at the Library of Congress. Bibliographic information published by the Deutsche Nationalbibliothek The Deutsche Nationalbibliothek lists this publication in the Deutsche Nationalbibliografie; detailed bibliographic data are available on the Internet at http://dnb.dnb.de. © 2016 Walter de Gruyter Inc., Boston/Berlin Typesetting: Meta Systems Publishing & Printservices GmbH, Wustermark Printing and binding: Hubert & Co. GmbH & Co. KG, Göttingen ♾ Printed on acid-free paper Printed in Germany www.degruyter.com

Contents List of Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

vii

Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

xi

CHRISTOPH HORN Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1

MATTEO DI GIOVANNI and OLIVER PRIMAVESI Who Wrote Alexander’s Commentary on Metaphysics Λ? New Light on the Syro-Arabic Tradition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

11

ENRICO BERTI The Program of Metaphysics Lambda (chapter 1) . . . . . . . . . . . . . .

67

CHRISTOF RAPP The Principles of Sensible Substance in Metaphysics Λ 2–5 . . . . . . . .

87

MICHEL CRuBELLIER What the Form Has to Be and What It Needs not Be (Metaphysics, Λ 3) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

119

MARCO ZINGANO Individuals, Form, Movement: From Lambda to Z–H . . . . . . . . . . .

139

STEPHAN HERZBERG God as Pure Thinking. An Interpretation of Metaphysics Λ 7, 1072b14–26 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

157

vi

Contents

SILVIA FAZZO Unmoved Mover as Pure Act or Unmoved Mover in Act? The Mystery of a Subscript Iota . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 181 ALBERTO ROSS The Causality of the Prime Mover in Metaphysics Λ . . . . . . . . . . . .

207

MARIA LIATSI Aristotle’s Silence About the Prime Mover’s noe¯sis . . . . . . . . . . . . . .

229

ISTVÁN BODNÁR Cases of Celestial Teleology in Metaphysics Λ . . . . . . . . . . . . . . . . .

247

CHRISTOPH HORN The Unity of the World-order According to Metaphysics Λ 10 . . . . .

269

Indices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

295

1. Index locorum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2. Index nominum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3. Index rerum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

295 305 309

List of Contributors Enrico Berti is Professor emeritus of Philosophy at the Università di Padova. In his scholarly career, he published dozens of monographs on Aristotle, including La filosofia del primo Aristotele (1962), Aristotele. Dalla dialettica alla filosofia prima (1977) and Dialectique, physique et métaphysique. Études sur Aristote (2008). István Bodnár is Professor of Philosophy at Eötvös University, Budapest, teaching also at Central European University. As author, editor and translator he published several monographs, for example Eudemus of Rhodes (with W. W. Fortenbaugh, 2002) and Simplicius: On Aristotle Physics 8.1–5 (with M. Chase and M. Share, 2014). Michel Crubellier is Emeritus Professor of Ancient Philosophy at Université Lille 3 and the former Vice-President of the University. His research bears mainly on Aristotle’s philosophy of knowledge, logic, psychology and natural philosophy. He has published Aristote: Le philosophe et les savoirs (with P. Pellegrin, 2002), Dunamis: Autour de la puissance chez Aristote (2008), Aristotle’s Metaphysics Beta: Symposium Aristotelicum (with A. Laks, 2009), and a French translation with commentary of the Prior Analytics (2014). Matteo Di Giovanni is Assistant to the Chair of Late Ancient and Arabic Philosophy at the Ludwig-Maximilians-Universität of Munich. He recently earned his Ph.D. in Graeco-Arabic and Islamic studies from Yale University with a dissertation entitled “Aristotle’s Metaphysics, Book Alpha Minor: Studies in the History of the Tradition and Critical Editions of the Greek and Arabic Texts.” His research focuses on both the textual and the philosophical tradition of Aristotle in Islam. Recent publications include The Commentator. Averroes’s Reading of the Metaphysics (2014), Averroes and Philosophy in Islamic Spain (2012), and Motifs of Andalusian Philosophy in the PreAlmohad Age (2011). Silvia Fazzo (Università della Calabria)’s writings mainly deal with Aristotle’s Metaphysics and the Aristotelian tradition, including philological reconstruction of Greek philosophical texts. See among her volumes Il libro Lambda della Metafisica di Aristotele (2012), Commento al libro Lambda della Meta-

viii

List of Contributors

fisica di Aristotele (2014), Aporia e sistema… nelle Quaestiones di Alessandro di Afrodisia (2002) and Alessandro di Afrodisia, La provvidenza (1999); among others: Aristotelianism as a commentary tradition, The Metaphysics from Aristotle to Alexander of Aphrodisias (2012), Heavenly matter in Aristotle, Metaphysics Lambda 2 (2013), and Towards a Textual History and Reconstruction of Alexander of Aphrodisias’s Treatise On the Principles of the Universe (with Mauro Zonta, 2014). Stephan Herzberg is Lecturer in Philosophy at the Philosophisch-Theologische Hochschule Sankt Georgen (Frankfurt). He mainly works on ancient and medieval philosophy, particularly in Aristotle. His publications include Wahrnehmung und Wissen bei Aristoteles. Zur epistemologischen Funktion der Wahrnehmung (2011) and Menschliche und göttliche Kontemplation. Eine Untersuchung zum bios theoretikos bei Aristoteles (2013). Christoph Horn is Professor of Ancient and Practical Philosophy at the University of Bonn and editor of the journal Archiv für Geschichte der Philosophie. He is author of Plotin über Sein, Zahl und Einheit (1995). Augustinus (1995), Antike Lebenskunst (1998), Politische Philosophie (2003), Philosophie der Antike (2013) and Nichtideale Normativität (2014). Maria Liatsi is Professor of Classics (Ancient Greek Philosophy) at the University of Ioannina, Greece. Her main subjects are Plato, Aristotle, Hellenistic philosophy, the reception of ancient Greek philosophy in mediaeval and modern Times. For examples of her works see Aristoteles’ De generatione animalium, Buch V. Einleitung und Kommentar (2000), Interpretation der Antike. Die pragmatistische Methode historischer Forschung (2006), Die semiotische Erkenntnistheorie Platons im Siebten Brief. Eine Einführung in den sogenannten philosophischen Exkurs (2008) or Der Gedanke der irdischen Unsterblichkeit in der Antiken Philosophie. Zur Metaphysik des Lebens in der Antike (working title, forthcoming). Oliver Primavesi holds the chair of Greek (I) at the University of Munich (LMU). His research has been focused mainly on Aristotle and the Presocratics, in particular on Empedocles. His books include Die Aristotelische Topik (1996) and Empedokles: Physika I (2008). He has co-authored, with A. Martin, L’Empédocle de Strasbourg (1999) and with J. Mansfeld, Die Vorsokratiker Griechisch / Deutsch (2011), he has edited, with K. Luchner, The Presocratics from the Latin Middle Ages to Hermann Diels (2011), and he has contributed a new critical edition of Metaphysics Alpha Major to the Symposium Aristotelicum volume devoted to that book (2012).

List of Contributors

ix

Christof Rapp is Professor of Ancient Philosophy at Ludwig-MaximiliansUniversität München and the editor of several journals, for example Phronesis. Although he has published on a wide spectrum of topics, a special focus is on Aristotle, including Identität, Persistenz und Substantialität. Untersuchung über das Verhältnis von sortalen Termen und Aristotelischer Substanz (1995) and Aristoteles zur Einführung (42012). Alberto Ross is Professor of Metaphysics at the Universidad Panamericana, Mexico. His research is focused on ancient philosophy, especially Aristotle and his reception in late antiquity. His works include Dios, eternidad y movimiento en Aristóteles (2007), Filópono y el Pseudo Justino contra la eternidad del movimiento (2010), The causality of the Prime Mover in Simplicius (forthcoming 2015) and Causality, nature and fate in Alexander of Aphrodisias (forthcoming 2016). Marco Zingano is Professor of Philosophy at the Universidade de São Paulo (USP). He mainly works on ancient philosophy, and takes a special interest in Aristotle. Some of his publications are Studies on Ancient Ethics (2005) and Nicomachean Ethics I 13–III 8: Treatise on Moral Virtue (2008).

Preface This volume contains the contributions of the 13th conference of the Karland-Gertrud-Abel foundation, held at the university of Bonn in 2010 (November, 28th – December 1st). I wish to thank the foundation for the generous financial support of this scholarly meeting which took place in a remarkably dense and fruitful atmosphere. The conference provided an excellent chance to discuss highly diverging views on one of the most interesting and most difficult Aristotelian texts. It is a great pleasure for me to present here the results of our discussions. The publication has also been facilitated by the foundation. I want to express my gratitude to the authors for their valuable papers as well as to the editors of the book series for their patience. Special thanks to Malte Kuhfuss, Julia Petz, and Henning Könitz for all the editorial work they did, especially on the Indices. Bonn, December 2015

Christoph Horn

Introduction CHRISTOPH HORN

Aristotle’s treatise known as book XII or Lambda of the Metaphysics is not an organic part of this extensive work. It comprises a self-contained discussion of substances, causes, principles, and motions. At the beginning of the treatise, Aristotle says that substance (ousia) is the topic of the text, and then he provides a coherent line of argument. In chapter 1, he differentiates between three types of substances, (a) the terrestrial, (b) the celestial and (c) the immaterial ones and characterizes them (a) as being moved (= changeable) and perishable, (b) as moved but imperishable, and (c) as unmoved (= unchangeable) and imperishable. The terrestrial (or sublunary), moved and perishable substances are discussed in chapters 2–5, whereas the eternal ones appear in chapters 6–10. In this second part, we encounter, on the one side, the celestial bodies, i.e. the fixed stars, the planets, the sun, and the moon and, on the other side, immaterial divine substances, namely the Unmoved or Prime Mover and 55 further immaterial movers. Concerning the causality that is at work in the sublunary world, Aristotle defends the idea that causality is relative to individual cases. There is ‘no general man’ bringing about each instantiation of men, no Platonic ‘man-initself’, but it is “Peleus who begot Achilleus and your father you” (5, 1071a22). For Aristotle, it is the father who generates the son, not the divine Prime Mover. At first glance, it seems to follow from this that, for Aristotle, there is no overarching causality. Causal relations have to be located within a species or as a transfer of forms by a techne¯: one can call this the ‘homoeidetic’ concept of causation. A human being begets a human being (anthro¯pos anthro¯pon genna: 1070a8.27–8), or a doctor applies his medical knowledge as the form of health to his patient. The term homoeide¯s is in the Lambda explicitly used in ch. 5, 1071a17. It refutes the idea that universal causes are at work. For Aristotle, some F is brought about by an entity which possesses F in a synonymous way, not, like a Platonic form, homonymously. Aristotle thinks that homoeidetic causality is sufficient to explain how individuals are brought about. A further aspect of Aristotle’s theory of causation in the first chapters of the Lambda is the description of elements, causes, and principles in ch. 4. There, he tells us that we should assume the existence of three elements

2

Christoph Horn

(stoicheia), and of four causes and principles (aitiai kai archai) (1070b25–6). The elements are form, privation, and matter (1070b18–9), and the same are also causes and principles, plus the efficient cause. But their generality as elements, causes, and principles is only by analogy. For Aristotle, it is not the case that the same entities, e.g. again Platonic Forms, can be taken as general principles of everything. As he emphasizes, it would be absurd (atopon) to assume that elements, causes and principles are numerically identical, but only analogically; between the entities in the world, there is nothing common (ouden estin koinon) except the structure of substance and the categories (1070a35–b4). The relation of this position with the rest of the Metaphysics is notoriously unclear. Of particular interest is how close it is to or how far it is from Met. Z–H. If there exist substantial differences (and this is quite unambiguously the case), then the question arises how it should be related to Z–H and whether the differences can be traced back to a developmental perspective: perhaps the Lambda is an early text of Aristotle (as Werner Jaeger claimed) or rather a late one. In the history of its reception, it was especially its second part, the chapters 6–10, which attracted the attention of the readers since it develops a view of God as Prime Mover, on the origin of cosmic motion, on the noetic activity of God, and on the divine world-order. But it must not be neglected that, primarily, the treatise is on substances and its principles. Prima facie, the notion of homoeidetic, synonymous, and analogous causality seems sufficient to explain the generation of individual beings. But this is mistaken; according to Aristotle, it does not suffice to explain how the everlasting process of causation in an imperishable universe is going on. The fact that celestial bodies are in eternal movement as well as the fact that, in the sublunary world, an eternal process of emerging and deceasing is happening, needs some further explication. To give an account for this requires additionally the assumption of an entity that is in eternal actuality. Aristotle describes the sun and the Prime Mover in Lambda 5 as parallel in their function as overarching causes, the first for the sublunary world, the second for the entire cosmos. One of the most controversial issues regarding book Lambda concerns the question whether (and if yes, to what extent) the book contains a theology. The older tradition – i.e. the commentaries of Alexander of Aphrodisias, Themistius, Simplicius, Ibn Rušd (Averroes) and Thomas Aquinas – understood the treatise in this sense. Those who are against the description of the Lambda as a theological treatise emphasize that the text is mainly about substances, principles, causes, motions, and celestial bodies; it has no close connection with popular or narrative religion as it was practised at Aristotle’s time. This is clearly correct; and it would also be completely mistaken to expect, within the book, pronouncements on providence, revelation, salva-

Introduction

3

tion, and the right sort of piety; these issues are rather typical for a Platonist or a Judaeo-Christian-Islamic context. There are, however, at least seven points in our treatise which merit to be called ‘theological’: (i) Aristotle denominates the Prime Mover, the noetic principle of the universe, explicitly as ‘God’ (1072b30), (ii) he provides an argument for the existence of this God (6, 1071b12–22), (iii) he develops an extensive list of ‘divine attributes’,1 (iv) he explains how God, as an unmoved entity, moves the celestial bodies, (v) he describes the way of life lead by God (7, 1072b13–30), (vi) he specifies the activity practised by God, i.e. the thinking of his thinking (9, 1074b28– 35) and (vii) he characterizes the universe as a well-ordered unity structured by God (10, 1075a11–25). Another crucial question debated in recent scholarship is that on the causality executed by the Prime Mover: is it efficient or final causality? Whereas Gregory Vlastos (1963/21995) claimed that we can exclude efficient causality, the efficient-causality reading has been defended by Sarah Broadie, Enrico Berti, and Lindsay Judson. Aristotle claims that “the heavens and the nature depend on such a principle”, i.e. on the Prime Mover (ek toiaute¯s ara arche¯s e¯rte¯tai ho ouranos kai he¯ phusis: 1072b13–4). The easiest way of understanding this is in terms of a minimal causal effect brought about by the First Mover; one might call this the domino-effect interpretation. It goes as follows: by exerting an impact (efficient causality) on the outermost celestial sphere, the Prime Mover indirectly causes the movement of all other spheres and finally also sets all sublunary things in motion. This line of interpretation, however, seems correct, but insufficient. Although it is right to say that the sphere of the fixed stars is set in motion by the impulse of the Prime Mover, and despite the fact that, according to Lambda 8, the motion of one sphere influences the motion of the next one, the role of the Prime Mover would then not be a prominent one: the Prime Mover only gave the first impulse, but this is only one of 56 impulses of the same kind. But in fact,

1

The Prime Mover is characterized by the attributes of being in pure act or actuality (energeia: 1071b20; 1072a25; 1072b8) and eternal (aidios: 1072a23 ff.). The Unmoved Mover is substance (ousia: 1072a25; 31), more precisely the first (pro¯te¯) and simple substance (haple¯); he is the most eminent object of thought (1072a30–33). He is without matter and hence without change and motion (einai aneu hule¯s: 1071b21). He is invariant and excludes every alteration (touto ouk endechetai allo¯s echein oudamo¯s: 1072b8). Moreover, he exists by necessity (ex ananke¯s ara estin on: 1072b10); Aristotle even says that the Unmoved Mover is simply (haplo¯s: 1072b13) by necessity. The Prime Mover is principle, even the first principle (arche¯: 1072b11; he¯ arche¯ kai to pro¯ton: 1073a23). He is without parts and indivisible (amere¯s kai adihairetos: 1073a6 f.). He leads the best life (zo¯e¯ ariste¯: 1072b28). He is unaffectable (apathes: 1073a11; 1074a19) and separated from the sensible things (kecho¯rismene¯ to¯n aisthe¯to¯n: 1073a3–4).

4

Christoph Horn

Aristotle ascribes an outstanding role to the Prime Mover. One should at least add something to the domino-effect interpretation, and for this, one should take a closer look at the traditional finalistic interpretation. One argument against this reading is that it seems unable to explain how non-psychic entities (the celestial spheres) should strive for the Prime Mover motivated by the love for it. Despite the critical discussion of this point and many others within recent Aristotelian scholarship the traditional reading is still the standard interpretation of what the causality of the Prime Mover consists of. Its defenders insist on the fact that the Prime Mover moves by attraction, as Aristotle clearly indicates by his famous formula kinei de¯ ho¯s ero¯menon (1072b3; cf. E. E. 1249b14 ff.). In a sense, however, the two interpretations, based on efficient and on final causality respectively, do not principally rule out one another. The scholarly research on the Lambda is currently more intense than ever before, and not only on this question. Starting with the volume edited by Michael Frede and David Charles (2000) the book has regained the status of an important topic of Aristotelian scholarship. The present volume is intended as a contribution to this ongoing debate. In their highly innovative article, Matteo Di Giovanni and Oliver Primavesi discuss the introduction to Ibn Rušds (Averroes’) commentary on the Aristotelian Metaphysics Lambda (1393,4–1405,12 Bouyges). This introduction contains a complete outline of Aristotle’s Metaphysics. Who wrote this outline? Whereas most of the interpreters rejected the idea that we can ascribe the paraphrase to Alexander of Aphrodisias (starting with J. Freudenthal 1885) and think that it derives from thy Syro-Arabic context, the two scholars now come to the conclusion that the text belongs to the ancient tradition. Ibn Rušd himself traces the material back to ‘Alexander’, and there is strong evidence for this. Even if the text which we now possess is not identical with that written by Alexander of Aphrodisias (but is a revised version of Alexander’s writing), it can be shown, as Di Giovanni and Primavesi believe, that it has genuinely ancient fundaments, namely in Ammonius and Asclepius. They examine the historical problem on the basis of Freudenthal’s six reasons advanced in favour of his decision. An important issue of their article concerns the order of the books of the Metaphysics, as the outline presents it: the Alpha Minor comes first, only then the Alpha Maior follows. While most modern scholars regard this as an Arabic peculiarity, Di Giovanni and Primavesi think that this sequence goes back to late antiquity, more precisely to the Neoplatonic school of philosophical teaching. There are remarkable parallels with the oral presentation of the Metaphysics by Ammonius, as we know from his collaborator Asclepius. Enrico Berti’s contribution is based on a close reading of chapter 1; it has the character of a dense linguistic and philosophical commentary. One

Introduction

5

of the merits of this article is that it takes ancient and medieval views on the chapter into consideration, namely those of Ps.-Alexander, Themistius, Ibn Rušd, and Thomas Aquinas. Berti’s topic is the ‘program’ of Metaphysics Lambda as it is developed in its first lines. He starts with considerations on the Incipit of the book. Reconsidering his own article from 2003, he takes the word theo¯ria in the sense of a ‘research into the principles and causes’ and emphasizes that this is in congruence with Met. Alpha Elatton. In both of these texts, philosophy is practiced, as Berti notes, as a ‘Prinzipienlehre’. In the following sentence of the chapter, he sees Aristotle explaining why this theo¯ria is about substance. While substances exist in an unqualified way, their attributes are dependent on substances. The way in which Aristotle describes the primacy of substance here is, according to Berti, close to the Categories, but does not presuppose Met. Zeta. He then turns to the question if Aristotelian physics is meant to be the science of all substances, or only of the moved ones. Berti claims that Aristotle opts in favour of a unique science covering both moved and unmoved substances. From all of this, he concludes that the Lambda and the Alpha Elatton are earlier than the Epsilon, the Gamma, and the Zeta. To some extent at least, Berti comes back, in the end, to Werner Jaeger’s idea that the Lambda is part of an ‘Urmetaphysik’. Christof Rapp gives a detailed reconstruction of the line of argument advanced in Lambda 2–5. The entire passage is about the principles of sensible substances. Aristotle starts with the straightforward statement that there are three principles: privation, form, and matter. Then he adds a fourth cause: the external mover. In the discussion of this quadruple, Aristotle introduces the synonymy principle (claiming that each ousia proceeds from something having the same name) and the distinction between pre-existing and simultaneously existing causes. There are both inherent principles (i.e. elements) and external ones. Aristotle claims that, in a sense, all things have the same principle, while, in another sense, they don’t. All things have the same principle, as he says, analogically. This means that the three elements of things as well as the external cause are different in various genera, but they exert the same sort of causality. And that implies that privation, form, and matter are the same because these principles are mutually related as potential and actual beings. Aristotle then brings in the ‘coincidence assumption’ which supposes that the external moving cause and the form might coincide. At the end of Lambda 4, Aristotle quite abruptly introduces a further external moving cause that is called ‘the first’. According to Rapp, this attribute cannot signify ‘immediate’ or ‘proximate’ cause since it is characterized by Aristotle as the ‘first of everything’. This first mover must hence be a non-synonymous cause, and Rapp argues that it hints at the Prime Mover as discussed in Lambda 6– 10. The causes of substances are the causes of non-substantial beings since substances are the causes of everything else. However, the famous formula

6

Christoph Horn

‘man begets man’ (anthro¯pos anthro¯pon genna) is intended to exclude the Platonic Form of man, the man-in-itself. On the other hand, Aristotle is not making a nominalist point claiming that only particulars can be proper causes. Michel Crubellier raises the question whether Lambda 3 is a mere résumé or short version of the line of thought known from Z 7–9 regarding the synonymy principle. He rejects this view and points out that chapter 3 fits very well into the general argument developed in the Lambda. Crubellier provides a close commentary on Lambda 3 and discusses, in the end, some textual problems. The general task Aristotle is following in the first chapters of the Lambda is to give a theory of change and motion. By contrast with chapter 2 which concentrates on the question of what is going on in the thing that changes, Lambda 3 introduces the notion of cause; additionally, Lambda 2 discusses all sorts of alteration while chapter 3 is restricted to the generation of substance. Crubellier then proposes a reading of the text that takes as its starting-point the synonymy principle. He assumes that Aristotle considers this principle, introduced in Z 7–9, as problematic. Crubellier believes that Lambda 3 is intended to reject the introduction of the theory of Forms in Plato’s Phaedo. While Plato uses this idea without restriction in his ‘intellectual autobiography of Socrates’, Aristotle confines the synonymy principle to the generation of substances. Concerning the problem of how luck and chance are compatible with the synonymy principle, Crubellier points out that the cases Aristotle refers to are similar to techne¯-based causation: a patient can e.g. regain his health by running and thereby producing exactly the amount of heat a doctor would order on the basis of his medical competence. The paper of Marco Zingano gives a general characterization of the philosophical project exposed in Met. Lambda. Zingano thinks that the theory of substance, as developed in the Lambda, should be located halfway between the two positions represented by the Categories, on the one hand, and by Met. Z–H, on the other. In the Categories, Aristotle investigates sensible substances without taking non-sensible substances into account, whereas in the Lambda, he discusses both. The Lambda mainly looks for the common principles and causes of sensible substances; these principles (namely, as we saw, matter, form, privation, and the external mover) are the same for all things merely by analogy. Additionally, Zingano emphasizes the importance of the dunamis-energeia distinction for the second part of the treatise (chs. 6–10). The Lambda is thus different from the ontology of individuals as evolved in the Categories, but it also differs from the perspective of Z–H: Zingano gives an interpretation of Lambda 3, 1070a9–13 which is directed against the readings of Ross and Jaeger and follows the text of the manuscripts (with Bekker’s punctuation). Contrary to the view of L. Judson (2000)

Introduction

7

he believes that, in this passage, Aristotle distinguishes three kinds of substances, matter, nature or form, and individuals such as Socrates or Callias which are composites of form and matter. According to Zingano, we should resist the temptation to conflate what is said here with the theory presented in Z–H, especially the form-division. The Lambda does also not contain the distinction between ousia and ousia of X. In the Lambda, as Zingano finally points out, the non-sensible substances are taken as unmoved movers, while in Z–H, unmoved movers turn out to be separate unchanging substances. In his article, Stephan Herzberg raises a crucial question regarding Lambda 7: how close is a human being, when practising theo¯ria, to the God who is permanently contemplating himself? While humans, following this text (and, in more detail, E. N. X 7–9), are dedicated to theo¯ria no more than for a short period of time, the Aristotelian God, by thinking himself, is realizing his essential activity. Is there only a gradual or also a principal difference between these two forms of thinking? Does the difference merely concern time, or is there a substantive dissimilarity between these two forms of noetic activity? Herzberg distinguishes three possible interpretations: (a) According to Alexander of Aphrodisias, God himself is thinking within us when we are theorizing; the active intellect of De anima III 5 is thus regarded as identical with the divine nous of the Lambda, (b) our thinking is not identical with the divine one, but equals it in species or genus, (c) our thought and that of God are more loosely correlated, namely in the sense of a pros hen-relation, i.e. our thought is directed towards divine thinking without immediately being identical with it and without being equal with it in species or genus. Herzberg argues in favour of option (c); he gets his most important evidences from a diligent interpretation of Met. Λ 7, 1072b14–26 (while the passage E. N. X 8, 1178b22–31 gives us insufficient information to clarify this problem). Herzberg rejects (b), defended by Ch. Kahn (1992), by showing that the similarity between our theo¯ria and divine noe¯sis rests only upon one aspect, namely that of pure energeia. Human thinking is primarily directed towards external objects of thought and is only partly connected with the kind of thought which is pure activity and actuality. In her contribution, Silvia Fazzo challenges the traditional reading of Λ 7, 1072a26; this is the passage where Aristotle characterizes the Unmoved Mover by the term energeia. As she claims, the reception of the text suffers from a mistake which might have occurred in late antiquity when scribes and scholars became more and more unfamiliar with the scriptio continua and the uncial script. At that point, the reading of energeia(i) with a subscript iota might have been altered by leaving out the iota, thereby changing the text from a dative into a nominative form. Her new critical edition (Fazzo 2012; see also 2014) is the first text since R. Torstrick in the 19th century which deviates from this reading. The (alleged) mistake might go back to

8

Christoph Horn

Alexander of Aphrodisias (Quaestio I.25) who, as Fazzo conjectures, was favourable to the idea to equate the Prime Mover and divine intellect with the form and with actuality. At any rate, the effect of this small alteration is enormous: if Fazzo is right, the original meaning ‘the Prime Mover is in act’ was changed to ‘the Prime Mover is (pure) act’. Fazzo thinks that the equation of the Unmoved Mover with energeia can be shown to be un-Aristotelian on the basis of parallel passages; she explains the interest of later recipients which lead to this misreading by the Neoplatonizing atmosphere in which the Lambda was interpreted. In later ancient philosophy, Plotinus’ treatise VI.8 [39] 16 might be seen as one of the reference passages for this reinterpretation. The doctrine of the Prime Mover as pure act is “a main tenet of Aristotelianism in modern accounts”, as Fazzo says. The philosophical point is: Aristotle conceives the Prime Mover as in energeia, not as the energeia in itself; but despite their mistaken reading, modern interpreters normally understand the text in the first sense, not in the second. In his article, Alberto Ross defends the traditional reading of Metaphysics Lambda according to which the causality of the Prime Mover has to be understood in the sense of final causality. He thereby defends a tradition that goes back to Alexander of Aphrodisias who took the famous formula kinei de ho¯s ero¯menon (1072b4; cf. E. E. 1249b14) in the sense of a teleological relation. Ross rejects the interpretations advanced during the last fifteen years which favour a reconstruction in terms of efficient causality, formulated (as mentioned above) e.g. by Sarah Broadie or by Enrico Berti. In his text, Ross gives a precise reconstruction of the traditional reading and then reports the objections that have been raised against it. They are based, inter alia, on observations concerning the words kine¯tikon kai poie¯tikon (6, 1071b12) which seem to characterize the Prime Mover’s activity as efficient causality, on the formulation ho¯s ero¯menon which seems to imply the anti-realist meaning of “as if”, and on the fact that the celestial spheres, following the traditional interpretation, would have to be living beings in order to be capable to strive for the First Mover. Ross then tries to refute these (and many more) arguments; according to him, there are even fewer textual tensions when we accept the teleological reading than by adopting the more recent one. Maria Liatsi discusses in her article an old, but still unresolved question: what is Aristotle’s Prime Mover thinking about? During the long history of reception of Lambda ch. 7, many readers of the text felt uncomfortable with the fact that the Unmoved Mover is said to think himself. Even if he is thereby thinking the most valuable cognitive object, this implies a problematic characterization of God. Some interpreters even spoke of a ‘Narcissus problem’ in Lambda 7. Thomas Aquinas, e.g., tried to overcome the difficulty by the assumption that God thinks ‘everything’ by thinking himself (cf. intelligendo se intelligit omnia alia and deus cognoscendo seipsum omnia

Introduction

9

cognoscit).2 Liatsi rejects this reading as well as the interpretation advanced by M. Bordt (2006) who completely denies the self-referentiality of the Prime Mover’s thought. She considers four possible options to understand the Aristotelian silence: (1) Lambda is no more than a short draft for lectures; Aristotle added his full opinions orally. (2) Lambda is incomplete: the text is a fragment. In both of these cases (1–2), the lack of Aristotelian statements would be merely contingent. (3) Aristotle did not want to openly present his opinions (reservatio mentalis). (4) Lambda, in fact, contains the full intentions of Aristotle; but modern readers normally misunderstand them. According to the assumptions (3–4), Aristotle deliberately remained silent. Liatsi goes for option (4) and emphasizes that we should respect, as a matter of fact, that Aristotle held his peace on the content of the Prime Mover’s noe¯sis. The Aristotelian deity is not a personal God, and we should avoid an Interpretatio Christiana. If one takes this silence seriously then, according to the author, we have to conclude that Aristotle said everything he was able to say – and there is no deficiency left. In his contribution, István Bodnár discusses some examples of celestial teleology which he calls ‘lesser instances’. His considerations are particularly based on De caelo II 12. Bodnár is especially interested in the imitation of eternal celestial motion by inanimate elements, and in the striving of living beings for eternal and divine existence. With the help of these cases, Bodnár intends to shed some light upon the famous Aristotelian claim that the Prime Mover is that for the sake of which the outermost sphere moves. These minor cases are characterized by the fact that the goal of their activity is within the celestial realm itself. Bodnár first points out the similarities and differences between Aristotelian ‘unmoved movers’ such as soul of individuals, arts and crafts, on the one hand, and the supra-physical, incorruptible substances of the celestial domain, on the other. The main point is that, in all cases, an unmoved mover is that for the sake of which the moved entity moves. Given that celestial motions are continuous and eternal locomotion has no startingpoints or end-points, the explanation Aristotle has in mind is based on final causality: all unmoved movers are said to be in a state of excellence that makes them desirable or lovable. There is some congruence, as Bodnár points out, between the unmoved mover and the celestial sphere that is moved – even if it is not called ‘imitation’. The expression ‘for the sake of’ is used not in the sense of a beneficiary, but of a goal; the beneficiary is the moved entity. But in the end of his paper, Bodnár also discusses the case that there are, in the celestial region, goals as beneficiaries. As this discussion shows, Aristotle’s celestial domain is a well-structured, purposeful, teleological order. 2

Sententia super Metaphysicam, XII lectio 11, 2614–5 Marietti.

10

Christoph Horn

Finally, my contribution The Unity of the World-order According to Metaphysics Λ 10 examines the question in which sense Aristotle describes the universe as a unified and well-ordered whole. According to Lambda 10, the order of the cosmos is that of an organized totality, not a merely contingent cluster. As Aristotle claims, the world shows arrangement and interconnection: the higher an entity, the more is its conduct determined by regularity. The cosmos thereby possesses, he says, “the good and the best”, and this is traced back to the fact that all entities do not exist independently from one another, but in a close interrelation. Moreover, we learn how this interrelatedness has to be understood: namely in the sense that everything is directed towards one centre (pros men gar hen to panta suntetaktai: 1075a18–9). The order goes back, according to Aristotle, to a divine entity which is at the centre (or at the top) of the universe (even if of course not in a spatial sense). This deity is compared with the general of an army, and briefly later it is equated with a king. I defend a version of a teleological-theological interpretation which I call ‘divine-design reading’. I attempt to explain how Aristotle justifies the idea of an overarching causality in the universe. Moreover, I offer a possible solution to the problem of how Aristotle’s account of cosmic order and unity can be reconciled with his general idea of teleology. Finally, I try to clarify the sense in which the entities in Aristotle’s universe are interconnected by pointing out that his teleology is a version of perfectionism.

Bibliography Bordt, M.: Aristoteles’ Metaphysik XII, Darmstadt 2006. Fazzo, S.: Il libro Lambda della Metafisica di Aristotele, Napoli 2012. Fazzo, S.: Commento al libro Lambda della Metafisica di Aristotele, Napoli 2014. Frede, M./Charles, D. (eds.): Aristotle’s Metaphysics Lambda. Symposium Aristotelicum, Oxford 2000. Judson, L.: Formlessness and the Priority of Form: Metaphysics Zeta 7–9 and Lambda 3, in: M. Frede/D. Charles (eds.), Aristotle’s Metaphysics Lambda. Symposium Aristotelicum, Oxford 2000. Kahn, Ch.: Aristotle on Thinking, in: M. C. Nussbaum/A. Oksenberg Rorty (eds.), Essays on Aristotle’s De Anima, 1992, 359–379. Vlastos, G.: A Note on the Unmoved Mover, in: Philosophical Quarterly 13, 246–7 (id., Studies in Greek Philosophy, vol. II, Socrates, Plato, and their Traditions, Princeton 1963/21995, 283–4).

Who Wrote Alexander’s Commentary on Metaphysics Λ? New Light on the Syro-Arabic Tradition 1

MATTEO DI GIOVANNI and OLIVER PRIMAVESI

I. The Problem Volume I of the series Commentaria in Aristotelem Graeca, edited in 1891 by Michael Hayduck, bears the title Alexandri Aphrodisiensis in Aristotelis metaphysica commentaria. Yet this title does not quite correspond to the contents of the volume as presented by the editor himself. For right at the beginning of the praefatio Hayduck introduces the general contention that only the commentaries on the first five books (Α ‒ α ‒ Β ‒ Γ ‒ Δ), which fill pages 1‒439 of his edition, were written by Alexander, whereas the commentaries on Books E to N are in their extant form not authentic.2 This general contention was by no means new:3 Valentin Rose had pointed out in 1854 that Alexander’s commentary on Λ as quoted by Ibn Rušd in his Long Commentary (Tafsı¯r) on the Metaphysics4 differs considerably from the Λ section

1

2

3 4

The authors would like to thank Peter Adamson (Munich) for his encouragement and contribution to a preliminary stage of the research; Shady H. Nasser (Cambridge) and Marwan Rashed (Paris) for expert advice on a difficult passage in Ibn Rušd’s Arabic and on the uses and areas of application of the “abgˇad” notation of Arabic letters; and Irina Wandrey (Hamburg) for help with the Hebrew translations of Ibn Rušd’s Tafsı¯r on Book Λ. The authors are solely responsible for any mistakes that remain. Hayduck, M. (ed.) 1891: Alexandri Aphrodisiensis in Aristotelis metaphysica commentaria, consilio et auctoritate Academiae Litterarum Regiae Borussicae edidit Michael Hayduck. (= CAG I), Berolini, V: posterioris partis interpres quin et ratione interpretandi et orationis proprietate a vero Alexandro, qui apparet in priore, longe distet, dubitari non potest. See Luna, C. 2001: Trois études sur la tradition des commentaires anciens à la Métaphysique d’Aristote. (= Philosophia antiqua, Vol. LXXXVIII), Leiden–Boston–Köln, 55 n. 108. Ibn Rušd’s commentary covers, in this order, books α, Α (truncated), Β, Γ, Δ, Ε, Ζ, Η, Θ, Ι, and Λ; see Bouyges, M. S. J. (ed.) 1952: Averroès, Tafsı¯r Ma¯ ba›d at-Tabı¯ ›at. Texte arabe inédit, établi par Maurice Bouyges, S. J., Notice, Beyrouth, XVIII. In Rose’s day, the Arabic original of the commentary had not yet been rediscovered, but it had been accessible through a 13th century Latin translation (executed between 1220 and 1224 by Michael Scot and printed in Padua as early as 1473), on which see Freudenthal, J./Fränkel, S. 1885: Die durch Averroes erhaltenen Fragmente Alexanders zur Metaphysik des Aristoteles: Berlin, 121‒123 (“Die lateinische Afterversion”), Bouyges (see above), LXVI‒LXXIX, and Hasse, D. N. 2010: Latin Averroes Translations of the First Half of the Thirteenth Century,

12

Matteo Di Giovanni and Oliver Primavesi

of the extant Greek commentary on Books Ε‒Ν.5 In addition Jacob Freudenthal had argued in 18856 that the Μ‒Ν section of the extant Greek commentary depended on the paraphrase of the Neo-Platonic scholar Syrianus (5thcentury AD),7 not vice versa, as had been thought by Bonitz (1847) and Usener (1870). If Freudenthal is correct, the possibility that Alexander of Aphrodisias composed the Μ‒Ν section of the extant Greek commentary is ruled out from the start by chronology. Furthermore, the general contention according to which the extant form of the commentary on Books Ε‒Ν (henceforth: “Ps.-Alexander’s Greek commentary”) cannot be ascribed to Alexander had also been introduced in the guise of two stronger and more specific claims. Claim A: Christian August Brandis in 1836 had not only taken it for granted that the authorship of the extant Greek commentary on Books Ε‒Ν must be denied to Alexander of Aphrodisias, but he also suggested the possibility that it was written by Michael of Ephesus instead, to whom the work is attributed in MS Parisinus 1876 (A).8 Félix Ravaisson (1837) had argued for Michael’s authorship more confidently, by quoting from a Latin translation Michael’s own reference to his commentary on Metaphysics Ζ‒Ν.9 The ascription had

5

6 7

8

9

Hildesheim–Zürich–New York. On Ibn Rušd’s quotations from the Arabic translation of “Alexander’s” Commentary on Λ see Bouyges (see above), CLXXVII‒CLXXIX. Rose, V. 1854: Valentini Rose de Aristotelis librorum ordine et auctoritate commentatio, Berlin, 150: Ad liquidum vero haec res perducitur inspecto Averrois in librum duodecimum (λ’) commentario, cui quidem ad hunc solum librum quum ad ceteros libros nullius interpretis scholia reperta essent, oblata est ut ipse ait … et Themistii paraphrasis et expositionis Alexandri maxima pars …, cuius in singulis textus articulis verba apposuit verbotenus ut Themistii comparatio comprobat … translata et a Pseudo-Alexandro manifesto diversa; cf. Freudenthal (see note 4), 12 with n. 1. Freudenthal (see note 4), 28‒34. Edited by Usener, H. 1870: Scholia in Aristotelem. Supplementum, in: Aristotelis opera. Edidit Academia Regia Borussica, Volumen quintum, Berolini and by Kroll, G. 1902: Syriani in Metaphysica commentaria, consilio et auctoritate Academiae Litterarum Regiae Borussicae edidit Guilelmus Kroll. (= CAG VI 1), Berolini. Brandis, A. (ed.) 1836: Scholia in Aristotelem. Collegit Christianus Augustus Brandis. Edidit Academia Regia Borussica, Berolini, 734: Ad libros seqq. Metaphysicorum non integros dedi Alexandri, qui feruntur, commentarios, sed scholia tantum ex iis excerpta, cum mihi dubium non sit falso eos Aphrodisiensis nomen prae se ferre, sive Michaelis Ephesii sunt, quem admodum cod. Reg. Par. 1876 autumat (Μιχαὴλ τοῦ Ἐφεσίου σχόλια εἰς τὸ ε’ τῶν Μετὰ τὰ Φυσικὰ τοῦ Ἀριστοτέλους), sive alius cuiusdam similis notae scholiastae. According to Hayduck’s apparatus criticus (Hayduck, 440), the first hand of Parisinus A had written σχόλιον εἰς τὸ ε τῶν Μετὰ τὰ Φυσικὰ τοῦ Ἀριστοτέλους, whereas the second hand prefixed Μιχαὴλ τοῦ Ἐφεσίου and changed σχόλιον to σχόλια. Ravaisson, F. 1837: Essai sur la Métaphysique d’Aristote, Paris, 65 n. 1. Cf. Michael In Parva Nat. 149,14‒15 Wendland, P. (ed.) 1903: Michaelis Ephesii in Parva naturalia com-

Who Wrote Alexander’s Commentary on Metaphysics Λ?

13

gained further support from Valentin Rose,10 whereas it was doubted by Freudenthal on two counts: (i) The fact that the Greek Ps.-Alexander refers back to the extant books of the genuine commentary by Alexander of Aphrodisias as if these had been written by himself shows that he is a forger;11 yet Michael of Ephesus is unlikely to have been a forger.12 (ii) The Greek Ps.Alexander explains Aristotle’s references to the divine in a way which betrays that he adhered to pagan polytheism; yet Michael of Ephesus was a Christian.13 Claim B: On a different note, Freudenthal had maintained that Ps.-Alexander’s Greek commentary on Books Ε‒Ν cannot even be derived, however indirectly, from Alexander’s authentic commentary.14 This view rested upon a more thorough investigation of Ibn Rušd’s quotations from “Alexander’s” commentary on Book Λ than had been attempted by Rose. (Notice that the author of the commentary on Λ quoted by Ibn Rušd is henceforth written within quotation marks – “Alexander” – since we shall be concerned with the question of the extent to which “Alexander” is in fact Alexander of Aphrodisias.) A German translation of these quotations had been prepared by Freudenthal on the basis of the two Hebrew translations of Ibn Rušd’s commentary,15 subsequently checked against the rediscovered codex unicus

10 11

12

13 14 15

mentaria, (= CAG XXII 1), Berolini: γέγραπται δέ μοι καὶ εἰς τὰ Μετὰ τὰ φυσικὰ ἐξ αὐτοῦ τοῦ ζῆτα ἕως τοῦ νῦ. Rose (see note 5), 147; cf. Freudenthal (see note 4), 53. Freudenthal (see note 4), 19: “Er ist also keiner jener Compilatoren, die in naiver Unbefangenheit Auszüge aus früheren Schriften den ihrigen einfügen, sondern ein Betrüger, der in bewusster Absicht fälscht.” Freudenthal (see note 4), 53: “Denn wir sind nicht berechtigt, diesen unklaren, aber für seine Zeit sehr gelehrten und um die Erklärung der aristotelischen Schriften eifrig bemühten Mann für einen Betrüger zu halten als welchen wir Ps.-Alexander erkannt haben.” Freudenthal (see note 4), 53: “Der Verfasser der letzten Bücher des Commentars ist kein Christ wie Michael, sondern ein Anhänger des griechischen Götterglaubens.” Freudenthal’s point is correctly summarized by Hayduck (see note 2), V‒VI. On the two Hebrew translations see Freudenthal (see note 4), 134 (for the sigla of manuscripts) and 116‒120, as well as Bouyges (see note 4), LXXXIV‒XCVII; for more recent assessments of the identity of the two translators cf. Zonta, M. 2001: Sulla tradizione ebraica di alcuni commenti alla Metafisica (Abu¯ l-Faragˇ ibn al-Tayyib e Averroè), in: Documenti e studi sulla tradizione filosofica medievale, 12, 164‒177 and Halper, Y. 2013: Revision and Standardization of Hebrew Philosophical Terminology in the Fourteenth Century: The Example of Averroes’s Long Commentary on Aristotle’s Metaphysics Δ, in: Aleph: Historical Studies in Science and Judaism, 13, 95–137. The first, anonymous, translation (Freudenthal: y, Bouyges: ‫ )ﬡ‬is transmitted by Parisinus hebr. 886 (Freudenthal: A, Bouyges: a). For the second translation (Freudenthal: z, Bouyges: ‫)ﬢ‬, a revision of the first which is attributed in some manuscripts to “Moses the son of Solomon of the sages of Salon”, both Freudenthal and Bouyges used Parisinus hebr. 887 (Freudenthal: E, Bouyges d) and Parisinus hebr. 888 (Freudenthal: B, Bouyges f). Apart from these, Freudenthal used Parisinus

14

Matteo Di Giovanni and Oliver Primavesi

of the Arabic original 16 by Siegmund Fränkel, and finally published by Freudenthal and Fränkel in 1885.17 Notably, in the introduction to this collection of “Alexander’s” fragments, Freudenthal attempts to show that Ibn Rušd’s quotations are both entirely unrelated to Ps.-Alexander’s Greek commentary and a faithful representation of Alexander’s genuine commentary on Λ. Freudenthal’s argument can be summarized as follows: (i) Ps.-Alexander’s Greek commentary on Λ is not merely different from but even completely incompatible with the commentary on Λ quoted by Ibn Rušd from an Arabic translation in which the commentary was presented as a work by Alexander of Aphrodisias.18 (ii) The transmitted attribution of the commentary quoted by

16

17

18

hebr. 889 (D), Parisinus hebr. 890 (C), Mich. hebr. 441 (O), and Taurinensis hebr. 14 (T), whereas Bouyges added Vaticanus Urbinas hebr. 46 (u) and Estensis hebr. I. C. 17 (75) (e). Further manuscripts are listed and discussed by Halper (see note 15). The proem to Lambda was later rendered into Latin from the Hebrew by Paulus Ricius (“Paulus Israelita”, d. 1541) and Jacob Mantino (d. 1549). In addition, two Latin translations of the proem are ascribed to the Jewish scholar Elijah Delmedigo (“Helias Cretensis”) as noted by Bouyges (see note 4), LXXIX‒LXXXI. It is unclear whether the proem was included in the ArabicLatin translation of the Tafsı¯r realized by Michael Scot (Bouyges 1948b: La Métaphysique d’Aristote chez les Latins du XIIIe siècle: Connurent-ils le Proœmium d’Averroès à son commentaire du livre Lam-Lambda?, in: Revue du moyen âge latin: Études − Textes − Chronique − Bibliographie, 5, 279–281) which, however, should cast no doubts on its authenticity (Bouyges (see note 4), CXI‒CXII). The existence of a manuscript containing the Arabic original of Ibn Rušd’s commentary became known only in 1873 when the manuscript was transferred from the Royal Library of The Hague to the Library of Leiden University (Cod. arab. 1692 = Cod. or. 2074; the manuscript had been owned by the Jesuits of Paris until the dissolution of the Society in 1763), and when the Leiden orientalist Michael Jan De Goeje published the following brief description of it (De Goeje 1873: Codices Orientales Bibliothecae Academiae LugdunoBatavae nuper acquisiti, in: De Goeje, M. J. (ed.), Catalogus Codicum Orientalium Bibliothecae Academiae Lugduno-Batavae, Leiden, 324‒325): “MMDCCCXXI. (Cod. 2074.) Commentarius Averrois (+ 595) magnus ad Aristotelis Metaphysicam …, cujus alterum exemplar non superesse videtur. … Codex, Mauritanâ manu non ita male scriptus, praesertim in fine grave damnum passus est. Oriundus videtur e bibl. Coll. Paris. Soc. Jesu, nam notam memoratam [scil. “paraphé au désir de l’arrest du 5 Juillet 1763”] inscripsit Mesnil”; see Bouyges (see note 4), XXXII‒XXXVIII. The somewhat convoluted genesis of Freudenthal’s study (see note 4) is due to the fact that Freudenthal worked on the Hebrew versions of Ibn Rušd’s commentary on the assumption that the Arabic original was lost. Only at a later moment, when a first draft of his translation with introduction had already been completed, was Freudenthal alerted by Moritz Steinschneider that the Leiden library had recently acquired a manuscript of the Arabic original, which was then checked for Freudenthal by Fränkel; see Freudenthal (see note 4), 66. Freudenthal (see note 4), 3‒10; see especially p. 8: “Nicht häufig wird man zwei Commentare desselben Textes finden, die so weit voneinander abstehen, wie der griechische Alexander und der des Averroes. Wo sie nicht Worte des Aristoteles einfach umschreiben, da haben sie überhaupt keinerlei Beziehung, oder stehen in schroffem Gegensatze zu einander. Dies Verhältniss ist unerklärbar, wenn man der Meinung ist, dass das wenn auch noch so lose Band gemeinsamen Ursprungs beide Commentare verbinde.”

Who Wrote Alexander’s Commentary on Metaphysics Λ?

15

Ibn Rušd to Alexander of Aphrodisias is to be accepted, since Ibn Rušd’s quotations nowhere contradict Alexander’s known doctrines,19 and since they are remarkably close, with regard to both style and contents, to Alexander’s authentic works.20 (iii) It follows that Ps.-Alexander’s Greek commentary on Λ does not go back in any way, be it directly or indirectly, to the authentic commentary on the same book by Alexander of Aphrodisias.21 (iv) There is reason to believe that in Ps.-Alexander’s Greek commentary the sections on Books Ε‒K and M‒Ν are no less unrelated to Alexander’s genuine work on the Metaphysics.22 It follows that Ps.-Alexander’s Greek commentary on Ε‒Ν as a whole does not go back to Alexander’s exposition of the Metaphysics. The two claims do not depend upon, but rather complement each other: (A) If Ps.-Alexander’s Greek commentary on Books Ε‒Ν was written, in its extant form, by Michael of Ephesus as suggested by Brandis 1836 and Ravaisson 1837, it cannot have been written by Alexander of Aphrodisias, but it may still contain material ultimately going back to Alexander. (B) If, however, “Alexander’s” commentary on Λ quoted by Ibn Rušd was written by Alexander, as suggested by Freudenthal, then Ps.-Alexander’s Greek commentary on Books Ε‒Ν was not only not written by Alexander (though it may still have been written by someone other than Michael of Ephesus), but was not even in any way derived from Alexander’s commentary. Yet in research subsequent to Hayduck’s 1891 edition the two claims have, despite their mutual relation, fared quite differently: Claim A: The pros and cons of Brandis’ attribution of Ps.-Alexander’s Greek commentary to Michael of Ephesus have been critically re-examined more than once.23 In his pioneering review of Hayduck’s 1904 edition of Michael’s commentaries on Part. An., Mot. An. and Inc. An., Karl Praechter was the first to make use of internal evidence (i.e. detailed stylistic comparison) for the purpose of showing that not only Ps.-Alexander’s Greek commentary on

19 20 21

22 23

Freudenthal (see note 4), 39: “Es lässt sich in ihnen kein Satz aufweisen, dessen Inhalt mit den Lehren des Alexander in Widerspruch stände.” Freudenthal (see note 4), 39‒51. Freudenthal (see note 4), 34: “Sie (scil. eine unparteiische Untersuchung des Ursprungs der averroistischen Auszüge) wird die Kette von Beweisen schliessen, welche darthun, dass der griechische Commentar zum zwölften Buche der Metaphysik weder in der uns vorliegenden Gestalt von Alexander verfasst ist noch als Epitome oder Bearbeitung einer Schrift Alexanders angesehen werden darf.” Freudenthal (see note 4), 64. Research on the issue until the year 2000 has been amply documented by Luna (see note 3), 53‒66.

16

Matteo Di Giovanni and Oliver Primavesi

Metaph. Ε‒Ν, but also the commentaries by Ps.-Philoponus on Gener. anim. and by Ps.-Alexander on Soph. el. should be ascribed to Michael of Ephesus.24 ‒ Leonardo Tarán (1987) attempted to reverse the relationship between Syrianus and the Greek Ps.-Alexander postulated since Freudenthal, and returned to the view that Syrianus depends on the Greek Ps.-Alexander,25 with the result that the attribution of the commentary on Metaph. Ε‒Ν to Michael of Ephesus would be ruled out by chronology. ‒ Concetta Luna (2001) published an important monograph on the relationship between Alexander, Syrianus, Asclepius, the Greek Ps.-Alexander (i.e. Michael of Ephesus, according to Luna), and Ps.-Philoponus (i.e. Georgios Pachymeres),26 in which, as a result of careful discussion and evaluation of both Praechter’s and Tarán’s work, she came to side firmly with Praechter 27 against Tarán.28 Finally Pantelis Golitsis (2014), while accepting Michael’s authorship for Books Ζ‒Ν,29 has denied it for Book E, proposing to attribute the extant Greek commentary on Book E to the author of the so-called recensio altera of the genuine parts of Alexander’s commentary. Claim B: By contrast, nobody seems ever to have seriously assessed Freudenthal’s contention that Alexander’s genuine commentary on Λ is represented by the testimony of Ibn Rušd. Tarán 1987 simply took the results of Freudenthal’s perusal of the Arabic “Alexander” for granted,30 and in Concetta Luna’s otherwise excellent monograph on the tradition of the ancient com24

25

26 27 28 29

30

Praechter, K. 1906: Review of Hayduck 1904, in: Göttingische gelehrte Anzeigen, 863 n. 3: “Ps.-Philoponos zu de gener. anim. (ed. Hayduck, Comm. vol. XIV pars III), Ps.-Alexander zu Metaph. E–N (ed. Hayduck, Comm. vol. I p. 440 ff.) und Ps.-Alexander zu sophist. elench. (ed. Wallies, Comm. vol. II pars III) nehme ich als Eigentum des Michael in Anspruch ‒ die Berechtigung dazu werde ich weiter unten nachweisen ‒ und bezeichne den ersten mit Pg., den zweiten mit Am., den dritten mit As.” Tarán, L. 21987: Syrianus and Pseudo-Alexander’s commentary on Metaph. E–N, in: J. Wiesner (ed.), Aristoteles Werk und Wirkung. Paul Moraux gewidmet. Bd. I. Aristoteles und seine Schule, Berlin–New York, 231: “For his commentary on books M–N Syrianus made use of Ps.-Alexander’s commentary, which he mistook for the work of Alexander himself.” Luna (see note 3), 187‒190 summarizes her results and illustrates them by means of a stemma. Luna (see note 3), 53‒66. Luna (see note 3), 37‒53. Only that much is claimed by Michael himself in the passage first quoted by Ravaisson (see note 9), 65 n. 1, i.e. Parv. Nat. 149,14‒15: γέγραπται δέ μοι καὶ εἰς τὰ Μετὰ τὰ φυσικὰ ἐξ αὐτοῦ τοῦ ζῆτα ἕως τοῦ νῦ. Tarán (see note 25), 217‒220; especially 220: “Freudenthal has shown that the Ps.-Alexander did not have direct or indirect access to the lost part of Alexander’s commentary on the Metaph.” That is, of course, not to deny that Tarán strongly disagrees with Freudenthal’s view on the relationship between Ps.-Alexander and Syrianus.

Who Wrote Alexander’s Commentary on Metaphysics Λ?

17

mentaries on the Metaphysics, Ibn Rušd’s quotations from the Arabic “Alexander” are nowhere mentioned.31 So it seems that, while Brandis’ identification of the Greek Ps.-Alexander with Michael of Ephesus has been thoroughly debated in utramque partem, Freudenthal’s identification of the Arabic “Alexander” with Alexander of Aphrodisias has remained unexamined so far. There is, however, a compelling reason for not leaving the matter here: between 1938 and 1948 an outstanding editio princeps of the complete Arabic text of Ibn Rušd’s Long Commentary (Tafsı¯r) on the Metaphysics was published by Maurice Bouyges, S. J. (1878–1951).32 As a result we are now in a far better position to examine both Freudenthal’s selection of the texts attributed to Alexander’s commentary and his own evaluation of the material.33 Bouyges himself, in the posthumously published introduction to his edi31

32

33

Concentrating on the Greek material, Luna’s monograph provides no entry in the pertinent Indexes on either of the following items: Ibn Rušd, cf. “Index des auteurs anciens” (Luna [see note 3], 229‒230); the editor princeps of Ibn Rušd’s Tafsı¯r, Maurice Bouyges, S. J., and the two modern translators of the Tafsı¯r’s section on Λ, Charles Genequand and Aubert Martin, cf. “Index des auteurs modernes” (Luna [see note 3], 231‒232); passages either from “Alexandre d’Aphrodise, In Metaphysica (livre Λ)” or Ibn Rušd, cf. “Index des textes cités” (Luna [see note 3], 233‒250). By contrast, there is an entry on Jacob Freudenthal, but in those places where Luna refers to Freudenthal 1885 she is concerned with the relationship between Ps.-Alexander and Syrianus (Luna (see note 3), 48‒51), or with the identification of Ps.-Alexander with Michael of Ephesus (Luna [see note 3], 55 n. 108 and 67 n. 153), and not with Ibn Rušd’s Arabic quotations from Alexander’s commentary on Book Λ. See Bouyges, M. S. J. (ed.) 1938: Averroès, Tafsı¯r Ma¯ ba›d at-Tabı¯ ›at. Texte arabe inédit, établi par Maurice Bouyges, S. J., Premier Volume: Livres petit alif, grand alif, ba’, gim, Beyrouth (Books α, Α, Β, Γ), Bouyges, M. S. J. (ed.) 1942: Averroès, Tafsı¯r Ma¯ ba›d atTabı¯ ›at. Texte arabe inédit, établi par Maurice Bouyges, S. J., Deuxième Volume: Livres dal, he, zay, hha’, tta’, Beyrouth (Books Δ, Ε, Ζ, Η, Θ), and Bouyges, M. S. J. (ed.) 1948a: Averroès, Tafsı¯r Ma¯ ba›d at-Tabı¯ ›at. Texte arabe inédit, établi par Maurice Bouyges, S. J., Troisième Volume: Livres ya’ et lam, Beyrouth (Books Ι, Λ). Bouyges observed that the Codex unicus brought to light by De Goeje (see note 16) and used by Freudenthal (see note 4) (Cod. arab. 1692 = Cod. or. 2074 = B) lost at least 26 folia, and he was able to show that 21 of these were bound together with, and now form the second part of, another Leiden manuscript of the same provenance (Cod. arab. 1693 = Cod. or. 2075 = C); see Bouyges (see note 4), XXXVIII‒XLII. On the basis of Bouyges 1948a (see note 32) Ibn Rušd’s commentary on Λ has been translated into English by Genequand, C. (ed.) 1984: Ibn Rushd’s Metaphysics. A Translation with Introduction of Ibn Rushd’s Commentary on Aristotle’s Metaphysics, Book La¯m, Leiden, and into French by Martin, A. (ed.) 1984: Averroès: Grand Commentaire de la Métaphysique d’Aristote (Tafsı¯r Ma¯ ba›d at-Tabı¯ ›at), livre Lam-Lambda: Traduit de l’arabe et annoté. (= Bibliothèque de la Faculté de Philosophie et Lettres de l’Université de Liège, Fasc. CCXXXIV), Paris (which does not include chapter 8). Moraux, P. 2001: Der Aristotelismus bei den Griechen von Andronikos bis Alexander von Aphrodisias. Dritter Band: Alexander von Aphrodisias. J. Wiesner, Berlin–New York, 491– 501 still takes Freudenthal’s selection for granted.

18

Matteo Di Giovanni and Oliver Primavesi

tion, considers it likely that Freudenthal’s collection of Ibn Rušd’s quotations from “Alexander” is incomplete.34 If Bouyges is correct, it seems urgent to verify whether Freudenthal’s crucial contention that Ibn Rušd’s quotations nowhere contradict Alexander’s known doctrines35 also applies to the material that is omitted by Freudenthal. If that test should fail, we would at least have to qualify, if not abandon, Freudenthal’s claim that the material quoted by Ibn Rušd from “Alexander” goes back directly to Alexander’s authentic commentary. For the purpose of this evaluation the most relevant passage appears to be the introduction of Ibn Rušd’s commentary on Λ (1393,4‒1405,12 Bouyges). It is this introduction that forms the object of the present study.36 The text of the introduction is organized as follows:37 (I)

(1393,4‒1394,2 B.): Preliminary remarks on two extra sources available only for Book Λ: “Alexander’s” commentary on roughly the first two thirds of Λ, and Themistius’ paraphrase of the whole book. (II) (1394,3‒1395,8 B.): Account of “Alexander’s” assessment of the status of Book Λ. (III) (1395,9‒1405,12 B.): Analysis of the argumentation that runs through all books of the Metaphysics, which is in turn subdivided as follows: a) (1395,9‒1397,2 B.): Diaeresis of the topics to be addressed in a study of being qua being; b) (1397,3‒1405,12 B.): Analysis of the individual books α‒Ν, showing that Aristotle has achieved, by his arrangement of the topics deduced in section III (a), an articulation of the study of being qua being which is perfectly suitable for the purposes of instruction.

34

35

36

37

Bouyges (see note 4), CLXXVIII: “Je crois que sa Liste est incomplète”. The judgement is reiterated by Martin 1984, 26: “Sans doute faut-il admettre avec Bouyges (Notice, CLXXVIII) que les citations faites par Averroès sont plus nombreuses que ne le laisse croire le choix, très critique, de Freudenthal.” Freudenthal (see note 4), 39: “Es lässt sich in ihnen kein Satz aufweisen, dessen Inhalt mit den Lehren des Alexander in Widerspruch stände; der Charakter der alexandrischen Commentare tritt vielmehr in unverkennbarer Entschiedenheit wieder hervor.” In addition to the two modern translations of Ibn Rušd’s commentary on Book Λ as a whole by Charles Genequand (1984: English, see note 32) and Aubert Martin (1984: French, see note 32), a Castilian translation of Ibn Rušd’s proem to the commentary on Book Λ has been provided by Ramón Guerrero, R. 1996: Averroes: el «Proemio» de su Comentario al libro lambda de la «Metafísica», in: R. R. Guerrero (ed.), Memoria-Homenaje a Adolfo Arias (1945–1993). Anales del seminario de historia de la filosofía, Número extraordinario, Madrid, 286‒295. Cf. the brief characterization by Bouyges (see note 15), 279: “C’est une sorte de Proœmium, dont la majeure partie est une analyse sommaire, non pas du seul livre Lam-Lambda, mais de chacun des livres de la Métaphysique d’Aristote.”

Who Wrote Alexander’s Commentary on Metaphysics Λ?

19

The commentary by “Alexander” mentioned in Part I contained Aristotle’s text, too.38 Clearly the two sources, i.e. a text with “Alexander’s” commentary and a paraphrase by Themistius, correspond to the two Arabic translations mentioned two centuries earlier by Ibn al-Nadı¯m in the Fihrist.39 What is not clear is how much of Ibn Rušd’s introduction goes back to “Alexander”: while it is universally agreed that in Part II Ibn Rušd reports what “Alexander” said on the status of Book Λ, the source of Part III has been a matter of controversy. Valentin Rose (1854), whose judgement was still based on the Latin translation, deemed it obvious that – just as Part II of the introduction is devoted to Alexander’s proem to Book Λ (Alexandri … in hunc ipsum librum prooemium) – Part III contains Ibn Rušd’s summary of Alexander’s analysis of the argument of the Metaphysics as a whole (eiusdem de omnium librorum Metaphysicorum argumento … disputationem).40 Jacob Freudenthal, by contrast, included only Part II in his collection of fragments from “Alexander”,41 whereas he left out Part III entirely: Freudenthal correctly observed that one important criterion for determining the authorship of the contents of Part III consists in the two framing sentences (1395,9‒10 B. and 1405,9‒12 B. respectively) by which it is introduced and concluded, and he claimed that it was the inaccurate Latin translation of these two sentences that tempted Rose into believing that also the contents of Part III go back to Alexander, whereas in fact they are entirely the work of Ibn Rušd.42 Furthermore, Freudenthal drew up a list of six features of 38

39

40

41 42

Bouyges (see note 4), CXXX. For an updated assessment of Ibn Rušd’s sources in his commentary on Lambda see Geoffroy, M. 2003: Remarques sur la traduction Usta¯t du livre Lambda de la Métaphysique, chapitre 6, in: Recherches de Théologie et Philosophie Médiévales, 70, 417–436. See Ibn al-Nadı¯m’s notice in Flügel, G. (ed.) 1871: Kitâb al-Fihrist, mit Anmerkungen herausgegeben von Gustav Flügel. Erster Band: den Text enthaltend, Leipzig, 251,25–252,1, to which we shall return later. Rose (see note 5), 151–152: “Maxime autem memorabilis est Averrois in eundem illum librum λ’ praefatio … in qua ille Alexandri et in hunc ipsum librum prooemium retulit et longam deinde eiusdem de omnium librorum Metaphysicorum argumento hic in ipsius doctrinae metaphysicae fine brevibus repetito disputationem, qua simul ordinem librorum quem ceteris adeo Aristotelis libris exactiorem praedicat ut solet defendit et dispositionem operis partiumque divisionem exhibet.” Cf. Freudenthal (see note 4), 67–69, in particular the section designated as Frgm. 1. Freudenthal (see note 4), 122–123 on the Latin translation of the introductory sentence 1395,9–10 B.: “Vgl. ferner die Übersetzung (p. 311C), durch welche Rose (a. a. O p. 151) verleitet wurde, die ganze dort folgende Erörterung des Averroes dem Alexander beizulegen: Nec absonum est huc quoque afferre ea quae ad summariam aliarum huius scientiae dictionum intelligentiam Alexander praeposuit”. Freudenthal 1885, 122 on the Latin translation of the concluding sentence 1405,9–12 B.: “Vgl. 312D: nos autem summa indagine atque amore prosecuti sumus hanc scientiam et iam ante huius viri in hac scientia dictiones compendio perstrinximus, in eruditionem eorum, qui non ad amussim Arist. verbis studuerint

20

Matteo Di Giovanni and Oliver Primavesi

Part III which, in his view, proved that it was composed by Ibn Rušd and not by Alexander. This list is based on the Hebrew translations, but in the following English rendering of Freudenthal’s points we shall add, in square brackets, references to Bouyges’ edition of the Arabic text where appropriate:43 − − − − − −

The analysis starts with the phrase ‘and we say’ (‫[ )וכאמר‬1395,11 B.]. The overall contents of the analysis show that it goes back to Ibn Rušd, not to Alexander. The books of the Metaphysics are labelled in accordance with the Arabic fashion. Book Alpha Minor is called “the first book” [1397,5‒13 B.] and Book Alpha Major comes second [1397,14‒1398,7 B.]. Towards the end there is a reference to Nicolaus of Damascus [1405,4‒ 8 B.]. The concluding sentence [1405,9‒12 B.] suggests that the analysis is by Ibn Rušd himself.

Freudenthal’s assessment was rejected by Maurice Bouyges in a short note published in 1948. According to Bouyges, Ibn Rušd’s proem (1393,4‒ 1405,12 B.) is mainly44 a summary – and in part even a copy – of the Arabic translation of the corresponding part of “Alexander’s” commentary on Book Λ.45 In particular, Bouyges specified that Freudenthal was wrong to exclude the analysis of the whole Metaphysics (1395,9‒1405,12 B., = our Part III) from his collection of fragments.46 In the posthumously published introduction (“Notice”) to his edition Bouyges added that the analysis of the Metaphysics offered in Part III reminded him of the technique of Greek or Byzan-

43

44 45 46

et ad epilogi instar his, qui eius verba penetraverint. Dass hier mit Unrecht Averroes’ Epitome der Metaphysik in die Übersetzung eingeschwärzt wird, ergiebt fr. 1 (S. 69).” Cf. Freudenthal (see note 4), 123: “Dass aber das Nachfolgende die Erörterungen des Averroes, nicht die des Alexander enthält, erweist das ‫‘ וכאמר‬und wir sagen’, womit der folgende Abschnitt beginnt, erweist auch der Inhalt desselben, sowie die bei den Arabern übliche Zählung der Bücher, die Voraufstellung des zweiten Buches (α) vor Α, das Citat aus Nikolaus von Damaskus und endlich das Schlusswort des Averroes (oben S. 69).” That is to say, apart from more or less obvious additions like, for instance, the preliminary remark on sources in the paragraphs between 1393,4 and 1394,2 B. Bouyges (see note 15), 280: “La majeure partie de ce Proœmium est, si je ne me trompe, un résumé (ici ou là une copie?) du « Commentaire d’Alexandre ».” Bouyges (see note 15), 280 n. (6): “J. Freudenthal n’a pas inclus ledit résumé, ou copie, dans sa précieuse collection des « Fragments d’Alexandre », traduits par lui en allemand, op. cit., p. 67–133. – Mais je crois qu’il a eu tort”. See also Bouyges (see note 4), XX n. (1).

Who Wrote Alexander’s Commentary on Metaphysics Λ?

21

tine commentators,47 and he further speculated that Freudenthal’s mistaken exclusion of that analysis was entirely due to the fact that in the introductory sentence of Part III (i.e. 1395,9‒10 B.) he adopted a wrong reading based on a variant in the Hebrew transmission: in Bouyges’ belief, the fact that Freudenthal preferred the variant in question makes it superfluous to suppose that his decision was also based on an internal assessment of the passage.48 Irrespective of whether Bouyges is or is not right on the main point, that is, in rejecting Freudenthal’s exclusion of Part III of the introduction from the fragments of “Alexander’s” commentary, certainly he is mistaken in ruling out that Freudenthal’s decision was based on an internal assessment of our Part III. Bouyges must have overlooked Freudenthal’s aforementioned list of the six features which supposedly undermine the attribution of Part III to Alexander. Freudenthal’s six arguments against the authenticity are nowhere addressed by Bouyges, and no one in more recent research has engaged in assessing Freudenthal’s arguments independently of Bouyges’ misrepresentation.49 Therefore, paradoxical as it may seem, it is still a desideratum that proper consideration be given to Freudenthal’s actual arguments, by checking them against the evidence provided by Ibn Rušd’s text and examining their real potential to undermine the optimistic judgements by Rose (1854) and by Bouyges (1948b). It is important to notice that we are faced with not just two options here, but with three: Option (i)

Ibn Rušd draws the analysis of the Metaphysics expounded in Part III from “Alexander’s” commentary on Book Λ, which in turn preserved – in Arabic translation – an authentic work by the historical Alexander of Aphrodisias (Bouyges). Option (ii) The analysis of the Metaphysics expounded in Part III is entirely the work of Ibn Rušd (Freudenthal). Option (iii) Ibn Rušd draws the analysis of the Metaphysics expounded in Part III from “Alexander’s” commentary on Book Λ, but the analysis cannot go back to the historical Alexander of Aphrodisias unless it was subjected to a revision prior to Ibn Rušd.

47

48 49

Bouyges (see note 4), CLXXVIII: “le Tafsı¯r arabe qu’Averroes avait entre les mains … débute par une analyse succincte de tous les Livres de la Métaphysique – ce qui rappelle bien, si je ne me trompe, la technique grecque ou byzantine des commentateurs.” Bouyges (see note 4), CLXXVIII, n. 7: “… la variante adoptée par lui p. 1395, 1042 … nous dispense, je crois, de supposer une raison tirée de la critique interne.” This holds not only for Genequand (see note 32) and Martin (see note 32), but also for Ramón Guerrero (see note 36), who has devoted a whole article to Ibn Rušd’s proem to the commentary on Book Λ without ever mentioning Freudenthal’s work.

22

Matteo Di Giovanni and Oliver Primavesi

Option (iii), while appearing at first sight as the cautious middle course, is potentially dangerous for Freudenthal’s fundamental claim from which we started, that is the attribution of “Alexander’s” commentary on Book Λ as a whole to the historical Alexander of Aphrodisias. For both options (i) and (ii), though radically opposed with regard to the analysis of the Metaphysics presented in Part III, leave open the possibility that the rest of “Alexander’s” commentary on Λ as quoted by Ibn Rušd goes back directly to the historical Alexander of Aphrodisias. But if we had positive reasons to admit both that Ibn Rušd does owe the analysis to what he quotes as “Alexander’s” commentary on Book Λ, and that the analysis, in its transmitted form, cannot be attributed to the historical Alexander of Aphrodisias, then the unqualified authenticity of Ibn Rušd’s later quotations from “Alexander” could no longer be taken for granted. In this case, Freudenthal’s arguments would fail to free Ibn Rušd’s “Alexander”-quotations from the potentially compromising company of Part III of the proem, and would instead cast serious doubts on the authenticity of the source of those quotations. Before we come to the analysis of the Metaphysics in Part III of the introduction it should be noted that the fourth feature of Freudenthal’s list, i.e. the unorthodox sequence Alpha Minor – Alpha Major, turns up already within Part II: in the account of “Alexander’s” assessment of the status of Book Λ, which Freudenthal did include in his collection,50 Book Alpha Major is explicitly called the “second book” of the Metaphysics.51 Freudenthal bracketed the phrase as a secondary addition made by a reader (possibly Ibn Rušd himself) who regarded Alpha Minor as the first book.52 It seems, however, rather odd that one of the features which induced Freudenthal to exclude Part III of the introduction from his collection appears also in Part II, where Freudenthal must athetize it because the attribution of Part II to “Alexander” is unavoidable. With this in mind we can now turn to Part III, which will be provided in a complete English translation, and selectively commented upon, as required by the argument.53

50

51 52

53

Cf. Freudenthal (see note 4) 68‒69 (from the translation of Frgm. 1): “Diese Gegenstände aber hat er schon in dem Buche, das mit gross A bezeichnet wird, [und das ist das zweite Buch dieses Werkes], erörtert …” Cf. Bouyges (see note 4), CL n. 1: “La qualification de « second Livre » est donnée à grand ALIF, p. 1395,4.” Freudenthal (see note 4) 69 n. 1: “Die eingeklammerten Worte sind offenbar der Zusatz eines Lesers, der die Metaphysik mit Buch α beginnen ließ und nicht mit Α, vielleicht des Averroes selbst.” Our translation will be subdivided in accord with the division into paragraphs in Bouyges’ edition.

Who Wrote Alexander’s Commentary on Metaphysics Λ?

23

II. “Alexander’s” Analysis of the Metaphysics in Ibn Rušd’s Tafsı¯r § c Introduction to “Alexander’s” Analysis: 1395,9‒10 B. “Yet what he [i.e. Alexander] says with regard to understanding (fı¯ tafahhum) the contents of the other books written on this discipline is open to interpretation (fı¯hi hø tima¯l), and the passage best suited for being expounded (bi-talh˚ ı¯søihi) is perhaps this passage (ha¯dß a¯ l-mawdø iʿ).”54 This introductory paragraph consists of two statements: (i) Ibn Rušd first describes “Alexander’s” remarks on the other books of Aristotle’s Metaphysics as containing ihø tima¯l. (ii) Subsequently he notes that one passage – which he identifies rather vaguely by means of the demonstrative pronoun “ha¯dß a¯” (“this”) – is best suited “bi-talh˚ ı¯søihi”: literally, “for expounding it” (with talh˚ ı¯sø read as having an active meaning) or “for its being expounded” (with talh˚ ı¯sø read as having a passive meaning). A fundamental question here is whether the remarks at issue in (i) are to be identified with the passage mentioned in (ii) or whether Ibn Rušd is referring to different texts. It should be observed in this respect that Ibn Rušd has just summarized, in Part II, the sketch of Aristotle’s Metaphysics as a whole by means of which “Alexander” intended to show that the entire treatise culminates in Book Λ. If Ibn Rušd then goes on, in statement (i), to evaluate “Alexander’s” remarks on the contents of the other books, this evaluation appears to refer back to what “Alexander” has said on the other books in the sketch summarized only a few lines above, in Part II.55 That is to say, our basic question is whether or not the passage that is said, in statement (ii), to be best suited for the exposition is to be identified with “Alexander’s” sketch just summarized. Our answer will depend on the meaning of ihø tima¯l, which is used in statement (i) to characterize the preceding sketch. As noted in Edward William Lane’s Arabic-English Lexicon the

54

55

The present translation is based on the Arabic text attested by the Leiden manuscript, with tafahhum and fı¯hi hø tima¯l in place, respectively, of the reading conjectured by Bouyges (tafhı¯m) and by Fränkel (fı¯hi igˇma¯l). See Bouyges’ apparatus criticus, ad loc. Gutas, D. 1987: Review of Charles Genequand: Ibn Rushd’s Metaphysics. A Translation with Introduction of Ibn Rushd’s Commentary on Aristotle’s Metaphysics, Book La¯m. Leiden 1984, in: Der Islam, 64, 124 n. 10: “The reference here is to the preceding synopsis of Alexander’s proem by Av[erroes], and specifically to lines 1394.5/4–9/9.”

24

Matteo Di Giovanni and Oliver Primavesi

verb ihø tamala (form VIII of hø amala, “to bear”), of which ihø tima¯l is the infinitive or verbal noun (masødar), is employed with the following meanings:56 “[In lexicology, said of a word or phrase or sentence, It bore, admitted, or was susceptible of, a meaning, a sense, or an interpretation, and, elliptically: it bore, admitted, or was susceptible of, two, or more, different meanings, senses or interpretations; it was equivocal.] In the conventional language of the lawyers, and the Muslim theologians [and men of science in general], … it is used, … or may be used, as importing supposition, and admissibleness, or allowableness; and thus used, it is intrans.” Accordingly, in statement (i) “Alexander’s” sketch is either said to be equivocal, or to be admissible. In order to decide between these two meanings we must first determine which of them can make the ascription of ihø tima¯l to “Alexander’s” sketch pertinent to the argument. Its function may consist either in showing that the sketch itself is best suited for being expounded, or that the sketch is deficient and Ibn Rušd feels urged to clarify it by means of another passage that he deems best suited for being expounded. In the first case, stating that the sketch is admissible or the like is so weak a compliment that it can scarcely sustain the inference that the sketch is of all passages the one which is best suited for being expounded.57 By contrast, stating that the sketch is equivocal is both plausible in itself – given the inherent allusiveness of a sketch where the content or function of single books is not detailed – and a likely antecedent for the introduction of another passage which is best suited for being expounded, presumably a passage in which “Alexander” does offer a more informative analysis of the Metaphysics.58 Accordingly we shall opt for the meaning “equivocal”, and assume that statement (ii) refers to another passage within “Alexander’s” commentary. Both interpretations discussed so far share the crucial assumption that the object of the exposition announced in statement (ii) is a passage taken 56 57

58

Lane, E. W. 1865: An Arabic-English Lexicon. Book I. – Part 2. ‫ج—خ‬, London–Edinburgh, 648. This becomes evident in Ramón Guerrero’s translation (see note 36, 288): “Lo que dice para comprender el contenido de los restantes libros que tratan sobre esta ciencia es una mera posibilidad, quizá este pasaje sea el más digno de su exposición” (our italics), even though Guerrero curiously goes on to explain his own translation, in the pertinent footnote, by exchanging it with its alternative rendering (“es algo ambiguo”). On this specific usage of the verbal form ihø tamala expressing equivocity or polysemy cf., in addition, the pertinent entry in Ibn Manz u¯r, Lisa¯n al-ʿArab 1884, vol. 13, p. 186: wa-fı¯ hø adı¯tß ʿAlı¯ la¯ tuna¯zø iru¯hum bi-l-Qurʾa¯n fa-inna l-Qurʾa¯n hø amma¯l dß u¯ wugˇu¯h ay yuhø malu ʿalayhi kull taʾwı¯l fa-yahø tamiluhu wa-dß u¯ wugˇu¯h ay dß u¯ maʿa¯n muh˚ talifa (“In the h adı¯t of ʿAlı¯ [it is found]: Do not argue with them on the basis of the Qurʾa¯n, because the Qurʾa¯n is a carrier (hø amma¯l) multifaceted – that is, every interpretation can be carried (yuhø malu) upon it and it bears (yahø tamilu) them all; ‘multifaceted’: that is, it has different meanings”).

Who Wrote Alexander’s Commentary on Metaphysics Λ?

25

from “Alexander’s” commentary, be it the sketch summarized in Part II or a passage different from and more informative than the sketch. This entails that the analysis of the Metaphysics as a whole which Ibn Rušd is going to present beginning with the following paragraph is introduced here as the exposition of a passage from “Alexander’s” commentary; it is, as such, an analysis by “Alexander”, and should, accordingly, be added to the fragments from or testimonies to his commentary. Jacob Freudenthal tried to avoid this conclusion by adopting another text for our statement (ii). Instead of “the passage best suited for an exposition of it” he assumed that we should read “the passage best suited for an exposition by me.” This attempt, however, is manifest only in his edition of the second Hebrew translation of our passage,59 where he treats the Hebrew text (“its expounding” = “expounding it”), which is equivalent to the Arabic form with third-person pronoun (talh˚ ı¯søihi), as a mere variant,60 whereas in the text he prints the Hebrew equivalent (“my expounding”) of an Arabic form with first-person pronoun (talh˚ ı¯søı¯).61 On this reading the object of the exposition is left unspecified, which allows the possibility that the object may well be the Metaphysics itself rather than Alexander’s analysis thereof, provided of course that the phrase “the passage best suited for my exposition” is taken as referring to the best possible place for the exposition – i.e. to the present passage in Ibn Rušd’s own commentary –, and not to its best possible object. The meaning of statement (ii) would then be: “and the place best suited for inserting my own exposition [scil. of the Metaphysics] is, perhaps, the present one.” In reality the manuscript authority for his reading is insufficient already within the Hebrew branch of the transmission. According to Freudenthal’s negative apparatus, the third-person variant (“its expounding = expounding it”) is transmitted by four of his six manuscripts of the second Hebrew translation,62 which implies that the first-person variant adopted in his Hebrew text (“my expounding”) is transmitted only by the remaining two.63 We must add that the third-person variant is also confirmed by the first Hebrew translation.64 What is more: when Siegmund Fränkel collated the codex unicus 59 60 61

62 63 64

Freudenthal (see note 4), 119‒120. Freudenthal (see note 4), 120: last entry in the critical apparatus, referring to line 11. Cf. Bouyges 1948a (see note 32), 1395 note 42 to line 10, who refers to text (= f) and apparatus (= f*) of the specimen of the second Hebrew translation in Freudenthal (see note 4) 119‒120. Parisinus hebr. 888 (Freudenthal: B, Bouyges f), Parisinus hebr. 890 (Freudenthal: C), Parisinus hebr. 889 (Freudenthal: D), and Taurinensis hebr. 14 (Freudenthal: T). Parisinus hebr. 887 (Freudenthal: E, Bouyges d) and Mich. hebr. 441 (Freudenthal: O). See the text of the first translation (Freudenthal: y, Bouyges: ‫ )ﬡ‬as transmitted by Parisinus hebr. 886 (Freudenthal: A, Bouyges: a), in Freudenthal (see note 4), 119.

26

Matteo Di Giovanni and Oliver Primavesi

of the Arabic original, it became clear that the unequivocal reading of this manuscript (bi-talh˚ ı¯søihi) confirms the third-person variant attested by the majority of the Hebrew manuscripts. For this reason Freudenthal was prompted to re-introduce, in the printed version of his German translation of statement (ii), a pronoun (“es”) referring to the object of the exposition. At the same time he gladly adopted, in statement (i), the conjectural substitution of ihø tima¯l by igˇma¯l suggested to him by Fränkel, with the result that the problem inherent in Alexander’s sketch would not be its equivocity but rather its summariness.65 Provided that Alexander had described the contents of the Metaphysics too summarily, Freudenthal deemed it plausible that Ibn Rušd decided to expound or rather explicate this description precisely by supplying a more detailed analysis of the Metaphysics of his own. Furthermore, Freudenthal remained convinced that “the passage best suited for the exposition” refers to the best possible place for the exposition. Relying on these two considerations, he felt entitled to maintain his initial interpretation even after abandoning the Hebrew manuscript reading on which it was based in the first place. This is shown by his German translation as finally printed:66 “(i) Das nun, was er zum Verständisse des Inhaltes der übrigen Bücher dieser Wissenschaft gesagt hat, ist summarisch ausgedrückt, (ii) und vielleicht ist hier der passendste Ort, e s [i.e. das was er zum Verständisse des Inhaltes der übrigen Bücher gesagt hat] zu erläutern.” Freudenthal even quoted this translation as evidence for his assumption that what follows in the next paragraphs is a summary of the Metaphysics composed by Ibn Rušd himself.67 We conclude that Freudenthal’s exclusion of Part III from his fragment collection does by no means stand or fall on his one editorial choice in the edition of the Hebrew versions, as claimed by Bouyges.68 In fact, Freudenthal’s six stated reasons for his decision may have suggested his initial adoption of the variant reading in the Hebrew text, but they do not depend upon it. Hence we need to examine these six reasons in their own right. We will 65

66 67

68

The emendation igˇma¯l is accepted by Martin (“un aperçu général”) and Genequand (“a summary”). Ramón Guerrero (see note 36), 288 n. 68, by contrast, keeps the reading ihø tima¯l transmitted by the Leidensis B (which he misleadingly calls “la variante de B”). Freudenthal (see note 4), 69. Freudenthal (see note 4), 122: “Dass hier [scil. in the Latin translation of the concluding sentence 1405,9–12 B.] mit Unrecht Averroes’ Epitome der Metaphysik in die Übersetzung eingeschwärzt wird, ergiebt fr. 1 (S. 69).” Bouyges (see note 4), CLXXVIII, n. 7: “A ceux qui demanderaient pourquoi J. Freudenthal n’a pas inclus l’Analyse parmi les « Fragmente Alexanders », je me contente de signaler la variante adoptée par lui p. 1395, 1042. Elle nous dispense, je crois, de supposer une raison tirée de la critique interne.”

Who Wrote Alexander’s Commentary on Metaphysics Λ?

27

do this by discussing, in each section, the potential evidence in favour of or against each one of Freudenthal’s points. § d Subject and Structure of the Metaphysics: 1395,11‒1396,5 B. “And so we say: Since this study inquires into being qua being, and this makes it necessary to inquire into both the principles of being qua being and the entities attributed to it (al-umu¯r al-la¯hø iqa lahu) ‒ for every theoretical study comprises precisely these two kinds of knowledge ‒ this study divides primarily into two parts. Moreover, since the Ancients set forth false views concerning the principles of beings, and it is incumbent upon [Aristotle] to oppose those [views], it follows that this is like a third part of this study. Therefore, the primary69 parts of this treatise are three: one part dealing with being qua being, another part dealing with the attributes (lawa¯hø iq) of being qua being, and yet another part dealing with the false views that were set forth with regard to the principles of being.” In this paragraph, we encounter the first item from Freudenthal’s list of suspicious features:70 As Freudenthal correctly observed, the analysis starts, in 1395,11 B., with the phrase “and so we say”, which must refer to a speechact by Ibn Rušd, not by “Alexander”. In the preceding paragraph c, however, Ibn Rušd has announced a passage from “Alexander’s” commentary which is best suited for being expounded. It would seem that the natural next step is to provide that exposition, which it is quite appropriate to introduce by means of the expression: “And so we say.” This expression, then, cannot cast doubts on the natural assumption that the section thus introduced contains Ibn Rušd’s exposition of “Alexander’s” analysis. § e Structure of the Metaphysics, continued: 1396,5‒1397,2 B. “At the same time, since every theoretical study divides into two kinds of inquiry, one of which comprises the mode of inquiry [typical] of that study – what causes it provides, where it starts and where it ends, and how definitions are used therein, this being that which is called logic specific (mantøiq h˚ a¯sø sø ) to that study –, whereas the second kind consists in knowing the contents of that study, this discipline, furthermore, divides primarily into two parts: one logical, specific to [the discipline], and one comprising that whose knowledge is aimed at therein. Conse-

69 70

This qualification is omitted in the two Hebrew manuscripts constantly used by Bouyges, i.e. a and d, see his note 48 on 1396,3. Cf. Freudenthal (see note 4), 123: “Dass aber das Nachfolgende die Erörterungen des Averroes, nicht die des Alexander enthält, erweist das ‫‘ וכאמר‬und wir sagen’, womit der folgende Abschnitt beginnt …”

28

Matteo Di Giovanni and Oliver Primavesi

quently, there are two major parts, one of which divides into three, so that the parts of this study come to be four. Moreover, since this study is in charge of counter-arguing against those who suppress inquiring and deny its principles, this comes to be as it were a fifth part of it; after all, this study must, insofar as it inquires into the subdivisions of beings and one of those is mental beings, inquire also into the principles of this kind of beings and knock down the false views set forth with regard to them.” Paragraph e brings the first section of Part III to a close. According to Ibn Rušd’s exposition, “Alexander” has shown that a satisfactory inquiry into being qua being must consist of the five parts mentioned. This result, however, would be rather pointless in a commentary on Aristotle’s Metaphysics if it were not to provide a criterion for testing the completeness of the work. So the section that ends here naturally introduces the following book-bybook analysis of the Metaphysics which shows that this work does actually meet the aforementioned requirements and which Ibn Rušd will expound in the second section of Part III, without hinting at any source other than “Alexander”.71 § f The Perfect Order of the Metaphysics (Announced): 1397,3‒4 B. “When Aristotle inquired into these kinds [of subjects] he chose the best arrangement for the purposes of instruction.” Paragraph f introduces the second section of Part III. Ibn Rušd makes clear that “Alexander”, in the second section of his summary, will not merely establish that Aristotle’s Metaphysics is a complete treatment of being qua being by the standards set in the first section. In addition, “Alexander” aims at demonstrating that Aristotle has arranged the books in the order best suited for purposes of instruction, and he will attain this goal thanks to the specific way in which he actually conducts the book-by-book-analysis.72 The remark is of considerable importance for our problem: given that “Alexander’s” main objective was to defend Aristotle’s book sequence, it would be rather odd if he himself had tampered with the very evidence for his claim,

71

72

On the conceptual homogeneity of Part III a) and III b), presenting a coherent understanding of the structure of the Metaphysics, see Di Giovanni, M. 2014: The Commentator. Averroes’s Reading of the Metaphysics, in: F. Amerini/G. Galluzzo (eds.), A Companion to the Latin Medieval Commentaries on Aristotle’s Metaphysics, Leiden, 82‒88, and, in particular, the chart ibid. 89. The motif of taxis is found also in the Greek Alexander, e.g. in his commentary on Metaph. Δ, p. 11: “That this is also the proper place (taxis) for the book [in the Metaphysics] is obvious from the fact that …”

Who Wrote Alexander’s Commentary on Metaphysics Λ?

29

that is, with the transmitted sequence of books, and – what is more – without explicitly saying so. § g Book Alpha (Alif) Minor, i.e. Book 1: 1397,5‒13 B. “In the first book, namely that whose title is the letter Alif Minor,73 [Aristotle] begins by making known to what extent this discipline is difficult and to what extent it is easy. He makes known that, if one brings together what every person has attained of [the discipline], something of import will result from it. And since a demand most incumbent upon the expert of this discipline is that through his work the inquiry into causes should come to an end (for [causes] are what is sought of existents), he shows in this book – after the beginning already mentioned – that the causes do come to an end. Subsequently he concludes [the book] by making known the principal causes of error regarding theoretical matters, and especially regarding the causes: that is, the false beliefs that people grow up74 with and are laid down for their benefit as the law75 in theoretical matters.” § h Book Alpha (Alif) Major, i.e. Book 2: 1397,14‒1398,7 B. “Since the inquiry into causes requires that he make known how many their first genera are and what has been said on the subject by his predecessors, he begins in the book whose title is the letter Alif Major76 to make known what the Ancients have attained with regard to the first genera of causes.77 He shows that they are four, and advances – by way 73

74

75

76

77

The Arabic elative søug˙ra¯ translates the Greek comparative ἔλαττον. This feature is captured better by the Latin comparative “minor” employed by Bouyges than by Martin’s “petit”, Genequand’s “small”, and Ramón Guerrero’s “pequeña”. “grow up”: reading with the Leiden MS (1397,12) yanšaʾuna (or yanšuʾuna) – variant spelling for yanšaʾu¯na (or yanšuʾu¯na) – as distinct from the text printed by Bouyges (yunšiʾu¯na), “build”, followed e.g. by Genequand (see note 32), 61: “namely what people build upon of the false views”. Clearly “Alexander” takes up Aristotle’s remark in Metaph. α 3, 995a5‒6 concerning the mythical elements that, through the laws, become familiar since childhood (παιδαριώδη … διὰ τὸ ἔθος). Cf. also Alexander’s comments on Alpha Minor (167,14‒20 Hayduck) particularly as regards legislators who keep citizens obedient to the law by means of tales familiar from childhood (διὰ τὴν τῶν λεγομένων συντροφίαν). “Law” (šarı¯ʿa) refers to Metaph. α 3, 995a3‒4: ἡλίκην δὲ ἰσχὺν ἔχει τὸ σύνηθες οἱ νόμοι δηλοῦσιν. The Arabic šarı¯ʿa means specifically “religious law”; in the present context it appears to be merely a cultural adaptation which originated within the Arabic tradition of the “Alexander” that is quoted by Ibn Rušd. The Arabic elative kubra¯ translates the Greek comparative μεῖζον. This feature is captured better by the Latin comparative “maior” employed by Bouyges than by Martin’s “grand”, Genequand’s “great”, and Ramón Guerrero’s “grande”. The expression “genera of causes” (agˇna¯s al-asba¯b) is a translation of the Greek γένη αἰτίων such as we find e.g. in Alexander’s commentary on Metaph. Δ 2 (352,10–16 Hayduck): Τρόπους τῶν αἰτίων λέγει τὰς κατὰ τὴν ἀπόδοσιν αὐτῶν διαφοράς, ὡς καὶ αὐτὸς ἐδήλωσεν

30

Matteo Di Giovanni and Oliver Primavesi

of corroboration – the fact that whatever causes [the Ancients] spoke about do not, upon examination, exceed the four causes. Subsequently he relates what [the Ancients] thought on the subject of the causes of the existents with special regard to those who practiced metaphysics rather than physics ([physicists’ views] being related in physics), namely those who claimed that the principles were the mathematical entities and the numbers. Some refutation of them is given in this book, indeed, but the full argumentation in refutation of them is delayed until the two books following the letter La¯m, namely Books Mı¯m and Nu¯n.” A first observation to be made here is that in paragraph h the contents of Book Alpha Major, Chapters 3‒7 (on the treatment of the four causes in Aristotle’s predecessors) is correctly analysed despite the fact that the only Arabic translation of Alpha Major used by Ibn Rušd, i.e. the translation by Naz ı¯f b. Ayman, does not begin before the end of A 5 (987a6), and leaves out Chapter 7 altogether. This fact invalidates Freudenthal’s second point according to which the contents of the analysis show that it was written by Ibn Rušd.78 A second observation concerns Freudenthal’s fourth point, which is perhaps the most crucial:79 in paragraphs g and h, Book Alpha Minor comes first and Book Alpha Major second; at the beginning of paragraph g Alpha Minor is even explicitly called the “first book”.80 This is not the order in which the books are both transmitted in our Greek manuscripts and commented upon in the extant part of Alexander’s genuine commentary (on Books Α ‒ α ‒ Β ‒ Γ ‒ Δ); the sequence of the two paragraphs rather corresponds to the order adopted by Ibn Rušd in his own Tafsı¯r. This discrepancy constitutes Freudenthal’s most important piece of evidence for his contention that the analysis here expounded was written by Ibn Rušd and not by Alexander of Aphrodisias. Before we can decide whether or not to side with Freudenthal, we will have to take into account what Alexander himself, in

78 79

80

ἐπενεγκὼν “λέγεται γὰρ τὰ αἴτια πολλαχῶς”. κατὰ μὲν οὖν τὸν ἀριθμὸν πλείους φησὶν εἶναι τοὺς τρόπους, ἀναγομένους μέντοι εἰς τὰ κοινὰ καὶ καθόλου ἐλάττους γίνεσθαι, ὥστε εἰς ἓξ ἀνάγεσθαι γ έ ν η α ἰ τ ί ω ν πάντας τοὺς τρόπους, ὧν ἕκαστον διχῶς λέγεσθαι, ἢ κατ’ ἐνέργειαν ἢ κατὰ δύναμιν, ὡς δείξει. πρῶτον δὲ τίνες οἱ τρόποι τῶν αἰτίων καὶ ὅτι πλείους δείκνυσι. Cf. Freudenthal (see note 4), 123: “Dass aber das Nachfolgende die Erörterungen des Averroes, nicht die des Alexander enthält, erweist … auch der Inhalt desselben …” Cf. Freudenthal (see note 4), 123: “Dass aber das Nachfolgende die Erörterungen des Averroes, nicht die des Alexander enthält, erweist … die Voraufstellung des zweiten Buches (α) vor Α …” In tune with p. 1395,4 B. where Book Alpha Major is explicitly called the “second book” of the Metaphysics, as we have seen.

Who Wrote Alexander’s Commentary on Metaphysics Λ?

31

the genuine part of his commentary, has to say about the sequence of the first two books. Concluding his commentary on Alpha Major, in fact, Alexander states that it might seem (δόξαι ἂν) that – as far as one can judge from what Aristotle says at the end of Alpha Major – the sequel to Alpha Major is Beta, although the announcement made in the final lines of Alpha Major is also taken up “in the following book, i.e. in Alpha Minor”.81 In his commentary on Alpha Minor, however, Alexander goes a step further. There he says that it will seem (δόξει) that Book Beta is better suited (scil. than Book Alpha Minor) to follow Book Alpha Major.82 What is more, he observes that what Aristotle claims in Book Alpha Minor, while being not completely irrelevant to the Metaphysics, “is like a preface and prolegomena to theoretical philosophy as a whole.”83 This statement clearly shows that Alexander has now become more sympathetic to the idea that Beta was meant to be the immediate continuation of Alpha Major. But it certainly does not prove also that he deemed Alpha Minor to be the true first book of the Metaphysics. Alexander rather believes that the “physical” conclusion of Alpha Minor might suggest that we are dealing here with an introduction to the Physics.84 It is true that at the end of his commentary on Alpha Minor Alexander considers a possible alternative: the conclusion of Alpha Minor might just be meant to remind us that the study of first philosophy presupposes the study of physics. But even there, Alexander does not hint at the possibility of regarding Alpha Minor as the first book of the Metaphysics.85 81

82 83

84

85

Alexander in Metaph. 136,13‒15 Hayduck: διὸ καὶ δόξαι ἂν ‒ ὅσον ἐκ τῶν νῦν εἰρημένων ‒ συνῆφθαι τούτῳ τῷ βιβλίῳ τὸ Β. οὐ μὴν ἀλλὰ λέγει γε περὶ ὧν ἐνταῦθα προτίθεται καὶ ἐν τῷ ἐλάττονι α. Alexander in Metaph. 138,2‒3 Hayduck: διὸ δόξει τῷ μείζονι Α τὸ Β μᾶλλον ἀκολουθεῖν· συνῳδὸς γὰρ ἡ τούτου ἀρχὴ τῷ ἐκείνου τέλει. Alexander in Metaph. 138,6‒9 Hayduck: οὐ μὴν οὐδὲ τὰ ἐν τούτῳ λεγόμενα ἀλλότρια πάντῃ τῆς προκειμένης πραγματείας εἶναί μοι δοκεῖ, ἀλλ’ ἔστιν ὥσπερ τινὰ προγραφόμενα καὶ προλεγόμενα κοινῶς πάσης τῆς θεωρητικῆς φιλοσοφίας τὰ ἐν τούτῳ λεγόμενα. Cf. Alexander in Metaph. 169,21‒26 Hayduck (ad 995a17–19 διὸ σκεπτέον τί ἐστιν ἡ φύσις· οὕτω γὰρ καὶ περὶ τίνων ἡ φυσικὴ δῆλον ἔσται): εἰ μὲν ἁπλῶς οὕτως εἴη εἰρημένον ὡς δοκεῖ, δεικτικὸν ἂν εἴη τοῦ τοῦτο τὸ βιβλίον μὴ τῆς Μετὰ τὰ φυσικὰ εἶναι πραγματείας, ὡς καὶ ἀρχόμενοι αὐτοῦ ἐπεσημηνάμεθα, ἀλλὰ προγραφόμενόν τι πάσης τῆς θεωρητικῆς φιλοσοφίας, ἧς ὡς πρὸς ἡμᾶς πρώτη ἡ φυσική, ἣν ὁποία τίς ἐστι καὶ περὶ τίνων, εὕροιμεν ἂν εἰ πρῶτον τί ποτέ ἐστιν ἡ φύσις ἐπισκεψαίμεθα. Alexander in Metaph. 137,12‒15 Hayduck: ὅσον δὲ πάλιν ἐπὶ τῷ τέλει αὐτοῦ, οὐ δόξει τοῦτο ἐκ ταύτης εἶναι τῆς συντάξεως, ἀλλὰ τῆς φυσικῆς πραγματείας προοίμιόν τι· ὡς γὰρ περὶ φύσεως ἐν τοῖς ἑξῆς ἐρῶν, καὶ ζητήσων τί ἐστιν ἡ φύσις, ἐπαύσατο. Alexander in Metaph. 169,26‒170,4 Hayduck: εἰ δέ τις ἀκούοι τοῦ λόγου οὕτως εἰρημένου πρὸς διάκρισιν τῶν τε φυσικῶν λόγων καὶ τῶν κατὰ τήνδε τὴν πραγματείαν (ὁ γὰρ ἐπεσκεμμένος πρῶτον τί ἐστι φύσις καὶ εἰδὼς περὶ τίνων ἡ φυσική, οἶδεν οὗτος ὅτι οἵδε οἱ λόγοι οὐ φυσικοὶ ἀλλ’ ἀκριβέστεροί τε καὶ περὶ ἀύλων), οὐχ ὅτι δὲ χρὴ νῦν ἐπισκέπτεσθαι τί ἐστι φύσις λέγοι ἄν,

32

Matteo Di Giovanni and Oliver Primavesi

Others, however, came much closer to actually reversing the order of the two “first” books and providing positive reasons in favour of the sequence Alpha Minor ‒ Alpha Major, as we know from Asclepius. At the beginning of his commentary on Alpha Minor Asclepius first summarizes the current view that Alpha Minor is misplaced. Against this background he affirms in his own name (i.e. in the name of his teacher Ammonius) that Aristotle wants to treat the first causes in the first place (πρότερον) and that he does so “here”, i.e. in Alpha Minor.86 At this point, Asclepius adds a reference to other, unnamed authors who have interpreted (εἰρήκασιν) the name “Alpha Minor” as designating the beginning of what is the real first book of the Metaphysics, i.e. of Alpha Minor + Alpha Major taken together: clinging to the (“physical”) conclusion (πρὸς τῷ συμπεράσματι) of Alpha Minor, Aristotle will turn in Alpha Major to the treatment of the physical causes.87 The original contribution reported here would appear to be that the “physical” conclusion of Alpha Minor does not introduce the Physics proper, as Alexander suggested, but the treatment of the four “physical” causes in Alpha Major. We can, of course, not exclude the possibility that this suggestion ultimately goes back to Alexander’s own lost introduction to Lambda, provided that his views had further developed by the time when he wrote it. But what Alexander explicitly says in the extant parts of his commentary does not provide the slightest evidence for this assumption. Furthermore, it seems unlikely that Alexander should have changed his mind on this fundamental question without in any way explaining his reasons, or that Ibn Rušd should

86

87

ἀλλ’ ὅτι πρὸ τῆσδε τῆς πραγματείας δεῖ περὶ τῶν φυσικῶν πεπραγματεῦσθαι· αὕτη γὰρ ἡ τάξις τῶν πραγματειῶν, καὶ ὁ ἐν τῇ περὶ ἐκείνων θεωρία γεγυμνασμένος οὕτως ἂν καὶ τοῖς εἰς τήνδε συντείνουσι παρακολουθεῖν δύναται. Asclepius in Metaph. 113,16‒20 Hayduck (with omissions): λέγομεν οὖν πρὸς ταῦτα ὅτι ἠβουλήθη πρότερον περὶ τῶν πρωτίστων ἀρχῶν διαλεχθῆναι … διὰ τοῦτο τοίνυν ἐνταῦθα πρότερον περὶ τῶν πρωτίστων ἀρχῶν διαλέγεται. Asclepius in Metaph. 113,20‒23 Hayduck: ἀμέλει τοι καὶ διὰ τοῦτο εἰρήκασιν ἔλαττον αὐτὸ Α προσαγορεύεσθαι, ὡς ἂν ἀρχῆς οὔσης τ ο ῦ ὄ ν τ ω ς π ρ ω τ ί σ τ ο υ β ι β λ ί ο υ (our emendation of the corrupt manuscript reading τοῦ †ὄντον πρωτίστου τοῦ βιβλίου; Hayduck’s emendation τοῦ ὄντος πρωτίστου τοῦ βιβλίου is unconvincing), ἐπειδὴ περὶ τῶν πρωτίστων ἀρχῶν διαλέγεται, εἶτα λοιπὸν πρὸς τῷ συμπεράσματι διαλέγεται καὶ περὶ τῶν φυσικῶν ἀρχῶν. VuilleminDiem, G. 1983: Anmerkungen zum Pasikles-Bericht und zu Echtheitszweifeln am größeren und kleineren Alpha in Handschriften und Kommentaren, in: P. Moraux/J. Wiesner (eds.), Zweifelhaftes im Corpus Aristotelicum. Studien zu einigen Dubia. Akten des 9. Symposium Aristotelicum, Berlin–New York, 180–181 with n. 72 urges that Asclepius does not subscribe to this theory himself, whereas Jaeger, W. 1912: Studien zur Entstehungsgeschichte der Metaphysik des Aristoteles, Berlin, 116 with n. 1 ascribes such a view to Asclepius (or rather Ammonius) on the basis of Asclepius in Metaph. 114,16‒17 Hayduck: οὐκ ἔστιν οὖν, ὡς εἴρηται, ἀλλότρια τὰ νῦν λεγόμενα τῆς προκειμένης πραγματείας, ἀλλ’ ἔστι π ρ ο γ ρ α φ ό μενα.

Who Wrote Alexander’s Commentary on Metaphysics Λ?

33

have omitted these reasons in his own account of Alexander, considering that the whole point of the analysis is precisely to defend the order of the books of the Metaphysics established by Aristotle. Our conclusion will therefore be that paragraphs g and h were not the work of Ibn Rušd, as evidenced by the well-informed reference to the treatment of the four causes in the missing first part of Alpha Major. At the same time the two paragraphs attest to a remarkable elaboration on the view surfacing in Alexander’s extant commentary as regards the relationship between Alpha Major and Alpha Minor: Alexander already considered that the most suitable sequel to Alpha Major appeared to be Beta, not Alpha Minor. But whereas he tentatively suggested considering Alpha Minor as an introduction to the Physics, or theoretical philosophy in general, Asclepius reports that Alpha Minor was regarded as a suitable introduction to Alpha Major itself. This latter view is presupposed in paragraphs g and h which, therefore, seem to go back to a later version of Alexander’s commentary, revised under the influence of views such as the one reported by Asclepius. § i Book Beta (Ba¯ʾ): 1398,8‒1399,8 B. “Since every scientific investigation comes to completion only if one has prior knowledge of dialectical arguments aiming at affirmation and negation (for one cannot realize how much knowledge of a thing one has acquired after being ignorant nor how much knowledge one lacked before acquiring it, unless one poses questions88 to oneself about it), [Aristotle] considers that the best for the purposes of instruction is carefully to examine in a single book the subjects of investigation of this discipline specifically in connection with the dialectical and aporetic arguments (alaqa¯wı¯l al-gˇadaliyya al-mušakkika) relating to each of such subjects. Subsequently, after doing so, in the other books of this treatise he begins to resolve the questions that arise in this discipline. This he does in the third book of this treatise, that whose title is the letter Ba¯ʾ. Necessarily this book follows the first and the second books as much as it precedes the other books: it follows the two preceding books because the [theses] discussed in them, too, are among those which the discipline of dialectic posits89 (i.e., that there are four causes and that for each of their genera 88

89

“Questions” here as in what follows corresponds to the Greek aporia – Lat. quaestio (here for the sake of consistency “question” and its cognates seem to be the best terms, offering the necessary flexibility, for rendering the Arabic vocabulary connected to the root š-k-k); cf. Ramón Guerrero (see note 36), 290: “cuando no se han planteado aporías acerca de algo …” Neither “examine critically” (Genequand) nor “mettre en doute” (Martin) are precise enough. The Arabic wadø aʿa and its cognates correspond here to the semantic area of the Greek τιθέναι in the Aristotelian logical sense (“posit”). Cf. Afnan, M. A. 1969: A Philosophical Lexicon in Persian and Arabic, Beirut, ad loc.

34

Matteo Di Giovanni and Oliver Primavesi

an infinite regress is not possible); it precedes the following ones because, clearly, the following books contain nothing else but one of these two: either a solution of the questions mentioned in this book or some information about things necessary for solving the questions mentioned in this book.” Notably the treatment of the four causes in Aristotle’s predecessors is the only subject mentioned in connection with Book Alpha Major, although the chapters dealing with that subject (A 3‒7) are largely missing in the translation used by Ibn Rušd. ˇ ı¯m): 1399,9‒1400,8 B. § k Book Gamma (G “Now, the questions mentioned in this book are of two kinds: questions concerning the mode of inquiry of this discipline and questions concerning the subjects of investigation falling within it. And the first duty incumbent upon everyone who begins to inquire into this discipline in a demonstrative way is to have preliminary knowledge of how these questions [concerning the mode of inquiry] are resolved. For, if he knows this,90 he will realize the discipline’s mode of inquiry that is built on demonstration, and through this mode he will distinguish the study that is called “philosophy”. Accordingly, he considers it necessary to start off by resolving the questions concerning the mode of inquiry belonging to this study. Furthermore, since an inquiry is sound only when primary propositions are admitted, he considers it necessary to dispute first with those who knock down these propositions and deny inquiring. Consequently he makes both of these points in a single book, and he makes it follow the book of the letter Ba¯ʾ: this [following book] is that whose ˇ ı¯m. As a result this book includes two kinds,91 one of title is the letter G which is the logic specific (al-mantøiq al-h˚ a¯sø sø ) to this discipline, whereas the other is the establishment of the first principle among those which we have by nature, I mean that which is prior to all. This [principle] is [Aristotle’s] claim that affirmation and negation do not simultaneously hang together because that is the principle of inquiry, so much so that no arguing, proving or disproving will be sound from anyone who denies it.”92 90

91 92

In the Arabic it is indeterminate whether the pronoun grammatically refers to the questions themselves (thus Martin: “la connaissance de ces problèmes”; Genequand: “knowing them”) or the solution thereof, as implied by the sense. In 1400,4 B., one may consider emending the attested gˇinsayni (“two kinds”) to gˇuzʾayni (“two parts”). Cf. Alexander’s commentary on Metaph. Γ where Alexander claims similarly to what is said here that in Metaph. Γ Aristotle resolves some aporiae related to the status of metaphysics as a science (246,13–24 Hayduck as translated by Arthur Madigan, S. J. [1993] 21):

Who Wrote Alexander’s Commentary on Metaphysics Λ?

35

The summary of Book Gamma given here is strikingly focused on the two subjects (four and five) postulated above in paragraph e, i.e. the logic specific (mantøiq h˚ a¯søsø) to metaphysics93 and counter-arguing against those who suppress inquiring and deny its principles. Such coherence provides additional evidence for the unity of the two sections (IIIa and IIIb) of the analysis. § l Book Delta (Da¯l): 1400,9‒1401,1 B. “Once he is done with this point, in this book, he considers it most necessary that this book be followed by a division of the meanings with which nouns are used in this discipline, and that the best for the purposes of instruction is to mention them in a single separate book. This he does in the book of the letter Da¯l, which is placed after the book of the letter ˇ ı¯m but before the other books for the reason that the first [task] to G start from for anyone who is determined to demonstrate a given notion that is sought is to analyse the meanings of the noun that is said of that notion, if it has more than one meaning, and especially those nouns which are said by analogy (bi-tana¯sub), namely the notions specific to this discipline, given that the noun “being” (mawgˇu¯d) is certainly said by analogy as he shows in this discipline.” § m Book Epsilon (Ha¯ʾ): 1401,2‒1402,3 B. “Once he has achieved his purpose in this book, it is necessary for him to examine that which is sought in this discipline, which is acquiring knowledge of the causes of beings (huwiyya¯t), and since the kinds of beings are three (the being that is by accident, that which is in the soul, and that which is outside the soul), he considers it necessary to begin first by making known the kinds of these beings, as well as the fact that that whose examination is conducted primarily falls within the kinds of beings that are outside the soul. He examines this point in the book that follows the book of the letter Da¯l. Since in this [following] book his inquiry con-

93

“Through the things he has said and established, he has solved some of the aporiae mentioned in Book Beta: whether it belongs to the same science to carry on the consideration of all substances, or whether there is one science of one kind of substance, another science of another kind of substance [Met. Β 1, 995b10–13; Β 2, 997a15–25]; further, the aporia whether it belongs to the same science to consider substance and to consider the accidents of substance [Met. Β 1, 995b18–20; Β 2, 997a25–34]. … And he next solves the next point of aporia, the one concerning sameness and otherness, likeness and unlikeness, and concerning contrariety for he shows that consideration of these things is proper to the science concerned with being.” On the notion of “specific logic” (mantøiq h˚ a¯søsø), cf. Di Giovanni (see note 71), 84‒88. Cf. also Hoffmann, R. 1991: La puissance argumentative de la logique spéciale dans la métaphysique d’Ibn Rushd, in: M. A. Sinaceur (ed.), Penser avec Aristote, Toulouse–Paris, 667–676.

36

Matteo Di Giovanni and Oliver Primavesi

cerns the distinction of the being that is by accident and in the soul from the being that is real (al-huwiyya al-hø aqı¯qiyya), and since that [distinction] can only be made by those who know that this study is that which inquires into all kinds of beings qua being, at the beginning of this book ˇ ı¯m by way of he recalls what has become clear in the book of the letter G a reminder, not as a repetition, and from new angles: namely that this study is that which inquires into being qua being, and that it is different from both physics and mathematics. He does all this in the book that follows that of the letter Da¯l, namely that one where he analyses the meanings of nouns: that is found in the book of the letter Ha¯ʾ which he places after that of the letter Da¯l. That this book [Ha¯ʾ] precedes that which follows it, and follows what precedes, is also necessary.” § n Book Zeta (Wa¯w) and Eta (Za¯y): 1402,4‒1403,10 B. “Once he has shown that the being that is by accident and that which is in the mind are defective (na¯qisøata¯n), and that the [being] whose examination is conducted by this discipline is the real being (al-huwiyya alhø aqı¯qa) that exists outside the soul, he begins after this to examine this being. Since substance is the principle of this being, he undertakes to investigate what the principles of substance are, and he starts this [investigation] from the principles of the substance generated and corruptible. He makes known that [these principles] are form and matter, and he traces a path leading to knowledge of the fact that forms are substances and ensuing from definitions. That is, since he shows that definitions indicate something of sensible substances which is substance to them, and that forms is what they indicate, from this he gathers that forms are substances, that [forms] are one and the same with the thing of which they are a form, and that for this reason accidents are not substances, nor does knowing things require introducing separate forms other than the sensible because what is indicated by the definitions of things would be other than the things [defined]. Likewise he shows also that this substance generated and corruptible can only be generated by the agency of a substance generated and corruptible that is alike in species and genus; that what is generated and corruptible is the thing composed of form and matter, whereas forms are generated and corruptible only by accident; that for this reason the Platonic Forms, supposing that they exist, are of no use in generation (I mean the separate forms that are professed by Plato). Moreover he shows that the universals are not substances existing outside the soul even though they indicate substances; that forms are substances not in the sense of being an element or being composed from some element, but rather in the sense of being a third substance, and he spells out the difference between the two substances. Subsequent-

37

Who Wrote Alexander’s Commentary on Metaphysics Λ?

ly he shows the primary differences of substances94 and discriminates the parts of forms from those of matter. He performs all this in the book whose title is the letter Wa¯w and in that whose title is the letter Za¯y.” In paragraph n we encounter for the first time a peculiarity that corresponds to the third point from Freudenthal’s list: the designation of the single books of the Metaphysics by means of unusual book-labels. The philosophical contents of the summary leave no doubt that the two books summarized here are to be identified with the books which in our Greek tradition bear the names Ζ and Η. This is only to be expected, given that the preceding paragraph (m) contained the summary of Book Ε. Yet in the present paragraph the two books in question are not labelled by means of the Arabic letters Za¯y and H  a¯ʾ, as customary to Ibn Rušd, but rather by means of the two Arabic letters Wa¯w and Za¯y. Freudenthal assumed that these labels correspond to the “usual Arabic way of counting books”, and concluded that they yield further evidence for his claim according to which the summary of the Metaphysics was composed by Ibn Rušd himself.95 But that argument will not stand close scrutiny. In order to see what is really at play here, it will be useful to recall 96 that the extant works of Aristotle and Theophrastus share with the poetry of Homer the peculiarity that the individual books of which these works consist are not counted, but labelled by means of the letters of the ordinary alphabet, which is best envisaged as consisting of three rows of eight letters each: Α Ι Ρ

94

95

96

Β Κ Σ

Γ Λ Τ

Δ Μ Υ

Ε Ν Φ

Ζ Ξ Χ

Η Ο Ψ

Θ Π Ω

While the adjective “primary” in the Arabic can in principle qualify either “differences” or “substances”, the former construal seems preferable in light of the passage in H 2, 1042b31–1043a1 that is alluded to here, where it is a question of tracing the differentiae in things to their primary pairs (transl. by W. D. Ross [1908]): “We must grasp, then, the kinds of differentiae (for these will be the principles of the being of things), e.g. the things characterized by the more and the less, or by the dense and the rare, and by other such qualities; for all these are characterized by excess and defect. And everything that is characterized by shape or by smoothness and roughness, is determined by the straight and the curved. And for other things their being will mean their being mixed, and their not being will mean the opposite.” Cf. Freudenthal (see note 4), 123: “Dass aber das Nachfolgende die Erörterungen des Averroes, nicht die des Alexander enthält, erweist … die bei den Arabern übliche Zählung der Bücher …” The evidence for the following remarks on Greek book labelling and book counting is set out in Primavesi, O. 2007: Ein Blick in den Stollen von Skepsis: Vier Kapitel zur frühen Überlieferung des Corpus Aristotelicum, in: Philologus, 151, 51–77.

38

Matteo Di Giovanni and Oliver Primavesi

By contrast, the individual books of all other works of Classical and Hellenistic literature are counted, by means of the Hellenistic system of alphabetic numerals. This system was created by re-introducing obsolete or introducing foreign signs which did not form part of the Ionic alphabet which became the standard Greek alphabet during the 4th century BC – F (Fαῦ, δίγαμμα),97 Ϙ (κόππα),98 and ϡ (σάμπι?),99 respectively – in each of the three rows of the ordinary alphabet. The result is a quasi-decimal system, consisting of three rows with nine letters each, one row for the units, one row for the tens and one row for the hundreds: Α 1

Β 2

Γ 3

Δ 4

Ε 5

F (ϛ) 6

Ζ 7

Η 8

Θ 9

Ι 10

Κ 20

Λ 30

Μ 40

Ν 50

Ξ 60

Ο 70

Π 80

Ϙ 90

Ρ 100

Σ 200

Τ 300

Υ 400

Φ 500

Χ 600

Ψ 700

Ω 800

ϡ 900

In principle it would be possible to replace the Peripatetic letter labels traditionally used for the works of Aristotle and Theophrastus by alphabetic numerals, which would involve a change from the sixth book onwards (the book which used to be labelled as Ζ would have to count as F, book Η would have to count as Ζ, and so forth).100 But in fact, the Peripatetic letter labels survived well into the Arabic reception of Aristotle’s Metaphysics, a fact nicely captured by the traditional Arabic characterization of the Metaphysics as “Book of Letters”.101 When Usta¯t, in the first half of the 9th cenJeffery, L. H. 21990: Local Scripts of Archaic Greece. A Study of the Origin of the Greek Alphabet and its Development from the Eighth to the Fifth Centuries BC, Oxford, 24–25; Heubeck, A. 1979: Schrift, Archaeologia Homerica. Denkmäler und das frühgriechische Epos, Göttingen, X 89–90 (No. 6). 98 Jeffery (see note 97), 33–34; Heubeck (see note 97), X 89 (No. 4). 99 Jeffery (see note 97), 38–39. 100 In the case of the complete 14-book Metaphysics the inconvenience would be even greater, since in that treatise the letter label A is employed twice: Αlpha Major and Αlpha Minor. Hence, the introduction of alphabetic numerals would entail that all but one label be changed: Book Αlpha Major would have to count as Α, Book Αlpha Minor as Β, Book Β as Γ, Book Γ as Δ and so forth. 101 As is immediately apparent from the title of al-Fa¯ra¯bı¯’s treatise on the “Aims of the Philosopher” (Aristotle): Fı¯ Ag˙ra¯dø al-hø akı¯m fı¯ kull maqa¯la min al-kita¯b al-marsu¯m bi-l-hø uru¯f. (Bouyges (see note 4), CXXV). Cf. Bouyges (see note 4), CXXIV: “Ce nom [scil. “Livre des Lettres”] est celui qui vient en premier lieu dans le Fihrist lequel fait allusion à son origine, à savoir, que les Livres de la Métaphysique sont désignées en grec par des lettres de l’alphabet”. What the Fihrist alludes to by “letters” was precisely letters of the Greek alphabet (hø uru¯f al-Yu¯na¯niyyı¯n) taken as alphabetical signs: thus Book La¯m is presented as the book 97

Who Wrote Alexander’s Commentary on Metaphysics Λ?

39

tury AD,102 prepared the translation of the Metaphysics into Arabic (probably leaving out Alpha Major and possibly Ny103), which was to become Ibn Rušd’s text of reference, he appears to have used a labelling system in which the Peripatetic letter labels – except for the two vowels Ε and Η – were rendered by what amounts to their approximate phonetic equivalents:104 Α Β Γ Δ Ε Ζ

= = = = = =

‫ا‬ ‫ب‬ ‫ج‬ ‫د‬ ‫ه‬ ‫ز‬

(Alif) (Ba¯ʾ) ˇ ı¯m) (G (Da¯l) (Ha¯ʾ) (Za¯y)

Η Θ Ι Κ Λ Μ

= = = = = =

‫ح‬ ‫ط‬ ‫ي‬ ‫ك‬ ‫ل‬ ‫م‬

 a¯ʾ) (H (Ta¯ʾ) (Ya¯ʾ) (Ka¯f) (La¯m) (Mı¯m)

Small wonder, then, that this system is presupposed not only in the headings of Ibn Rušd’s Tafsı¯r,105 but also in its main text. In his commentary on Lambda, in particular, he refers to Book Z106 by means of its phonetic equivalent

102

103 104

105

106

corresponding to letter 11 (al-hø a¯diya ʿašra min al-hø uru¯f) because La¯m corresponds to the 11 th letter of the Greek alphabet or, which is the same, the 11th letter used as an alphabetical sign; cf. also Bertolacci, A. 2005: On the Arabic Translations of Aristotle’s Metaphysics, in: Arabic Sciences and Philosophy, 15, 246 n. 13. On the designation of books in the Arabic Metaphysics see Bouyges (see note 4), XVIII‒XX. As apparent from Ibn al-Nadı¯m’s testimony, according to which Usta¯t worked for the affluent philosopher al-Kindı¯ (ca. AD 800‒870); see Endress, G./Adamson, P. 2012: Abu¯ Yu¯suf al-Kindı¯, in: U. Rudolph (ed.), Philosophie in der Islamischen Welt. Band 1: 8.–10. Jahrhundert, Basel, 100‒101. See Bouyges (see note 4), CXXVIII‒CXXIX and Bertolacci (see note 101), 246‒247. Relevant evidence of Usta¯t’s nomenclature is supplied by the codex Leidensis. The same scribe (‘V’ in Bouyges’ notation) who copied in the margins Usta¯t’s translation of Books α and Λ (the two books quoted from other translations in the main text of the commentary) added what appear to be Usta¯t’s book titles also for other books, labelling Book Z as Za¯y, Book Θ as Ta¯ʾ, and Book I as Ya¯ʾ; cf., respectively vol. 2 of Bouyges’ edition, p. 744; ibid., ˇ ı¯m (ibid., vol. 1, p. 296); Delta is p. 1103; vol. 3, p. 1235. Similarly, Gamma is named G  a¯ʾ (ibid., vol. 2, Da¯l (ibid., vol. 2, p. 473); Epsilon is Ha¯ʾ (ibid., vol. 2, p. 697); Eta is H p. 1022). Thus, in the title of the Tafsı¯r on Book H we find that H is called H  a¯ʾ (1022,2‒4). On the assumption that the titles might have been assigned by Ibn Rušd see Bouyges 1948a (see note 32), Index B, p. (24), N.B. 3°). The numerals added in the “Usta¯tian” marginal titles are lower by one unit than those found in the main text of Ibn Rušd’s commentary, apparently since Usta¯t did not translate Book Alpha Major. Cf. pp. 1417,14–16 B.: “[Plato], holding to this belief, posited the universals as principles and Forms for the sensible substance, and [Aristotle] has previously, in Book Za¯y, disproved the thesis that the universals are substances” (reference to Z 13); p. 1463,11‒12 B.: “[Aristotle] laid down in Book Za¯y that what comes into being by the agency of art is from things synonymous” (reference to Z 7–9); p. 1464,4–7 B.: “When [Alexander] states that Aristotle mentioned [the things subject to spontaneous generation] in the Physics, this is a mistake and God knows best. It appears, rather, that he mentioned them in Book Za¯y because the

40

Matteo Di Giovanni and Oliver Primavesi

Za¯y, and to Books Z H by means of the couple Za¯y H  a¯ʾ,107 the only seeming exception being clearly due to textual corruption: at p. 1576,5 B. what is being referenced by Ibn Rušd is, in all evidence, our Book Θ (i.e. Ta¯ʾ both in the Arabic rendering of the Peripatetic label and in the abgˇad-system when accounting for two “First books”; and H  a¯ʾ in the abgˇad-system, when accounting for one “First book” only), and yet the Arabic tradition offers Ya¯ʾ, which cannot refer to our Book Θ on any available system. We conclude that (pace Martin 1984, 202, n. 1) the two letters Ta¯ʾ (for Θ) and Ya¯ʾ must have been confused at some point in the Arabic tradition.108 Yet in places where Ibn Rušd either quotes or explains “Alexander” on individual passages of Book Λ, he tends to use a different system:109 he labels Books Z and H as Wa¯w and Za¯y (instead of Za¯y and H  a¯ʾ) 110 – just as in the

107

108

109

110

book that has come down to us contains some lacunae” (the reference is to the analysis of generation in Z 7 which in fact contains three lacunae, duly signaled by Ibn Rušd, in the Arabic translation corresponding to the Bekker lines 1032a29–b5; 1032b14–20; 1032b29– 1033a1). Cf. p. 1425,1‒4 B.: “Subsequently [the metaphysician] shows that the principles of the generated and corruptible substance are substances, and that the universals are not the substances of these [generated and corruptible substances] nor are the numbers and, in general, neither the Forms nor the mathematical entities. This is what he shows in Books  a¯ʾ.” Za¯y and H The same confusion, albeit in reverse direction, appears in a passage of Ibn Rušd’s commentary on Book Δ (543,7–9 B.): “Therefore the celestial body encompasses unity by continuity, universal form, and unitary continuous movement as [Aristotle] will say in what follows, in Book T a¯ʾ”. This book label would seem to correspond to the letter Θ, although the contents of the remark suggests that the reference is perhaps rather to Book Iota (Ch. 1,1052a25 ff.) than to Book Θ (Ch. 8). If that is correct, the error could again be explained by the assumption that the Arabic letters Ta¯ʾ and Ya¯ʾ were confused. A theoretical alternative might be the hypothesis that we are dealing with an isolated use of the abgˇad-system, a¯ʾ corresponds to 9. Yet Ibn Rušd could have called Book Iota the ninth book in which T only if, when commenting on Book Δ, he was still ignorant of the existence of Alpha major. This hypothesis, however, would not cohere with the book-labelling in Ibn Rušd’s commentary on Book Λ, where the coexistence of the two Alphas is clearly presupposed and where they are identified as the first and the second book, respectively, as we have seen. That is not to deny that even within a quotation from “Alexander”, he is also capable of keeping his usual habit of labelling Book Z as Za¯y; cf. 1459,3‒4 B.: “Subsequently Alexander says: [Aristotle] has discussed this [i.e. generation] at great length in Book Za¯y of this treatise” (reference to Metaph. Z 7‒9). Cf. on this passage Martin (1984, 96 n. 11): “Il est à noter qu’il faut ici donner à ZAY la valeur de Z, bien qu’il s’agisse d’une citation d’Alexandre, pour qui cette lettre désigne un autre livre […] Sans doute, Averroès a-t-il corrigé de lui-même.” 1407,1‒5 B.: “According to this interpretation we should understand by ‘substance’ the kind generated and corruptible and the one eternal, and that his purpose in this book [La¯m/ Lambda] which he primarily has in mind is to discuss the principles of the eternal substance, because he has shown the principles of the sensible substance generated and corruptible in Books Wa¯w and Za¯y” (reference to Books Z‒H as Wa¯w‒Za¯y within the context of a quota-

Who Wrote Alexander’s Commentary on Metaphysics Λ?

41

last sentence of paragraph n from which we started –, Book Θ as H  a¯ʾ (instead of Ta¯ʾ),111 and Book I as T a¯ʾ (instead of Ya¯ʾ).112 The presence of this unusual labelling system in the analysis of the Metaphysics expounded in our Part III does certainly not indicate that the analysis was composed by Ibn Rušd himself, as Freudenthal claimed. On the contrary: when compared with Ibn Rušd’s standard practice elsewhere and taken jointly with its presence in Ibn Rušd’s quotations of, or references to, “Alexander’s” comments on individual passages of Book Λ, the unusual system appears clearly confined to those cases where Ibn Rušd draws on “Alexander”.113 The unusual labelling system employed by Ibn Rušd in connection with “Alexander’s” commentary on Λ may be perceived as corresponding either (i) to the Aramaic alphabet as developed in the Syriac language, or (ii) to the numerical use made of the Aramaic ordering in the Arabic notation known as “abgˇad”, or else (iii) to the Greek alphabetic numerals. (i) In Syriac, the letters of the alphabet were used both as ordinary letters and as quasi-decimal alphabetic numerals. (ii) It is exclusively this latter, i.e. numerical, use of the “Aramaic” arrangement of letters that has survived in Arabic (= “abgˇad”). (iii) The numerical use of the Aramaic alphabet also corresponds to the alphabetic numerals of the Greeks which we have described earlier. The fact that in the Greek system the old F (Fαῦ) = 6 was restored resulted in phonetic similarity between the Greek and the Syriac system, as far as the first 14 signs go: in this respect, the Hellenistic alphabetic numerals might be characterized as a partial return to the Phoenician origins of the Greek alphabet.114 Consequently, for the first 14 signs – which is all we can possibly be concerned with as regards the 14 books of the Metaphysics –, the alphabetic numerals proper to Greek and Arabic, and the letters/numerals of the Syriac

111

112

113

114

tion from Alexander, cf. qa¯la l-Iskandar 1406,5). We are faced here, in the words of Martin (see note 32, 46 n. 6), with an “emprunt implicite à Alexandre.” 1439,9‒1440,1 B.: “Alexander says that this is something that [Aristotle] has shown in Book H  a¯ʾ, namely where he discussed potentiality and actuality” (reference to Book Θ as H  a¯ʾ in a quotation from Alexander; cf. Bouyges (see note 4), LXIIn7 and CLIV, 2°). 1431,3‒6 B.: “How change taking place into and from [the intermediates] is also from contraries on account of the fact that the things falling between two contraries result from a mixture of the two contraries is something that [Aristotle] has shown in Book T a¯ʾ” (reference to Book I as Ta¯ʾ within an elucidation of Alexander, cf. yaqu¯lu l-Iskandar fı¯ tafsı¯rihi 1430,5). This is also the opinion of Bouyges (see note 4), XIX‒XX, in his Remarques sur la designation des Livres ou Maqa¯la¯t, where the unusual references to books of the Metaphysics are presented as a perturbation occurring in the Prologue (Part III) “où les Sommaires de Livres dépendent du Tafsı¯r d’Alexandre” (ibid., XXn1). On the place and date of the adoption of the Phoenician script (φοινικήια γράμματα) by Greeks see Jeffery (note 97), 5–21 and Heubeck (see note 97), X 73–87.

42

Matteo Di Giovanni and Oliver Primavesi

alphabet, are very similar not only with regard to their quasi-decimal numerical use, but also phonetically: Numerical value

Greek numerals

Syriac letters

Arabic numerals

1 2 3 4 5 6 7 8 9 10 20 30 40 50

Α Β Γ Δ Ε F (ϛ) Ζ Η Θ Ι Κ Λ Μ Ν

      



 

‫ا‬ ‫ب‬ ‫ج‬ ‫د‬ ‫ه‬ ‫و‬ ‫ز‬ ‫ح‬ ‫ط‬ ‫ي‬ ‫ك‬ ‫ل‬ ‫م‬ ‫ن‬

(Alaph) (Beth) (Gamal) (Dalath) (He) (Waw) (Zayn)  eth) (H (Teth) (Yod) (Kaph) (Lamad) (Mem) (Nun)

(Alif) (Ba¯ʾ) ˇ ı¯m) (G (Da¯l) (Ha¯ʾ) (Wa¯w) (Za¯y)  a¯ʾ) (H (Ta¯ʾ) (Ya¯ʾ) (Ka¯f) (La¯m) (Mı¯m) (Nu¯n)

In light of these correspondences the four problematic Arabic book labels in question, i.e. Wa¯w for our Book Ζ, Za¯y for Book Η, H  a¯ʾ for Book Θ, and Ta¯ʾ for Book I, can be easily decoded. For the Arabic numerals Wa¯w, Za¯y, H  a¯ʾ, and Ta¯ʾ correspond to the Greek numerals F, Ζ, Η and Θ, which is to say that they occupy exactly the same positions – the 6th, 7th, 8th, and 9th respectively –, which in the ordinary Greek alphabet are occupied by letters Ζ, Η, Θ, and Ι; and the latter are precisely the Peripatetic labels of the books in question. It follows that at some point in the transmission of “Alexander’s” commentary the Peripatetic letter labels must have been replaced either by Greek numerals, or by Arabic “abgˇad” numerals, or by the letters/numerals of the Syriac alphabet. According to Freudenthal we would be faced here with the “usual Arabic way of counting books”;115 Bouyges, instead, seems to think that the introduction of the system goes back to the Greek text of “Alexander”.116 Both hypotheses are implausible. In addition to contradicting the standard desig-

115 Cf. Freudenthal (see note 4), 123: “… die bei den Arabern übliche Zählung der Bücher …”. 116 Bouyges (see note 4), CLIV n. 2 refers his reader to the Greek alphabetic numerals in order to account for the presence of the peculiar labelling system in “Alexander’s” commentary. The same view is taken by Martin (see note 32, 39 n. 50): “D’où vient cette discordance? Il est manifeste qu’Averroès cite, en ces pages 1403–1404, le Tafsı¯r d’Alexandre […] Le F que les Grecs appellaient ἐπίσημον Fαῦ, est donc devenu, à travers les traductions sémitiques (syriaque puis arabe) wa¯w, avec la valeur 6”. Cf. also ibid., 56 n. 4.

Who Wrote Alexander’s Commentary on Metaphysics Λ?

43

nation of the books of the Metaphysics established in both Greek and Arabic, any use of alphabetic numerals – no matter whether Greek or Arabic, no matter whether introduced by “Alexander” himself or by a translator – must be excluded specifically in the present context, since on p. 1395,4 B. and on p. 1397,5 B., Book Alif Major is identified explicitly as the “second book” and Book Alif Minor as the “first book” of the Metaphysics, as we have seen. Hence, Book Β would have to be counted as the third item of the list, not as the second, with the result that using alphabetic numerals would enˇ ı¯m, not as Ba¯ʾ, and so forth. Yet Book force the designation of Book Β as G Β, the book devoted to the aporiae, clearly is labelled Ba¯ʾ in paragraph i, and Books Γ ‒ Δ ‒ Ε follow suit. If, then, the new book labels could not have been introduced into our analysis on account of their numerical value, we must look for a context in which they could have been used for labelling, instead of counting, the items of a list in accordance with the ordinary alphabet. This requirement is met neither by the Arabic “abgˇad” notation, nor by the corresponding Greek system, since the use of both is strictly numerical, as we have seen. The only plausible occasion for the introduction of the new labels appears to be the translation of an original Greek version of “Alexander’s” commentary into Syriac, where the order of the alphabetic numerals and the ordinary alphabet are identical. The actual existence of a Syriac intermediary for the Arabic translation of “Alexander’s” commentary is documented by the valuable information that is provided by the notice on Aristotle’s Metaphysics in the 10th century Arabic bibliographical catalogue (Fihrist) compiled by Ibn al-Nadı¯m (d. AD 995):117 “Book of Letters (Kita¯b al-H ø uru¯f), known as Divine Matters (al-Ila¯hiyya¯t): The arrangement of this book follows that of the Greek letters (hø uru¯f alYu¯na¯niyyı¯n). It begins with Alif Minor (al-alif al-søug˙ra¯), translated by Ish a¯q [b. H  unayn], and what can be found of it goes up to letter Mu¯, translated by Zakarı¯ya¯ʾ Yah ya¯ b. ʿAdı¯. [Also] letter Nu¯ can be found: in Greek together with Alexander’s commentary. These letters were trans-

117 For the text of Ibn al-Nadı¯m’s notice, see Flügel/Roediger (see note 39), 251,25–252,1, reproduced in Bouyges (see note 4), CXVII. English translations are provided by Peters, F. E. 1968: Aristoteles Arabus. The Oriental Translations and Commentaries of the Aristotelian Corpus, Leiden, 49 and by Dodge, B. 1970: The Fihrist of al-Nadîm: A Tenth-Century Survey of Muslim Culture. 2 vols, New York–London, 606–607, a Castilian one by Ramón Guerrero (see note 36), 277. An analysis of the notice and its possible interpretations can be found in Bertolacci (see note 101), 244–249.

44

Matteo Di Giovanni and Oliver Primavesi

lated by Usta¯t for al-Kindı¯, who provided some information about that. A b u¯ B i š r M a t t a¯ t r a n s l a t e d B o o k L a¯ m , w h i c h i s t h e e l e v e n t h l e t t e r, t o g e t h e r w i t h A l e x a n d e r ’s c o m m e n t a r y ( m a q a¯ l a t a l - L a¯ m b i - t a f s ı¯ r a l - I s k a n d a r ) i n t o A r a b i c . H  unayn b. Ish a¯q translated this book (ha¯dß ihi l-maqa¯la) into Syriac. Themistius commented on Book La¯m, which was translated by Abu¯ Bišr Matta¯ together with Themistius’s commentary. It was translated by Šamlı¯ [too].  unayn translated several books. Syrianus commented on Book Ish a¯q b. H Ba¯ʾ, and this was rendered into Arabic (I saw it written, in Yah ya¯ b. ʿAdı¯’s own handwriting, in the catalogue of his books).” Ibn al-Nadı¯m informs us about an Arabic translation of Book Λ accompanied by Alexander’s commentary by Abu¯ Bišr Matta¯ b. Yu¯nus (d. AD 940),118 who is known to have translated from Syriac, not from Greek.119 The existence of a Syriac intermediary, a translation of Λ plus “Alexander”, is clearly presupposed here, irrespective of the further question as to whether this Syriac translation is identical with the translation of “this book” into Syriac  unayn b. Ish a¯q (AD 808‒873). We will which Ibn al-Nadı¯m attributes to H leave open this latter question, since the indeterminate expression “this book” (ha¯dß ihi l-maqa¯la) can, within the context, refer either to Book Λ of Aristotle120 or to the unit mentioned one line above (m a q a¯ l a t a l - L a¯ m b i - t a f s ı¯ r a l - I s k a n d a r) with Book Λ and “Alexander”.121 Leaving aside 118 Cf. Bouyges (see note 4), CXXII: “D’après le Fihrist (l. 28), le Livre LAM avec commentaire d’Alexandre a été traduit en arabe par Abou¯ Bišr; et aussi (l. 30) le Livre LAM avec commentaire de Thémistius”; more hesitantly Genequand (see note 32), 5–6: “Another point about which it will be prudent to keep an open mind is that of the authorship of the translation of Alexander’s commentary. It is, on the whole, very clear and readable, and I therefore hesitate to ascribe it to Abu¯ Bishr Matta¯.” 119 On Matta¯ b. Yu¯nus, who was “einer der bedeutendsten Initiatoren für die Rezeption der peripatetischen Philosophie durch Übersetzungen aus dem Syrischen in ihrer letzten Periode”, see Endress (see note 102), 295–301. 120 This interpretation is implied in the rendering of the Fihrist passage in Steinschneider, M. 1893: Die arabischen Uebersetzungen aus dem Griechischen. Zwölftes Beiheft zum Centralblatt für Bibliothekswesen, Leipzig, 67 = 195 (§ 35 = § 59): “Matta hat den Tractat Lamda, welcher der elfte Buchstabe ist, mit dem Commentar des Alexander ins Arabische übersetzt; dieser Tractat wurde auch von Honein syrisch übersetzt”. 121 This latter interpretation is accepted by Gutas, D. 2012b: Gelehrte als Vermittler philosophischen Denkens, in: U. Rudolph (ed.), Philosophie in der Islamischen Welt. Band 1: 8.– 10. Jahrhundert, Basel, 490 (No. 42). It is already presupposed in what Brockelmann, C. 1898: Geschichte der Arabischen Litteratur. I. Band, Weimar, 206 (ch. 11/5) presents as an  unayn; Brockelmann is innocent summary of Steinschneider’s rendering of the phrase on H inaccurate on three counts, however, for in ascribing to H  unayn “dess[elben, i.e. Aristoteles’] Metaphysik und Alexanders Comt.”, he omits the reference to Book Λ, adds the mention of Alexander’s commentary, and fails to make clear that we are dealing with a Syriac translation. This has earned him the criticism of Bouyges (see note 4), CXXIV: “Le célèbre

Who Wrote Alexander’s Commentary on Metaphysics Λ?

45

this issue, not relevant to our purposes, we shall remain content to regard as established the fact that there was a Graeco-Syriac translation of “Alexander”, calling the translator from Greek into Syriac “first translator”. In conjunction with the above analysis, this result suggests the following hypothesis: the first translator, when faced with the Peripatetic letter labels, did not attempt to represent these by what are – as far as Greek consonants go – their more or less close phonetic equivalents in Syriac, let alone to replace them by alphabetic numerals, which was excluded by the problem of the two Alphas. Instead he “translated” them, as it were, by replacing each occurrence of a given Greek letter by the letter of corresponding position in the Syriac alphabet: the first letter of the Greek alphabet by the first letter of the Syriac alphabet, the second by the second, and so forth. Only on this assumption does the labelling of Books Ζ ‒ Η ‒ Θ ‒ Ι as Wa¯w ‒ Za¯y ‒ H  a¯ʾ ‒ Ta¯ʾ in the Arabic “Alexander” not conflict with his counting Alpha Minor and Alpha Major as “first book” and “second book”, respectively: Peripatetic labels

/ “translated” into Syriac

/ rendered in Arabic

α ἔλαττον Α μεῖζον Β Γ Δ Ε Ζ Η Θ Ι Κ

       



‫ا‬ ‫ا‬ ‫ب‬ ‫ج‬ ‫د‬ ‫ه‬ ‫و‬ ‫ز‬ ‫ح‬ ‫ط‬ ‫ي‬

(Alaph) Minor (Alaph) Major (Beth) (Gamal) (Dalath) (He) (Waw) (Zayn) (H  eth) (Teth) (Yod)

(Alif) Minor (Alif) Major (Ba¯ʾ) ˇ ı¯m) (G (Da¯l) (Ha¯ʾ) (Wa¯w) (Za¯y) (H  a¯ʾ) (T a¯ʾ) (Ya¯ʾ)

In order to account for the presence of these “Syriac” book-labels in Part III of the proem and other passages connected with “Alexander”, it will certainly not do to assume, with Ramón Guerrero 1996, that they are isolated survivals of a passage in Alexander’s original commentary on Book Λ, and that Ibn Rušd has otherwise replaced the passage in question by an introductory section of his own which is “put into the mouth of Alexander” while

H  ounayn, dont le nom revient souvent quand on étudie le passage des œuvres grecques en arabe, est nommé à propos de la Métaphysique dans le Fihrist, l. 29, mais comme traducteur du onzième Livre en langue syriaque. C. Brockelmann dit qu’il aurait traduit en arabe « Metaphysik mit Alexanders Cmt »; mais il ne s’appuie sur aucune autorité nouvelle, et ce qu’avait écrit Mor. Steinschneider ne saurait étayer une telle affirmation. Tenons-nous en au renseignement du Fihrist.”

46

Matteo Di Giovanni and Oliver Primavesi

containing Ibn Rušd’s “personal observations”:122 had Ibn Rušd taken such liberties with “Alexander’s” text, why should he have faithfully preserved only the confusing “Syriac” book-labels rather than replacing them with his own (“Peripatetic”) labels? The “Syriac” labels rather indicate that the whole book-by-book analysis expounded by Ibn Rušd in Part III b) as well as the quotations from “Alexander” inserted into the main part of his commentary go back to a source which had been transmitted through the intermediary of a Syriac translation. This source was a codex containing a text of, and a commentary on, Book Λ; it had been translated from Greek into Syriac by the first translator, and from Syriac into Arabic by Matta¯ b. Yu¯nus. § o Book Theta (H ø a¯ʾ): 1403,11‒15 B. “Once he has made clear the principles of the sensible substance generated and corruptible in these two books, he considers that what is incumbent upon him to begin with, after this, is to examine the common attributes of being qua being. He first examines potentiality and actuality, reducing them to their principles, and he shows that actuality is prior to potentiality, which he does in the book whose title is the letter H  a¯ʾ placed after the letter Za¯y.” § p Book Iota (Tøa¯ʾ): 1403,16‒18 B. “Subsequently in the book that follows this, namely that whose title is the letter Ta¯ʾ, he examines the one and the many, the identical, the similar, the contrary, and the other common attributes that belong to being qua being.” § q Book Kappa (Ya¯ʾ): 1404,1‒8 B. “After this, in Book Ya¯ʾ, he discusses movement and the infinite, given that the expert in this discipline inquires into these from a different perspective than the expert in physics. Since [Aristotle] is determined after this book to discuss the principles of the first sensible substance, which is what he aims at primarily in this discipline (for it is the goal aimed at in it), at the beginning of this book, namely that whose title is the letter Ya¯ʾ, he recalls all the questions that he mentioned in the book of the letter Ba¯ʾ and sketches out the solutions given in the previous books.

122 Cf. Ramón Guerrero (see note 36), 285: “El origen de este Proemio puede estar en el comentario de Alejandro de Afrodisia al libro Λ de Metafísica, como parece deducirse de una atenta lectura del texto, sobre todo por la designación por medio de letras que se hace de los libros de la Metafísica por no corresponder la enumeración que aquí se hace con la que establece Averroes en otros pasajes y en su mismo comentario. No obstante, Averroes sólo pone en boca de Alejandro un parágrafo al comienzo y luego él mismo hace observaciones personales.”

Who Wrote Alexander’s Commentary on Metaphysics Λ?

47

Likewise he sketches out the absurdities that follow from the denial of the principles of knowledge.”  a¯ʾ – actuality In paragraph o, the subject matter associated with Book H (ἐντελέχεια) and potentiality (δύναμις) – clearly shows that what is alluded to is Book Θ (cf. Θ 1, 1045b34‒35: διορίσωμεν καὶ περὶ δυνάμεως καὶ ἐντελεa¯ʾ – the one χείας). Similarly, in paragraph p, the subject matter of Book T and the many, the identical, the similar, the contrary – makes clear that the reference is to Book Ι; cf. Ι 3, 1054a29‒32: ἔστι δὲ τοῦ μὲν ἑνὸς ὥσπερ καὶ ἐν τῇ διαιρέσει τῶν ἐναντίων διεγράψαμεν, τὸ αὐτὸ καὶ ὅμοιον καὶ ἴσον, τοῦ δὲ πλήθους τὸ ἕτερον καὶ ἀνόμοιον καὶ ἄνισον. Finally, in paragraph q, the peculiar combination of topics both from the Physics (movement and the infinite) and from the Metaphysics (the aporiai from Metaph. Β, the law of noncontradiction from Metaph. Γ) which is connected with Book Ya¯ʾ indicates that the book in question is our Book Κ. On this basis we conclude that in the three paragraphs o, p, and q the first translator must have rendered the Peripatetic labels Θ–Ι–Κ by means of the Syriac letters (H  eth), (Teth), and  (Yod), which Matta¯ b. Yu¯nus in  a¯ʾ, Ta¯ʾ, and Ya¯ʾ. Clearly, up to this point the first turn transposed into H translator has been consistently following the principle first apparent in paragraph n, i.e. the principle of rendering “Alexander’s” letter labels by “translating” them into the Syriac letters occupying the corresponding position in the Syriac alphabet. § r No Book Ka¯f found: 1404,9‒11 B. “This is what we find concerning the sequence of the books that have come down to us prior to Book La¯m: we do not find Book Ka¯f as expected according to the sequence of letters, nor has it come down to us (faha¯dß a¯ huwa lladß ı¯ nagˇiduhu fı¯ tartı¯b al-maqa¯la¯t allatı¯ waqaʿat ilayna¯ llatı¯ hiya qabla maqa¯lat al-La¯m wa-lasna¯ nagˇidu bi-hø asab tartı¯b al-hø uru¯f maqa¯lat al-Ka¯f wa-la¯ waqaʿat ilayna¯).” § s Book Lambda (La¯m): 1404,12‒16 B. “As far as Book La¯m is concerned, namely that which we begin to comment upon (šaraʿna¯ fı¯ tafsı¯riha¯), his primary purpose in it is to make known the principles123 of the first sensible substance. At the beginning, 123 There is no consistent use, in relevant passages of the Tafsı¯r, of the singular/plural number for referring to what the eternal sensible substance (the heavens) depends on, and the Arabic manuscript tradition fluctuates greatly: sometimes this is called “principle” (in the singular, cf. e.g. 780,15 ff.), sometimes “principles” (in the plural, cf. e.g. 1394,10; 1425,11). The thought behind the inconsistency seems to be that the principles of the eternal sensible substance are in fact reducible, upon examination, to one which is God (this line of interpretation is suggested by 1425,10–13). With this understanding in mind, we translate the Arabic text as printed by Bouyges.

48

Matteo Di Giovanni and Oliver Primavesi

however, he makes known the principles of all substances without qualification, starting from the principles of the substance generated and corruptible, and he recalls what has become clear on the subject in Books Wa¯w and Za¯y. Subsequently he shows the principles of the first substance, that this124 is substance, and in how many senses it is a principle.” The problems posed by paragraphs r and s are best addressed jointly. (i) A first problem is posed, in paragraph r, by the statement that there is an alphabet based on which one would expect some Book “Ka¯f” to be summarized, a book which is, however, not extant. What book of the Metaphysics could this refer to? Who could have possibly held this expectation? (ii) A second problem arises in paragraph s, where “Book La¯m” is explicitly characterized as “that which we begin to comment upon.” The label “Book La¯m” clearly refers to our Book Λ, since this is the book which will in fact be commented upon,125 and thus the Arabic letter La¯m is used here as a phonetic equivalent of the Greek letter Λ, whereas the Arabic book-labels assigned to the preceding books appear to be delayed by one letter with respect to the phonetically equivalent Peripatetic booklabels. Why does the rendering of Λ differ from the rendering of the preceding Greek book labels? Let us begin by addressing problem (ii). A first observation to be made here is that both methods of labelling books of the Metaphysics must go back to the first translator: he is not only responsible for the fact that the Arabic  a¯ʾ, Ta¯ʾ, and Ya¯ʾ do not stand for Ζ, Η, Θ, and Ι, but for book labels Za¯y, H Η, Θ, Ι, and Κ; he is also responsible for the labelling of Book Λ by its phonetic equivalent. For right at the beginning of “Alexander’s” assessment of Book Λ (1394,4 B. = Fr. 1 Freudenthal), this book is introduced as “Book La¯m”; and in paragraph s we find a discussion of Book La¯m = Λ which includes a back-reference to Books Ζ–Η by means of the “Syriac” book labels (Wa¯w and Za¯y); if the labelling of Book Λ as Book La¯m occurs in one and the same paragraph as the “Syriac” book labels (Wa¯w and Za¯y), it does not seem plausible to ascribe the two notations to different authors.

124 “This” (ha¯dß a¯) must refer to the principles as reducible to God (see the previous note). 125 An analogous observation is to be made in paragraph r, where the sequence of books so far discussed is characterized as “the sequence of the books that have come down to us prior to Book La¯m”: this shows that “Book La¯m” itself is regarded as a point of reference, and the natural point of reference here would seem to be the very book which is going to be commented upon, i.e. our Book Λ.

Who Wrote Alexander’s Commentary on Metaphysics Λ?

49

The challenge, then, is to understand why this one Syriac translator treated the Aristotelian book-label Λ in one way and the labels used by “Alexander” for the preceding books in another. There are two conceivable solutions, neither of which is free from difficulties: on the first hypothesis (A) our Syriac translator lacks information, whereas on the second hypothesis (B) he lacks consistency. A) On the first hypothesis, the Syriac translator was not acquainted with other parts of the Metaphysics. On this assumption the book-label Λ displayed in the title would have been known to him as an isolated item, being difficult to interpret due to the lack of context. Even a native Greek speaker could have interpreted it in various ways: either as the twelfth letter of the alphabetical numerals and, accordingly, as designating the number 30 – this being its standard function as a book number in virtually all of Greek literature –, or as the eleventh letter of the ordinary alphabet – this being its peculiar function as a book label in Aristotle as well as Theophrastus and Homer.126 A Syriac scholar was exposed to the additional challenge of seeing through a fairly subtle technical difference between the Greek system of alphabetic numerals and his own, Syriac alphabet: when used for numbering purposes, the first fourteen letters of the Greek system and the Syriac one are parallel with regard to both phonetics and function, as we have seen; but when used for the non-numerical purpose of drawing up an alphabetically ordered list, the Greek omits the sixth letter, i.e. F, whereas the Syriac retains its sixth letter, i.e.  (Waw). It is quite possible that our Syriac translator was unaware of this last-mentioned subtlety. In this case he may have interpreted the Aristotelian book-label Λ as being the twelfth letter of Aristotle’s labelling alphabet, just on the model of the Syriac letter Lamad. As a result, an isolated Aristotelian book-label Λ could have been perceived in any one of these three ways: 1) as an alphabetic numeral indicating 30; 2) as the eleventh letter of a Greek alphabet deprived of F; or 3) as the twelfth letter of a Greek alphabet containing F (and yet used for non-numerical labelling). We have seen in what precedes that in the present context a use of alphabetic numerals is made unlikely by the fact that Book Alif Minor is explicitly identified as the “first book” and Book Alif Major as the “second book” of the Metaphysics, whereas the third item, Book Β, is nevertheless called Ba¯ʾ, ˇ ı¯m. We have also seen, however, that for all books preceding Λ our not G first translator, by replacing “Alexander’s” book-label Z by the label Waw, made a (non-numerical) use of an alphabet in which Lamad would have to count as the twelfth letter, and thus departed from “Alexander” who must 126 Cf. Primavesi (see note 96), 65–68.

50

Matteo Di Giovanni and Oliver Primavesi

have used what we know to be the Peripatetic book-labels: this is the standard Greek alphabet in which F is omitted and Λ is the eleventh letter. If the first translator’s re-labelling of the preceding books is to be at all consistent with his rendering of the Aristotelian book-label Λ by means of Lamad, we must assume that he ascribed to Aristotle a labelling alphabet in which Λ counts as the twelfth letter:127 “Alexander’s” “faulty” labels

Aristotle’s labels as reconstructed by first translator

First translator’s Syriac labels

Arabic rendering by Matta¯ b. Yu¯nus

α ἔλαττον Α μεῖζον Β Γ Δ Ε Ζ Η Θ Ι Κ –

α ἔλαττον Α μεῖζον Β Γ Δ Ε F Ζ Η Θ Ι (Κ, not paraphrased by “Alexander”) Λ Μ Ν

       

 (

‫ا‬ ‫ا‬ ‫ب‬ ‫ج‬ ‫د‬ ‫ه‬ ‫و‬ ‫ز‬ ‫ح‬ ‫ط‬ ‫ي‬ (‫ك‬

Alif Minor Alif Major Ba¯ʾ ˇ ı¯m G Da¯l Ha¯ʾ Wa¯w Za¯y H  a¯ʾ Ta¯ʾ Ya¯ʾ Ka¯f)

‫ل‬ ‫م‬ ‫ن‬

La¯m Mı¯m Nu¯n

Λ Μ Ν

Alaph Minor Alaph Major Beth Gamal Dalath He Waw Zayn H  eth Teth Yod Kaph)

Lamad  Mem  Nun

On this hypothesis our translator could indeed have labelled Book Λ as (Lamad) in a way consistent with his re-labelling of Books Ζ, Η, Θ, Ι, and Κ, since he regarded as errors two interrelated features of “Alexander’s” commentary: (i) the consistent use of a labelling system marked by the absence of F; and (ii) the absence of a summary of the book which according to the “correct” system should be called Κ. The difficulty of attributing both defects to his ancient source (“Alexander’s” analysis) might – and perhaps even should – have induced our translator to rethink his own attribution to Aristotle of a system in which Λ counts as the twelfth letter. But in fact, on solution A), the difficulty would not have come to the translator’s mind; rather, he would have been so convinced of his own reconstruction that he 127 This interpretation of the label Λ is shared by the two 16th century translators of Ibn Rušd’s proem from Hebrew into Latin, who still render “Alexander’s” reference to Λ by “Duodecimo” (m), or by “literam Lamech, scilicet duodecima litera alphabeti” (p). See Bouyges 1948a (see note 32), apparatus on 1394, 413, and, on the two Hebrew-Latin translations in general, Bouyges (see note 4), LXXIX–LXXX.

Who Wrote Alexander’s Commentary on Metaphysics Λ?

51

took the bold initiative to correct “Alexander’s” labels in line with what he himself assumed to be Aristotle’s “original” system and what was in fact in use, albeit for numerical purposes, in the book-titles of almost all other Greek texts. Even so, however, solution A) can only work on the assumption from which we started, namely on the assumption that the first translator was not acquainted with other parts of the Metaphysics. For if he had consulted the original Greek text of the Metaphysics more extensively, he would have certainly noted that this text confirmed both of the suspected features of “Alexander”, namely the omission of the label F and the absence of an additional book between our Book Κ and our Book Λ. Thus, put on guard by such evidence, he would have hardly ventured to correct “Alexander” in the way just described. This consideration would suggest placing the first translator in a time when not only the preceding books of the Metaphysics but even the Peripatetic labelling system as such was still unknown in the East.128 However, a translation of a work so complex and technical in nature as Book Λ with “Alexander’s” commentary is perhaps more likely to have been undertaken when the Graeco-Arabic translation movement was already more advanced.129 All in all, solution A) implies that the first translator employed the Syriac book-labels with impeccable consistency but displayed both a considerable lack of knowledge concerning the other books of the Metaphysics and the Peripatetic system of book-labels in general (which would suggest that he worked in a surprisingly early stage of the translatio studiorum),130 and an unlikely light-heartedness in attributing fairly elementary errors to Alexander of Aphrodisias. We will do well, therefore, to consider an alternative solution. B) On the second hypothesis, the first translator was perfectly aware of the fact that Λ is the eleventh letter of the Peripatetic labelling-alphabet, so that his decision to render Λ as Lamad was not based on the position of Lamad

128 In this case, the identification of the first translator with H  unayn b. Ish a¯q (AD 808‒873) becomes problematic. See Steinschneider (see note 120), 67 = 195 (§ 35 = 59) and the criticism levelled by Bouyges (see note 4), CXXIV at Brockelmann (see note 121), 206 (ch. 11/5), all already quoted. 129 For a general survey on methods, phases, and significance of the translation movement see Gutas, D. 2012a, Die Wiedergeburt der Philosophie und die Übersetzungen ins Arabische, in: U. Rudolph (ed.), Philosophie in der Islamischen Welt. Band 1: 8.–10. Jahrhundert, Basel, 62‒66. For a detailed periodization of Metaphysics translations in the Islamic world see Bertolacci (see note 101). 130 Ibn al-Nadı¯m (d. 995 AD) in the passage of his Fihrist translated above emphasizes that La¯m is the eleventh letter label used for Aristotle’s Metaphysics, as we have seen; on hypothesis A) this remark might be regarded as a correction of the system introduced by our first translator.

52

Matteo Di Giovanni and Oliver Primavesi

in the Syriac alphabet but on its phonetic correspondence with Λ. Nevertheless, as we have seen, he re-labelled the previous books in accordance with a system in which the Peripatetic labels Ζ, Η, Θ, Ι, and Κ are “translated” into Waw, Zayn, H  eth, T eth, and Yod, so that Aristotle’s eleventh letter (Λ) would have to be rendered not by Lamad but by Kaph. Hence the question arises: is it at all possible to explain this apparent oddity? Any attempt to answer in the affirmative must start from Genequand’s assumption (p. 8) that the rendering of the isolated label Λ by Lamad was already fixed in learned circles of the day, (presumably 9th century Baghdad). The easiest way of accounting for this state of affairs would be to envisage that the Syriac translation of Themistius’ short paraphrase of Book Λ (which is presupposed by the aforementioned testimony of the Fihrist) antedated that of the much more demanding “Alexander”. For the paraphrase of Themistius, unlike “Alexander’s” commentary, did not expose the translator to a series of preceding Aristotelian book labels with which an initial transliteration of Λ to Lamad could have appeared inconsistent. Under these circumstances, the subsequent translator of “Alexander” would have been well advised to follow the standard already in place lest his readers be misled by the coexistence of different labels for one and the same Aristotelian book. In order to see what could have led the first translator to re-label the previous books nevertheless in the particular way he did, we should put ourselves in his position when faced with the task of rendering into Syriac the labels used by “Alexander” in his book-by-book analysis. He could have employed the ordinary Syriac alphabet in a way that compensates for the surplus of Syriac letters (due to the letter Waw) by making use of the coexistence, in “Alexander’s analysis”, of two books labelled as Book A. Thus, he would have safely arrived at the equation Λ = Lamad: α ἔλαττον Α μεῖζον Β Γ Δ Ε Ζ

 Alaph  Beth  Gamal  Dalath  He  Waw  Zayn

Η Θ I Κ Λ Μ Ν





 

 eth H Teth Yod Kaph Lamad Mem Nun

This rendering, however, would have had the unpleasant result of concealing what is the most peculiar feature of the Greek labelling that is attested by Alexander: the fact that the first two books, the two Alphas, share the same book-label. Moreover, supposing again that the Syriac translation dates from the 9th century, the rendering in question would have obscured in addition the striking fact that Alexander’s analysis diverges from the edition of the

Who Wrote Alexander’s Commentary on Metaphysics Λ?

53

Metaphysics which was known in the East since Usta¯t: whereas Usta¯t presumably included only one book labelled as Book A (Alpha Minor),131 in “Alexander’s” analysis there are two of them (Alpha Minor and Alpha Major). If the first translator wanted to bring out this contrast, one option might have been to make use of Syriac phonetic equivalents, to the extent possible, in a way similar to what Usta¯t did in Arabic: α ἔλαττον Α μεῖζον Β Γ Δ Ε Ζ

 Alaph Minor  Alaph Major  Beth  Gamal  Dalath  He  Zayn

Η Θ I Κ Λ Μ Ν





 

 eth H Teth Yod Kaph Lamad Mem Nun

In hindsight, i.e. with Usta¯t’s rendering in mind, this may seem to be a natural way of proceeding, but it need not have been so for the translator who was to render into Syriac the labels of the books preceding Book Λ. For the omission of Waw in a Syriac rendering of the Greek series Α‒Β‒Γ‒Δ‒Ε‒Ζ‒Η‒ Θ‒Ι‒Κ is definitely more disturbing than the omission of Wa¯w in an Arabic rendering of the same Greek series. In Syriac, the result would correspond to a sequence of the first eleven letters of the alphabet in which only one letter (Waw) seems to have been incompetently omitted, whereas in Arabic the result would differ so widely from the standard Arabic alphabet that one would allow for a specific underlying rationale. All things considered, then, the Syriac first translator may well have deemed it a lesser evil to keep the established equation Λ = Lamad, while rendering the preceding series of Greek book-labels in a way that both preserves the coexistence of two Alphas and follows the ordinary Syriac alphabet as close as possible, leaving out a letter (Kaph) only after the end of the series: α ἔλαττον Α μεῖζον Β Γ Δ Ε Ζ

 Alaph Minor  Alaph Major  Beth  Gamal  Dalath  He  Waw

Η Θ I Κ Λ Μ Ν





 

Zayn H  eth Teth Yod Lamad Mem Nun

131 On the absence of Alpha Major from Usta¯t’s translation see the analysis in Bertolacci (see note 101), 246.

54

Matteo Di Giovanni and Oliver Primavesi

With regard to our overall argument, it is not necessary to commit ourselves either to hypothesis A) or to hypothesis B), since on both hypotheses the book-labels in question are firmly embedded in a Syriac context. From a historical point of view, however, the price to be paid for the assumption of complete consistency on the part of the first translator (hypothesis A) seems so high that the present hypothesis (B) is perhaps to be preferred: the first translator would seem to have treated Book Λ and the series of the preceding Books Alpha Minor – Kappa as two separate entities, the latter of which he has re-labelled in accordance with the Syriac alphabet; consequently, he made no use of the one Syriac letter (Kaph) which falls between the two. This result takes us back to our problem (i), which is the problem of explaining the note according to which the book-labels used in “Alexander’s” book-by-book analysis suggest postulating an additional book, namely Book Ka¯f, which “has not come down to us.” To begin with, this remark can on no account go back to the first translator. On hypothesis A), the first translator would have attributed to “Alexander’s” book-by-book analysis the two shortcomings of omitting the booklabel F, and of not summarizing the “true” Book Κ. This presupposes that the translator was ignorant of the text of the other books of the Metaphysics, or else he would have observed that in this text the book-label F is not used at all and that there is no book possibly “missing” between the two that are labelled as Book Yod (= our Book Κ) and as Book Lamad (= our Book Λ). In the statement with which we are now concerned, however, the diagnosis that Book Ka¯f is missing is not merely based on the fact that in “Alexander’s” summary there is no book between Book Yod/Ya¯ʾ (= our Book Κ) and Book Lamad/La¯m (= our Book Λ); in effect, it is explicitly stated that no such book “has come down to us”. This clearly presupposes an inspection of a more or less complete text of the Metaphysics, which was ruled out on hypothesis A). On hypothesis B), by contrast, the first translator would have replaced “Alexander’s” book-label Κ by Yod and left out the book-label Kaph on purpose, in order to compensate for the incompatibility between labelling Book Λ as Lamad and labelling the preceding books in accordance with the order of the complete Syriac alphabet. If so, it seems fairly evident that the same translator cannot have regarded his own device of omitting the booklabel Kaph as evidence for the actual loss of one book of the Metaphysics. At the same time the statement on the missing book is not likely to be by Ibn Rušd either. It is true that it was attributed to Ibn Rušd in previous scholarship,132 thus contributing to the view that Part III is his own elabora132 This attribution is attested for the first time in a gloss found in the Leiden manuscript between Book I and Λ (fol. 146r) where the absence of Book K is noted as follows: “There would follow Book 10 – whose title is the letter Ka¯f – but Abu¯ l-Walı¯d [Ibn Rušd] did not

Who Wrote Alexander’s Commentary on Metaphysics Λ?

55

tion. But in fact, Ibn Rušd cannot have described our Book Κ as not extant, since (while omitting it in the Tafsı¯r) he did paraphrase it in his Middle Commentary (Talh˚ ı¯sø) which is extant only in Hebrew.133 Thus, he would have noticed that Book Κ was summarized in paragraph q, albeit in the guise of “Book Ya¯ʾ”.134 Neither had he evidence to postulate a further Book “Ka¯f”, other than our K, between Ya¯ʾ and La¯m, as maintained by Freudenthal. Freudenthal appears to have constructed this hypothesis when he was still working on the basis of the Hebrew translation alone. For in Hebrew, the standard alphabet motivates the expectation of a sequence Yod-Kaph-Lamed,135 comment on it. He said that he did not get hold of it. I did, however, and inserted it here. God permitting I will comment on it, as well as Books 12 and 13, in accord with Abu¯ l-Walı¯d’s mind” (our italics; for the text of the gloss see Bouyges (see note 4), LXII under ‘Annotation 25’). Books 10, 12, and 13 correspond to K, M, and (apparently) N in Usta¯t’s translation due to the absence of Alpha Major, and are all absent from the Tafsı¯r. 133 Freudenthal (see note 4), 128; cf. more recently Zonta, M. 2011: Il Commento medio di Averroè alla Metafisica di Aristotele nella tradizione ebraica. Edizione delle versioni ebraiche medievali di Zerah yah Hen e di Qalonymos ben Qalonymos con introduzione storica e filologica, 2 vols, Pavia, 12. This information invalidates the remarks of Genequand (see note 32, 8–9): “Ibn Rushd interspersed Alexander’s text with remarks of his own. Such a one is the reference to the absence of book Ka¯f. This clearly was not written by Alexander whose text of the Metaphysics cannot have been very different from ours […] I think it more likely that K was never translated or any rate did not figure in any of the versions used by Ibn Rushd”. Steinschneider (see note 120), 15, by contrast, acknowledged that Book K was included in the Talh˚ ı¯sø and so he attempted to explain the problematic evidence with the hypothesis that the book may have become available to Ibn Rušd after the completion of the Tafsı¯r and was inserted in the former (but not in the latter) at a later time. This scenario is not impossible but, as admitted by Steinschneider himself, the hypothesis should be supported by some evidence in order to be accepted (“Auch diese Hypothese müsste von irgend einer Seite her unterstützt sein, um angenommen zu werden”). 134 As acknowledged by Martin (see note 32), 41 n. 65, who nevertheless defends the ascription to Ibn Rušd: “Cette allusion à l’absence de K semble prouver qu’Averroès, qui parle maintenant en son nom, n’a pas compris que sous la letter YA¯, Alexandre parlait bel et bien de K […] Il est surprenant qu’un grand connoisseur de la Métaphysique se soit laissé abuser par la dénomination des livres, malgré une analyse qui ne laisse aucun doute sur l’identité des livres cités.” 135 Freudenthal (see note 4), 129: “Die arabischen Buchstaben aber entsprechen den griechischen nicht. So ward, wie aus dem arabischen Texte des Averroes ersichtlich ist, Buch Ι mit ‫ = ط‬Θ; Κ mit ‫ = ي‬Ι bezeichnet, während für Λ die feststehende richtige Bezeichnung ‫ = ل‬Λ beibehalten wurde. Demnach musste ein Jeder, der das griechische Original nicht kannte, musste auch Averroes ein zwischen ‫ ي‬und ‫ ل‬liegendes Buch, das dem arabischen ‫ =( ك‬K) entsprach, vermissen”. Cf., along similar lines, Bertolacci (see note 101), 250 n. 22: “In the introduction to his commentary on Λ, Averroes provides an accurate description of book K, which he designates by means of the Arabic letter Ya¯ʾ (Tafsı¯r, p. 1404, 1–8). Immediately afterwards (p. 1404, 9–11), before the description of book Λ (La¯m), he states: ‘This is what we find concerning the order of the books which have come down to us and which come before La¯m, but we do not find book Ka¯f in the order of letters, nor has it come down to us’ (Engl. transl. in Genequand, Ibn Rushd’s Metaphysics, p. 64). This statement, isolated

56

Matteo Di Giovanni and Oliver Primavesi

whereas in Arabic, Ya¯ʾ is not immediately followed by Ka¯f either in the ordinary Arabic alphabet or in the “abgˇad”-system of alphabetic numerals (Ya¯ʾ meaning 10, Ka¯f 20), unless the latter are taken in their purely sequential value to enumerate items, which in our case is again excluded by the labelling of the two Alphas. The probable author of the statement about a missing Book Ka¯f appears in fact to be the second translator, Matta¯ b. Yu¯nus. A translator from Syriac into Arabic who had no access to the Greek original is quite likely to have been misled by the first translator’s unconventional use of book-labels: the lacuna that was found between the Syriac letters Yod and Lamad due to the missing Kaph may well have induced Matta¯ to believe that the Metaphysics described by “Alexander” presented a corresponding lacuna between our Books Κ and Λ.136 In any event both the (non-numerical) re-labelling of Books Ζ, Η, Θ, Ι, and Κ and, presumably, the statement that Book Ka¯f is missing go back to a Syriac milieu, since both features are at odds with the two options also available in a Greek or Arabic context, i.e. with the authentic Peripatetic booklabels and with alphabetic “abgˇad”-numerals. The terminus ante quem for the composition of the analysis of the treatise as a whole, as contained in Part III of the proem, is the translation from Greek into Syriac of a codex which contained the Greek text of Book Λ and “Alexander’s” commentary on it. § t Books My (Mı¯m) and Ny (Nu¯n): 1405,1‒3 B. “Once he finishes considering the principle of the universe, he returns to refuting what was maintained by his predecessors concerning the principles of substance. This is done in the two books whose titles are the letter Mı¯m and the letter Nu¯n.” The fact that in paragraph t our Books Μ and Ν are labelled Mı¯m and Nu¯n does not make an independent contribution to the evidence bearing on our problem since it is a mere consequence of Book Λ having been labelled La¯m. § u The Perfect Order of the Metaphysics (Established): 1405,4‒8 B. “It is finally clear from the foregoing what each of the books of this discipline that are ascribed to Aristotle contain, that their arrangement from the context, has been taken as indicating that Averroes did not know book K at all (see Notice, p. cli). On the contrary, it only attests that he did not know this book as book Ka¯f, but as book Ya¯ʾ.” 136 On Matta¯’s activities as a glossator in general see Endress (see note 119), 297: “Die Kommentare Matta¯’s und seiner Schule waren – so weit uns erhalten – in der Form mehr oder weniger umfangreicher Anmerkungen gehalten, sei es als Marginalien oder an Lemmata des Textes angehängt (taʿa¯lı¯q).”

Who Wrote Alexander’s Commentary on Metaphysics Λ?

57

follows the best order, and that in them nothing is encountered that is not arranged and ordered, contrary to what we find Nicolaus of Damascus claiming in his work, where he states that he, for this reason, chose a better arrangement for didactic purposes, according to what he claims.” In paragraph u “Alexander” notes that he has by now succeeded in proving the claim made as early as paragraph f (1397,3‒4 B.): Aristotle has indeed chosen the best arrangement for the purposes of instruction. It is now added that this result amounts to a refutation of the view of Nicolaus of Damascus, who claimed to have improved, in his own exposition of the Metaphysics, upon Aristotle’s arrangement.137 This reference to Nicolaus of Damascus was thought by Freudenthal to provide further evidence for the assumption that the analysis contained in Part III was composed by Ibn Rušd.138 This argument, however, is again inconclusive. It is true that Ibn Rušd mentions and quotes Nicolaus’ Περὶ τῆς Ἀριστοτέλους φιλοσοφίας, Books II–III (Θεωρία τῶν μετὰ τὰ φυσικά) 139 fairly often in other books of his Tafsı¯r,140 and in two of these passages Ibn Rušd also aims at the same goal as here, i.e. refuting Nicolaus’ claim to have introduced a better arrangement of topics than that produced by Aristotle himself.141 It is also true that in Alexander’s commentaries extant in Greek Nicolaus is nowhere mentioned.142 On these grounds Ibn Rušd may well have added the reference to Nicolaus in the present passage. But such an addition toward the end of Part III could certainly not prove anything about the authorship of Part III as a whole.

137 Cf. Moraux, P. 1973: Der Aristotelismus bei den Griechen von Andronikos bis Alexander von Aphrodisias: Erster Band: Die Renaissance des Aristotelismus im 1. Jh. v. Chr., Berlin– New York, 468–469. 138 Cf. Freudenthal (see note 4), 123: “Dass aber das Nachfolgende die Erörterungen des Averroes, nicht die des Alexander enthält, erweist … das Citat aus Nikolaus von Damaskus”. 139 On this work see Moraux (see note 137), 465–487, especially 473–475. 140 See the German translation of passages from the Tafsı¯r which are pertinent to Nicolaus in Freudenthal (see note 4), 126–127. An English translation of these passages has been provided by Drossaart Lulofs, H. J. (ed.) 1965: Nicolaus Damascenus on the Philosophy of Aristotle. Fragments of the first five books translated from the Syriac with an introduction and commentary, Leiden, 11–12 (Testimona 7.2–6) and 76–79 (Fragment 22). 141 Freudenthal (see note 4), 126–127, fragments 1–2; Drossaart Lulofs (see note 140), 11–12, testimonia 7.4–5. 142 Drossaart Lulofs (see note 140), 6: “The indexes to the splendid edition of the Commentaria in Aristotelem graeca, published between 1882 and 1909 confirmed what was already known, viz. Nic. is nowhere else mentioned by any of them, except in Simpl.’s commentaries to the Physics and De caelo”. In Simplicius Cael. 398,36–399,4, however, there is a quotation from Nicolaus’ Περὶ τῆς Ἀριστοτέλους φιλοσοφίας, Book IV (fragment 29 Drossaart Lulofs), which is inserted between two references to Alexander’s Commentary on De caelo in a way which suggests that the quotation from Nicolaus might go back to Alexander.

58

Matteo Di Giovanni and Oliver Primavesi

§ x Praise of Alexander: 1405,9‒12 B. “Out of our love and intense desire to learn and teach this discipline we have given a concise account (lah˚ h˚ asøna¯) of what this man said about the discipline, so that for those who have not devoted themselves to inquiring into the books (maqa¯la¯t) of Aristotle, [the discipline] might be easier to access, whereas for those who have studied Aristotle’s books this [account] might serve as a reminder.” The phrase “what this man said about the discipline” in the first sentence of this concluding remark was taken by Freudenthal as a direct reference to what Aristotle said about the discipline.143 But the whole first sentence (“We have … given a concise account of what this man said”, Ar. fa-qad lah˚ h˚ asøna¯ l-qawl alladß ı¯ li-ha¯dß a¯ l-ragˇul) answers almost verbatim, by way of a recapitulation, to the program announced in 1393, 7–8: “I considered that the best was to give an account of what Alexander says (an nulah˚ h˚ isøa ma¯ yaqu¯luhu l-Iskandar) about each section of [this book] as clearly and concisely as I could.” On this ground we conclude that “this man” is not Aristotle but “Alexander”. At the same time, it is clear that what “Alexander” is being credited with must include the book-by-book analysis that is found in Part III: for, according to the present remark of Ibn Rušd, “Alexander’s” text was summarized in order to serve as a reminder of what is contained in the single books (maqa¯la¯t) of Aristotle’s Metaphysics, and the section of the proem that is devoted to an analysis of the Metaphysics which is detailed book by book is precisely its Part III.

III. Conclusion In 1885, Jacob Freudenthal attempted to show that the extant Greek commentary on Metaphysics Λ, ascribed by others to Michael of Ephesus, can by no means go back, however indirectly, to Alexander of Aphrodisias. Freudenthal’s argument was that the Greek commentary is not only different from, but incompatible with the quotations from Alexander inserted by Ibn Rušd in his Tafsı¯r on Book Λ, the source of these quotations being a special edition of Aristotle’s Book Λ, complemented by Alexander’s commentary, 143 Cf. Freudenthal (see note 4), 123: “Dass aber das Nachfolgende die Erörterungen des Averroes, nicht die des Alexander enthält, erweist … das Schlusswort des Averroes (oben S. 69)”. Freudenthal was followed in this by Martin: “nous avons analysé les théories d’Aristote”, and note ad loc. Cf. the undecided remark by Ramón Guerrero (see note 36, 294n106): “La expression “este hombre” puede designar a Alejandro, como sobreentiende Genequand, p. 65, o a Aristóteles, como lee Martin, p. 42.”

Who Wrote Alexander’s Commentary on Metaphysics Λ?

59

which had been translated from Syriac into Arabic by Abu¯ Bišr Matta¯ b. Yu¯nus. Freudenthal’s argument requires that the authenticity of this source is beyond any doubt. But in fact, there is a problem with authenticity. In Part II of his proem to Book Λ, Ibn Rušd summarizes a passage from “Alexander” which comes obviously from the same source as “Alexander’s” comments on individual passages of Book Λ which Ibn Rušd will quote later: according to this passage, the whole Metaphysics culminates in Book Λ. In Part III b), Ibn Rušd expounds a concise book-by-book-analysis of the Metaphysics designed to show that the treatise is a well-ordered whole.144 This analysis has been thought by more than one scholar – suffice it here to mention Valentin Rose and Maurice Bouyges – to come from the same work by Alexander as the passage concerning the status of Book Λ. And yet the analysis presupposes that Book Alpha Minor is the first book of the Metaphysics and Book Alpha Major the second, which is clearly neither the order known to, nor the possible alternative envisaged by, Alexander in his extant commentary on Books Α‒Δ. If this book-by-book-analysis really comes from the same source as Ibn Rušd’s quotations from “Alexander’s commentary”, the authenticity of the latter can no longer be taken for granted. Faced with that situation, Freudenthal aimed to defend the authenticity of Ibn Rušd’s source by claiming that this source must not be held responsible for the problematic book-by-book-analysis. So he advanced six arguments designed to prove that the analysis of the Metaphysics presented by Ibn Rušd in Part III b) of his proem to Book Λ was composed by Ibn Rušd himself, rather than being an exposition of a corresponding passage in Alexander’s commentary.145 Since Maurice Bouyges upheld the opposite thesis146 without taking notice of Freudenthal’s arguments,147 these arguments have never been properly examined so far, and the starting point of the present study was the pressing need to fill that gap. Our results may be summarized as follows. Freudenthal’s first argument, concerning the phrase ‘and we say’ in 1395,11 B. fails to show that the ensuing analysis (Part III a‒b) was composed by Ibn Rušd, since the phrase in question refers to Ibn Rušd’s activity as an expositor. According to Freudenthal’s second argument, the overall contents of the analysis expounded in Part III b) would also show that it was the work of Ibn Rušd. This argument is invalidated, in particular, by the section on book

144 145 146 147

1395,9‒1405,12 B. Freudenthal (see note 4), 123. Bouyges (see note 15), 280 with n. (6). See also Bouyges (see note 4), XX n. (1). Bouyges (see note 4), CLXXVIII, n. (7).

60

Matteo Di Giovanni and Oliver Primavesi

Alpha Major (1397,14‒1398,7 B.): here Aristotle’s interpretation of his predecessors’ treatment of the four causes (Α 3‒7) is correctly summarized – even in Alexander’s terminology148 –, despite the fact that the greater part of Aristotle’s interpretation (Α 3‒5) and especially Aristotle’s own summary in ch. Α 7 were inaccessible to Ibn Rušd. In his third argument Freudenthal pointed to the unusual letter labels which are used within the analysis in order to refer to the single books of the Metaphysics. These letter labels are indeed of fundamental importance for reconstructing the history of the analysis. But Freudenthal was wrong to identify the use made of them here with the “usual Arabic method of counting books.” On the contrary: despite the fact that Alpha Minor and Alpha Major have explicitly been called “the first book” and “the second book” respectiveˇ ı¯m, so that the alphabet ly, the third book is nevertheless called Ba¯ʾ, not G cannot be used for counting purposes here. On the other hand, the letter Wa¯w, Arabic equivalent of the old Greek letter F, appears as the sixth letter of the alphabet in question. An alphabet used for purposes other than counting and where the phonetic equivalent of the letter F is used as the sixth letter can neither be identified with the ordinary Greek or Arabic alphabet nor the Greek or Arabic alphabetic numerals, but only with the Syriac alphabet. Therefore, the book-labels in question must have been introduced by the scholar who translated the Greek original into Syriac. Also the striking gloss on p. 1404, 9‒11 B., according to which we should expect a Book Ka¯f which “has not come down to us,” can neither go back to the first translator, as we have seen, nor is it coherent with the use of the Arabic standard alphabet or the system of “abgˇad”-numerals or the phonetic rendering of the Peripatetic book labels as introduced by Usta¯t. Accordingly, it can only go back to a later Syriac scholar who had no independent access to the Greek original, in all probability Abu¯ Bišr Matta¯ b. Yu¯nus. Thus, the book-by-book analysis shows clear marks of the translation process throughout, which rules out the possibility that it might have been composed by Ibn Rušd himself, as claimed since Freudenthal. This conclusion has received further corroboration from the fact that the “Syriac” book labels in question, while not corresponding to Ibn Rušd’s usual practice, do reappear in connection with “Alexander’s” comments on individual passages of Book Λ. Thus, they prove that the analysis forms part and parcel of the source from which Ibn Rušd quotes “Alexander” in later passages of his commentary. This source is a commentary on Book Λ which is ascribed to Alexander of Aphrodisias: it was, according to Ibn alNadı¯m’s Fihrist, translated into Arabic by Matta¯ b. Yu¯nus who is known to have worked from Syriac translations rather than original Greek texts. 148 See n. 77 above.

Who Wrote Alexander’s Commentary on Metaphysics Λ?

61

Freudenthal’s fourth argument has proved to be of equally crucial importance. He correctly remarked that, in the analysis expounded by Ibn Rušd, the order of the two “first” books of the Metaphysics is reversed. While in Alexander’s extant commentary on Books Α–Δ and in our Greek manuscripts Book Alpha Major comes first and Book Alpha Minor second, in 1397,5‒ 1398,7 B. Book Alpha Minor comes first and Book Alpha Major second; accordingly, Alpha Minor is explicitly called the “first book” in 1397,5 B., and Book Alpha Major is explicitly called the “second book” in 1395,4 B. It is true that this matches the order in which Ibn Rušd comments upon both books in his Tafsı¯r; yet Ibn Rušd is very unlikely to have changed the order he found in his source. For in his exposition of the book-by-book analysis he carefully preserves the odd book labels of Books Ζ, Η, Θ, Ι, and Κ, as we have noted, even though they strikingly differ from his own (Peripatetic) book labels. If he also reported the claim that Book Ka¯f is missing, he clearly displays the same scholarly objectivity, given that this claim is inexplicable within his own labelling system. So faithful a reporter would most certainly not have changed, without warning, the order of the two “first” books explicitly attested in the Arabic “Alexander”. Therefore, this feature of the analysis, as disturbing as it may seem, cannot be explained away by the assumption that it goes back to a punctual intervention – or rather to two punctual interventions (1395,4 B. and 1397,5 B.) – by Ibn Rušd himself: the order Alpha Minor ‒ Alpha Major turns out to be firmly embedded in the book-by-book-analysis. On this basis, we have tried to determine when the preference for this order may have originated. While it is true that Alexander of Aphrodisias doubted the authenticity of Alpha Minor’s position between Alpha Major and Beta, it is also true that he did not regard Alpha Minor as a suitable first book of the Metaphysics, but rather as an introduction to theoretical philosophy in general or to the Physics. Consequently, Freudenthal was correct in maintaining that the reversed order of the first two books does cast doubts on the attribution of the analysis to Alexander of Aphrodisias. He was too rash, however, in concluding that for this very reason the whole analysis should be attributed to Ibn Rušd. For Alpha Major is explicitly called the second book of the Metaphysics on p. 1395, 4 B. of the proem, i.e. in a passage unequivocally ascribed to “Alexander”. Furthermore, already Asclepius reports the view that the “true first book” of the Metaphysics should be conceived of as consisting of a smaller part (Alpha M i n o r ) and a larger part (Alpha M a j o r ), with Alpha Minor serving as an introduction to Alpha Major. We have concluded that the analysis of the Metaphysics expounded by Ibn Rušd may well go back in one way or another to Alexander’s genuine commentary on Λ, but that the analysis and possibly the whole commentary on Λ is likely to have been revised in later antiquity in the light of proposals such as the one attested by Asclepius.

62

Matteo Di Giovanni and Oliver Primavesi

Freudenthal’s fifth argument, concerning the reference to Nicolaus of Damascus in 1405,4‒8 B., succeeds in identifying a brief polemical addition probably made by Ibn Rušd himself. Yet this observation does not at all warrant Freudenthal’s attempt to extend the authorship of Ibn Rušd to the rest of the analysis. Freudenthal’s sixth argument, concerning Ibn Rušd’s concluding remark in 1405,9‒12 B., presupposes a very implausible interpretation of the reference to “what this man said”: “this man” is, in all evidence, “Alexander” and not Aristotle. It seems, then, that “Alexander’s” commentary on Book Λ as expounded and quoted in Arabic by Ibn Rušd cannot be ascribed directly to the historical Alexander of Aphrodisias, but rather to an unknown Greek reviser of an original work by Alexander of Aphrodisias. For we have shown that the main part of “Alexander’s” commentary on Book Λ 1–7 as quoted by Ibn Rušd must belong – pace Freudenthal – with the same translation of a Greek original into Syriac as the book-by-book analysis expounded in Part III of Ibn Rušd’s proem, and that this analysis cannot be attributed – pace Bouyges – to the historical Alexander of Aphrodisias without qualification. The bad news is that the authenticity of the quotations from “Alexander” interspersed in Ibn Rušd’s commentary on Book Λ 1–7 will probably have to be examined afresh, as well as the conclusions to be drawn from that evidence with regard to the Greek commentary ascribed to Michael of Ephesus. The good news, however, is that the book-by-book analysis expounded in Ibn Rušd’s proem has turned out to be a document pertaining to the ancient reception of the Metaphysics – partly by Alexander of Aphrodisias, partly by one of his successors.

Bibliography Afnan, M. A.: A Philosophical Lexicon in Persian and Arabic, Beirut 1969. Bertolacci, A.: On the Arabic Translations of Aristotle’s Metaphysics, in: Arabic Sciences and Philosophy, 15 2005, 241–275. Bonitz, H.: Alexandri Aphrodisiensis commentarius in libros metaphysicos Aristotelis, Berolini 1847. Bouyges, M. S. J. (ed.): Averroès, Tafsı¯r Ma¯ ba›d at-Tabı¯ ›at. Texte arabe inédit, établi par Maurice Bouyges, S. J., Premier Volume: Livres petit alif, grand alif, ba’, gim, Beyrouth 1938. Bouyges, M. S.: Averroès, Tafsı¯r Ma¯ ba›d at-Tabı¯ ›at. Texte arabe inédit, établi par Maurice Bouyges, S. J., Deuxième Volume: Livres dal, he, zay, hha’, tta’, Beyrouth 1942.

Who Wrote Alexander’s Commentary on Metaphysics Λ?

63

Bouyges, M. S.: Averroès, Tafsı¯r Ma¯ ba›d at-Tabı¯ ›at. Texte arabe inédit, établi par Maurice Bouyges, S. J., Troisième Volume: Livres ya’ et lam, Beyrouth 1948a. Bouyges, M. S.: La Métaphysique d’Aristote chez les Latins du XIIIe siècle: Connurent-ils le Proœmium d’Averroès à son commentaire du livre LamLambda?, in: Revue du moyen âge latin: Études − Textes − Chronique − Bibliographie, 5 1948b, 279–281. Bouyges, M. S.: Averroès, Tafsı¯r Ma¯ ba›d at-Tabı¯ ›at. Texte arabe inédit, établi par Maurice Bouyges, S. J., Notice, Beyrouth 1952. Brandis, A. (ed.): Scholia in Aristotelem. Collegit Christianus Augustus Brandis. Edidit Academia Regia Borussica, Berolini 1836. Brockelmann, C.: Geschichte der Arabischen Litteratur. I. Band, Weimar 1898. De Goeje, M. J.: Codices Orientales Bibliothecae Academiae LugdunoBatavae nuper acquisiti, in: De Goeje, M. J. (ed.), Catalogus Codicum Orientalium Bibliothecae Academiae Lugduno-Batavae, Leiden 1873, 317–328. Di Giovanni, M.: The Commentator. Averroes’s Reading of the Metaphysics, in: F. Amerini/G. Galluzzo (eds.), A Companion to the Latin Medieval Commentaries on Aristotle’s Metaphysics, Leiden 2014, 59–94. Dodge, B.: The Fihrist of al-Nadîm: A Tenth-Century Survey of Muslim Culture. 2 vols, New York–London 1970. Drossaart Lulofs, H. J. (ed.): Nicolaus Damascenus on the Philosophy of Aristotle. Fragments of the first five books translated from the Syriac with an introduction and commentary, Leiden 1965. Endress, G.: Der arabische Aristoteles und seine Lehrüberlieferung in Baghdad: Abu¯ Bišr Matta¯ ibn Yu¯nus, in: U. Rudolph (ed.), Philosophie in der Islamischen Welt. Band 1: 8.–10. Jahrhundert, Basel 2012, 290– 301. Endress, G./Adamson, P.: Abu¯ Yu¯suf al-Kindı¯, in: U. Rudolph (ed.), Philosophie in der Islamischen Welt. Band 1: 8.–10. Jahrhundert, Basel 2012, 92–147. Flügel, G. (ed.): Kitâb al-Fihrist, mit Anmerkungen herausgegeben von Gustav Flügel. Erster Band: den Text enthaltend, Leipzig 1871. Freudenthal, J./Fränkel, S. 1885: Die durch Averroes erhaltenen Fragmente Alexanders zur Metaphysik des Aristoteles: Berlin. Genequand, C. (ed.): Ibn Rushd’s Metaphysics. A Translation with Introduction of Ibn Rushd’s Commentary on Aristotle’s Metaphysics, Book La¯m, Leiden 1984. Geoffroy, M.: Remarques sur la traduction Usta¯t du livre Lambda de la Métaphysique, chapitre 6, in: Recherches de Théologie et Philosophie Médiévales 2003, 70, 417–436.

64

Matteo Di Giovanni and Oliver Primavesi

Golitsis, P.: La Recensio Altera du commentaire d’Alexandre d’Aphrodise à la Métaphysique d’Aristote et le témoignage des manuscrits byzantins Laurentianus Plut. 87,12 et Ambrosianus F 113 Sup., in: J. Signes Codoñer/I. Pérez Martín (eds.), Textual Transmission in Byzantium: between Textual Criticism and Quellenforschung, Turnhout 2014, 201– 232. Gründler, B.: Arabic Alphabet: Origin, in: L. Edzard/R. de Jong (eds.), Encyclopedia of Arabic Language and Linguistics (Online Edition), Leiden, 2016. Gutas, D.: Review of Charles Genequand: Ibn Rushd’s Metaphysics. A Translation with Introduction of Ibn Rushd’s Commentary on Aristotle’s Metaphysics, Book La¯m. Leiden 1984, in: Der Islam, 64 1987, 122–126. Gutas, D.: Die Wiedergeburt der Philosophie und die Übersetzungen ins Arabische, in: U. Rudolph (ed.), Philosophie in der Islamischen Welt. Band 1: 8.–10. Jahrhundert, Basel 2012a, 55–91. Gutas, D.: Gelehrte als Vermittler philosophischen Denkens, in: U. Rudolph (ed.), Philosophie in der Islamischen Welt. Band 1: 8.–10. Jahrhundert, Basel 2012b, 480–511. Halper, Y.: Revision and Standardization of Hebrew Philosophical Terminology in the Fourteenth Century: The Example of Averroes’s Long Commentary on Aristotle’s Metaphysics Δ, in: Aleph: Historical Studies in Science and Judaism, 13 2013, 95–137. Hasse, D. N.: Latin Averroes Translations of the First Half of the Thirteenth Century, Hildesheim–Zürich–New York 2010. Hayduck, M. (ed.): Asclepii in Aristotelis metaphysicorum libros Α–Ζ commentaria, consilio et auctoritate Academiae Litterarum Regiae Borussicae edidit Michael Hayduck. (= CAG VI 2), Berolini 1888. Hayduck, M. (ed.): Alexandri Aphrodisiensis in Aristotelis metaphysica commentaria, consilio et auctoritate Academiae Litterarum Regiae Borussicae edidit Michael Hayduck. (= CAG I), Berolini 1891. Hayduck, M. (ed.): Michaelis Ephesii in libros De partibus animalium, De animalium motione, De animalium incessu commentaria, consilio et auctoritate Academiae Litterarum Regiae Borussicae edidit Michael Hayduck. (= CAG XXII 2), Berolini 1904. Heubeck, A.: Schrift, Archaeologia Homerica. Denkmäler und das frühgriechische Epos, Göttingen 1979. Hoffmann, R.: La puissance argumentative de la logique spéciale dans la métaphysique d’Ibn Rushd, in: M. A. Sinaceur (ed.), Penser avec Aristote, Toulouse–Paris 1991, 667–676. ˇ ama¯l al-Dı¯n Muh ammad: Lisa¯n al-ʿArab, vol. 13 Ibn Manz u¯r, Abu¯ l-Fad l G 1884 (h arf al-la¯m), Bu¯la¯q Misr.

Who Wrote Alexander’s Commentary on Metaphysics Λ?

65

Jaeger, W.: Studien zur Entstehungsgeschichte der Metaphysik des Aristoteles, Berlin 1912. Jeffery, L. H. : Local Scripts of Archaic Greece. A Study of the Origin of the Greek Alphabet and its Development from the Eighth to the Fifth Centuries BC, Oxford21990. Kroll, G.: Syriani in Metaphysica commentaria, consilio et auctoritate Academiae Litterarum Regiae Borussicae edidit Guilelmus Kroll. (= CAG VI 1), Berolini 1902. Lane, E. W.: An Arabic-English Lexicon. Book I. – Part 2. ‫ج—خ‬, London– Edinburgh 1865. Luna, C.: Trois Études sur la tradition des commentaires anciens à la Métaphysique d’Aristote. (= Philosophia antiqua, Vol. LXXXVIII), Leiden–Boston–Köln 2001. Madigan, A. (trans.): Alexander of Aphrodisias: On Aristotle’s Metaphysics 4, Ithaca–New York 1993. Martin, A. (ed.): Averroès: Grand Commentaire de la Métaphysique d’Aristote (Tafsı¯r Ma¯ ba›d at-Tabı¯ ›at). Livre Lam-Lambda, Traduit de l’Arabe et annoté. (= Bibliothèque de la Faculté de Philosophie et Lettres de l’Université de Liège, Fasc. CCXXXIV), Paris 1984. Moraux, P.: Aristotelismus bei den Griechen von Andronikos bis Alexander von Aphrodisias: Dritter Band: Alexander von Aphrodisias. J. Wiesner (Hrsg.), Berlin–New York 1973. Moraux, P.: Der Aristotelismus bei den Griechen von Andronikos bis Alexander von Aphrodisias. Dritter Band: Alexander von Aphrodisias. J. Wiesner, Berlin–New York 2001. Müller, A.: Die griechischen Philosophen in der arabischen Überlieferung, Halle 1873. Peters, F. E.: Aristoteles Arabus. The Oriental Translations and Commentaries of the Aristotelian Corpus, Leiden 1968. Praechter, K.: Review of Hayduck 1904, in: Göttingische gelehrte Anzeigen, 168 1906, 861–907. Primavesi, O.: Ein Blick in den Stollen von Skepsis: Vier Kapitel zur frühen Überlieferung des Corpus Aristotelicum, in: Philologus, 151 2007, 51– 77. Ramón Guerrero, R.: Averroes: el «Proemio» de su Comentario al libro lambda de la «Metafísica», in: R. R. Guerrero (ed.), Memoria-Homenaje a Adolfo Arias (1945–1993). Anales del seminario de historia de la filosofía, Número extraordinario, Madrid 1996, 275–295. Ravaisson, F.: Essai sur la Métaphysique d’Aristote. Ouvrage couronné par l’institut, Paris 1837. Rose, V.: Valentini Rose de Aristotelis librorum ordine et auctoritate commentatio, Berlin 1854.

66

Matteo Di Giovanni and Oliver Primavesi

Ross, W. D. (trans.): Metaphysica, (J. A. Smith, W. D. Ross [eds.], The Works of Aristotle Translated into English, Volume VIII), Oxford 1908. Rudolph, U. (ed.): Philosophie in der Islamischen Welt. Band 1: 8.– 10. Jahrhundert, (Grundriss der Geschichte der Philosophie. Begründet von Friedrich Ueberweg. Völlig neu bearbeitete Ausgabe. Herausgegeben von Helmut Holzhey), Basel 2012. Steinschneider, M.: Die Metaphysik des Aristoteles in jüdischen Bearbeitungen: Ein Versuch von M. Steinschneider, in: Curatorium der Zunz-Stiftung (ed.), Jubelschrift zum neunzigsten Geburtstag des Dr. L. Zunz, Berlin 1884, 1–35. Steinschneider, M.: Die arabischen Uebersetzungen aus dem Griechischen. Zwölftes Beiheft zum Centralblatt für Bibliothekswesen, Leipzig 1893. Tarán, L.: Syrianus and Psudo-Alexander’s commentary on Metaph. E–N, in: J. Wiesner (ed.), Aristoteles Werk und Wirkung. Paul Moraux gewidmet. Bd. I. Aristoteles und seine Schule, Berlin–New York 21987, 215–232. Usener, H.: Scholia in Aristotelem. Supplementum, in: Aristotelis opera. Edidit Academia Regia Borussica, Volumen quintum, Berolini 1870. Vuillemin-Diem, G.: Anmerkungen zum Pasikles-Bericht und zu Echtheitszweifeln am größeren und kleineren Alpha in Handschriften und Kommentaren, in: P. Moraux/J. Wiesner (eds.), Zweifelhaftes im Corpus aristotelicum. Studien zu einigen Dubia. Akten des 9. Symposium Aristotelicum, Berlin–New York 1983, 157–192. Wendland, P. (ed.): Michaelis Ephesii in Parva naturalia commentaria, (= CAG XXII 1), Berolini 1903. Zonta, M.: Sulla tradizione ebraica di alcuni commenti alla Metafisica (Abu¯ l-Faragˇ ibn al-Tayyib e Averroè), in: Documenti e studi sulla tradizione filosofica medievale, 12 2001, 156–177. Zonta, M.: Il Commento medio di Averroè alla Metafisica di Aristotele nella tradizione ebraica. Edizione delle versioni ebraiche medievali di Zerah yah Hen e di Qalonymos ben Qalonymos con introduzione storica e filologica, 2 vols, Pavia.

The Program of Metaphysics Lambda (chapter 1) ENRICO BERTI

I. The first problem raised by Lambda 1 is its incipit: “Substance is the subject of he¯ theo¯ria, for the principles and the causes which are researched are those of substances” (1069a18–19).1 The ancient commentators interpreted he¯ theo¯ria as the science contained in the whole treatise of which the book Lambda is a part, i.e. the science of being qua being, or “first philosophy”, or “metaphysics”. So it was for Alexander, quoted by Averroes in fr. 2 Freudenthal (“in dieser Wissenschaft”),2 Themistius (“nous avons pour but de chercher les principes de la chose qui est”);3 pseudo-Alexander (“la scienza proposta, la scienza prima, la sapienza”);4 Averroes (“the inquiry in this book, the speculation in this science”);5 Thomas Aquinas (“ista scientia, sapientia”);6 commentators who unanimously consider book Lambda as the summit of the whole Metaphysics, and the science exposed in this treatise as a natural theology (Thomas Aquinas concluded his commentary with this book). The modern commentators, having realized that book Lambda is not so clearly connected with the whole Metaphysics (thanks to Jaeger and Ross, who nevertheless considered this book as “the theology” of Aristotle), tend to interpret he¯ theo¯ria as the inquiry that Aristotle is practising in this book, and consequently translate: “unsere Betrachtung” (Bonitz), “our inquiry” (Ross),

1 2 3 4 5 6

I usually follow ROTA, modifying it where it seems necessary to me. Freudenthal, J./Fränkel, S. 1885: Die durch Averroes erhaltenen Fragmente Alexanders zur Metaphysik des Aristoteles: Berlin. Thémistius, Paraphrase de la Métaphysique d’Aristote (livre Lambda) traduit de l’hébreu et de l’arabe, introduction, notes et indices par R. Brague, Paris 1999, 47. Alessandro di Afrodisia, Commentario alla Metafisica di Aristotele, a cura di G. Movia, Milano 2007, 1871. Ibn Ruschd’s Metaphysics, A Translation with Introduction of Ibn Ruschd’s Commentary on Aristotle’s Metaphysics, Book Lam, by Ch. Genequand, Leiden 1986, 65. S. Thomae Aquinatis In duodecim libros Metaphysicorum Aristotelis Expositio, edd. M.R. Cathala et R. Spiazzi, Taurini–Romae 1964, n. 2416.

68

Enrico Berti

“notre speculation” (Tricot), “mein Thema” (Düring), “la nostra indagine” (Reale), “este studio” (Calvo). The first who has widely discussed the problem is M. Frede in his essay on Lambda 1.7 Here he observes that the definite article has the force of a demonstrative, i.e. it means “this inquiry” and alludes to an ongoing enterprise well known by the listeners, which is not only the inquiry which follows, but also the inquiry pursued in the past by the Presocratics and the Platonists – i.e. by all the philosophers. This same inquiry, following Frede, is continued by Aristotle in his Metaphysics, and is the research of the principles of the totality of what there is, stated in books Alpha, Gamma and Epsilon and partially realized in books Zeta, Eta, Theta. A slightly different interpretation of he¯ theo¯ria has been proposed by S. Fazzo, who interprets it very generally as “la theo¯ria in senso aristotelico”, i.e. the theoretical philosophy, practised by Aristotle himself as by the preceding philosophers, but in conclusion identifies it with the science exposed in books Alpha, Beta, Epsilon and Zeta, exactly as Frede.8 In my contribution to the Festschrift für Wolfgang Wieland I have taken this interpretation again, finding a confirmation of it in the book Lambda, where Aristotle speaks about the theo¯ria as about “what is most pleasant and best” (1072b24) and says that astronomy “makes theo¯ria about substance which is perceptible but eternal”, while the other mathematical sciences, i.e. arithmetic and geometry, treat no substance (1073b5–7).9 But, differently from Fazzo, I have seen an allusion to this theo¯ria in book Alpha elatton of the Metaphysics, where Aristotle speaks of “the theo¯ria of truth” (993a30), identifying it with the theo¯retike¯ philosophia or episte¯me¯, which has truth as its end and which researches the cause, because “we do not know a truth without its cause” (993b20–24, ed. Jaeger).10 Now I would like to confirm this interpretation of the incipit of Lambda, because Aristotle characterises theo¯ria as a research of principles and causes without any justification, therefore presupposing a definition of theo¯ria which coincides with that of Alpha elatton, while the affirmation that its subject is substance, still absent in Alpha elatton, needs a justification, which

7 8 9

10

Frede, M. 2000b: Metaphysics Λ 1, in: M. Frede/D. Charles (eds.), Aristotle’s Metaphysics Lambda. Symposium Aristotelicum, Oxford, 53–80. Fazzo, S. 2008: L’esordio del libro Lambda della Metafisica, in: Rivista di filosofia neoscolastica, 110, 159–181. Berti, E. 2003: Il libro Lambda della Metafisica di Aristotele: Tra fisica e metafisica, in: G. Damschen/R. Enskat/A. Vigo (eds.), Plato und Aristoteles – sub ratione veritatis. Festschrift für Wolfgang Wieland zum 70. Geburtstag, Göttingen, 177–193. I knew the article of S. Fazzo before its publication. A complete status quaestionis is available in Salis, R. 2005: Il commento di pseudo-Alessandro al libro Λ della Metafisica di Aristotele, Soveria Mannelli, 55–60.

The Program of Metaphysics Lambda (chapter 1)

69

is furnished by the arguments that follow. It is true that Alpha meizon conceives philosophy as knowledge of the causes and principles, but it speaks of sophia or philosophia in general, not of theo¯ria.11 This is the first affinity, or parallelism, between Lambda and Alpha elatton, which will be followed by others, as we will see, so that all these affinities taken together will have important consequences for the interpretation of the whole chapter 1. Differently from Frede, who discusses widely the sentence “for the principles and the causes which are researched are those of substances”, I think that Aristotle proposes it as a quite obvious definition of the theo¯ria as a research of principles and causes, which results just from Met. Alpha elatton.

II. The sentence which follows the incipit is: “for if the universe is of the nature of a whole, substance is its first part; and if it coheres by virtue of succession, on this view also substance is first, and is succeeded by quality or by quantity” (1069a19–21). I accept the version of manuscripts E and J, which on line 21 have e¯ instead of the second eita reported by Ab and accepted by Ross and Jaeger, for reasons I will go on to explain. Here Aristotle presents two possibilities of conceiving the universe (to pan), i.e. the totality of what exists, either as a whole or as a series, and claims that in both cases the substance is the first reality, to justify in this way the incipit, following which the theo¯ria, i.e. the search for principles and causes, is about the substance. Alexander (apud Averroes, fr. 3 Freudenthal) explains only the first possibility, saying that the whole can be interpreted either as a “zusammenhängendes Ganzes” or as a genus (“Gattung”), which corresponds to the two meanings of holon distinguished by Aristotle in Met. Delta 26, but he excludes the whole as genus, because among the species of a genus there is no first. So he concludes that the substance is the first part of the whole because the existence of other parts depends on it, referring to Met. Zeta 1. In this way he establishes a parallelism between Lambda and Zeta which will remain a constant of the commentators. Themistius, on his side, interprets the first possibility as that of a natural whole, as the body of an animal or of a plant, saying that in this case the substance is like the heart, on which the life of the body depends, and the second possibility as a series like that of numbers or of geometrical figures. 11

With the exception of 989b24–25: “those who make their theo¯ria about all things which exist”, where the term theo¯ria is only used, to say so, en passant, without any official definition, as in Alpha elatton.

70

Enrico Berti

Pseudo-Alexander refers without hesitation to the Aristotelian theory of categories, speaking explicitly of the substance and of the nine categories or ten categories. In this interpretation, the first possibility is that of a whole constituted by the substance and the accidents, and the second possibility is that of a series of categories, ordered in a rigorous succession (substance, quality, quantity and so on). This interpretation is based on the version of manuscript Ab (eita … eita), or perhaps it is as its origin. But Aristotle never considers the categories as to be ordered in a rigorous series, and for this reason I prefer the version of E and J. Averroes refutes the interpretation of Themistius, agrees with Alexander and adds that Aristotle’s thesis is only the second one, where he sees the series of the categories, whose unity is that of the healthy things or of the medical things, i.e. relative homonymy. Thomas Aquinas, on the contrary, agrees with Themistius, but mentions the disagreement of Averroes. Among the modern commentators Bonitz presents the first possibility as that of a whole constituted by matter and form, saying that substance is just its form, and the second one as that of the series of categories. Tricot and Reale agree with Bonitz and with pseudo-Alexander. All these interpreters think that Aristotle is presenting two possibilities derived from his own philosophy. Only Ross proposes a different interpretation, supposing that with the second possibility, that of the series, Aristotle alludes to the position of Speusippus, who conceived reality as constituted by a series of plans (numbers, geometrical magnitudes, perhaps souls, sensible things – every one with its proper principles), which Aristotle criticises in ch. 7 and 10 of Lambda. Elders thinks that the terms evoke a division of being related to the Platonic scheme of reality. For in such a classification to poion denotes the differentiae specificae, and hence follows upon ousia, together with which it belongs to the beings per se.12 Frede takes up the interpretation of Ross, adding that the second possibility is not only the series admitted by Speusippus, but also the series admitted by Plato (ideal numbers, forms, mathematical objects, sensible things) or by Xenocrates (ideal numbers identified with forms, mathematical objects, sensible things). Besides, by looking at the following part of the text, where Aristotle will quote “the ancients” (1069a25: hoi arkhaioi), likening them to his contemporaries (1069a26: hoi … nun), i.e. the Platonists, Frede proposes to interpret the first possibility, the conception of the universe as a whole, as the position of the Presocratics. Frede, moreover, retains the text of E and J (eita …e¯), observing that it is confirmed also by the manuscripts C and M, which belong to the same

12

Elders, L. (ed.) 1972: Aristotle’s Theology, A Commentary on Book Λ of Metaphysics, Assen, 76.

The Program of Metaphysics Lambda (chapter 1)

71

family of Ab, and is read also by Alexander (ap. Averroes) and Themistius. I completely agree with Frede, because in the conception of the universe of some Presocratics (the Milesians, Heraclitus, the Parmenides of the second part of his poem, Empedocles) the universe is a sort of immense body (a sphere), at the base of which there are the elements (water, air, fire, earth), which for Aristotle are true substances (ousiai), while the other things are mere affections of the substance (pathe¯, cf. Met. A 3.983b9–10). Concerning the second possibility, the series, I observe that for all the Platonists each plan of reality is made by substances, so that the series to which Aristotle alludes is not the series of categories (which for him is not a true series), but the series of the substances. Platonists also admitted other things besides the substances, as qualities or quantities, but they conceived them as dependent on substances.13 So the primacy of substance is justified, at the eyes of Aristotle, not only by his own philosophy, i.e. by the doctrine of the categories, which is certainly present in book Lambda, but also by the philosophies of all his predecessors.

III. “At the same time these latter [quality and quantity] are not even beings in the unqualified sense (haplo¯s), but (alla) are qualities [ROTA: quantities?] and movements – else even the not-white and the not-straight would be; at least we say even these are, e.g. ‘this is not white’.” (1069a21–24) On line 22 the manuscripts E and J have allá, edited by Ross, while Alc has hoion, edited by Jaeger under the influence, as we will see, of pseudo-Alexander. In fact, Themistius sees in this passage a mere opposition between substances, which are in the unqualified sense, and their attributes, which are only as attributes of substances, and reads the examples as examples of attributes, i.e. “ceci n’est pas blanc”, or “ceci n’est pas droit”. Pseudo-Alexander, on the contrary, sees in this passage a true reference to the whole doctrine of categories, and for this reason considers “qualities and movements” only as examples of categories, reading hoion instead of alla. Moreover he introduces a very complicated theory about the not-white and the not-straight, saying that they exist only as privations, and that privation presupposes possessions, and that these presuppose substances. Averroes also sees an opposi-

13

In Tim. 49 d–e Plato says that the sensible things are not a “this” (touto), i.e. a substance, but a “such” (toiouton), i.e. a quality of the elements.

72

Enrico Berti

tion between the substance, which exists “by itself”, and the attributes, which exist “in substance”. Thomas distinguishes very well substances as simpliciter entia and attributes as entis entia. Ross and Tricot, on their side, consider “not-white” and “not-straight” as subjects, and translate “a not-white exists”, while I prefer to consider them, with Themistius, as attributes, because Aristotle often says that being something (einai ti) does not imply being in an unqualified sense (being haplo¯s) (Soph. el. 25, 180a36–38), and that we say that non-being is non-being (Met. Gamma 2, 1003b10), without implying that it is a true subject. The meaning of this argument, in my opinion, is that things as qualities and movements, e.g. “white” and “walks”, are not being in the unqualified sense, but they are mere attributes of substances, i.e. of subjects, in the same way in which “not-white” and “not-straight” are attributes of other things, i.e. of substances. This is the first argument drawn by Aristotle from his own philosophy in favour of the primacy of substance. The second one is immediately following: “further, none of the others can exist apart (kho¯riston)” (1069a24). Here evidently kho¯riston does not mean separable from matter, but separable from other things, a property which holds only for substances, because attributes cannot exist apart from substances. Themistius explains this passage very well, saying that the substance can be separated from the accidents not in the sense that it could exist without accidents, but in the sense that it continues to be the same even if its accidents change, while each accident can exist only when it belongs to a substance. Although there are many parallelisms between these doctrines and those exposed by Aristotle in book Zeta, as many interpreters have noted, this does not mean that Lambda presupposes Zeta. Undoubtedly Lambda presupposes the conception of substance exposed in the Categories, following which substance is what neither is said of a subject nor is in a subject (2a11–13), because it is itself a subject, and for this reason it is prior to the other categories. In Categories there is also the notion of existing apart (1a25: kho¯ris). But the doctrine of Met. Z is much more sophisticated, because it establishes the primacy of substance in notion, in knowledge and in time (ch. 1), and it mentions the theory of the relative homonymy (ch. 4), which in Lambda is completely absent. Besides, the aim of book Zeta is not to search for the principles and causes of substance, as the aim of Lambda, but to establish “what is substance” (1028b4), i.e. which, among the various candidates to the title of substance are the most appropriate. For all these reasons I cannot think like M. Burnyeat, followed by Fazzo, that Lambda 1 is a résumé of Zeta 1.14 I think, on the contrary, that it is an earlier sketch of the primacy of substance, not yet as sophisticated as Zeta. 14

Burnyeat, M. F. 2001: A Map of Metaphysics Zeta, Pittsburgh, 148 ff.

The Program of Metaphysics Lambda (chapter 1)

73

IV. “And the old philosophers also in effect (ergo¯i) testify to this; for it was of substance that they sought the principles and elements and causes. The thinkers of the present day have ranked rather (mallon) the universals as substances (for genera are universals, and these they describe rather as principles and substances, owing to the logical nature of their inquiry); but the old thinkers ranked particular things as substances, e.g. fire and earth, but not what is common to both, body.” (1069a25–30) The mention of preceding philosophers, that with Frede we have supposed implicit at the beginning of the chapter, here becomes explicit, because they are presented as witnesses of the primacy of substance, but only “in effect”, because they did not speak openly of substance; they did not use – at least the Presocratics – the word ousia, but they searched for the principles of substance and they treated these principles as substances, on the basis of a tacit axiom, shared by Aristotle, following which the principles of substances have to be substances. The testimony of preceding philosophers becomes in this way a further argument for the primacy of substance. Themistius explains this passage saying that the ancient philosophers searched for the principles of substances and not of states, dimensions, or other similar things, i.e. accidents. This is not clearly said by Aristotle, but can be implied. The contemporary philosophers also search for the principles of substances, but they think that the genera, i.e. the universals, are more worthy to be substances and so they identify the principles with the genera. This is due to their logical training, because the genera exist only in the thought. The earlier thinkers, on the contrary, following the senses, thought that the particular things were more worthy to being substances and so they posed as principles of substances particular things like fire or earth. In this way Themistius admits different degrees of right to the title of substance and explains the difference between ancient and contemporary thinkers as an opposition between a research based on the senses and a research based on the reason. Pseudo-Alexander is more radical: he identifies the logical inquiry (logiko¯s) with an empty one (keno¯s) and he declares that universals are non-beings, i.e. do not exist, but he wants to save Plato from this disastrous result (he is a Neoplatonist!), saying that the contemporary philosophers of which Aristotle is speaking are not Plato but others. Averroes, on the contrary, recognises that the contemporary philosophers are first of all Plato, observing that Aristotle has refuted his thesis in book Zeta. He interprets the logical way of searching as a dialectical one, i.e. of a research which moves from inadequate

74

Enrico Berti

premises, or from premises which follow persuasion. Aquinas agrees that the contemporaries are Plato, who researched logice, and identifies the ancients with Democritus and Empedocles. Among the modern commentators Bonitz identifies the contemporaries of Aristotle with the Platonists, Plato included, saying that he does not see which others they could be, and explains their doctrine as produced by a total engagement in notionibus cognoscendis, not necessarily in a pejorative sense. Ross agrees with Bonitz, including, against pseudo-Alexander, Plato among the contemporaries, while Tricot, though including Plato, agrees with pseudo-Alexander, interpreting their logical inquiry in a pejorative sense of an empty one. Elders observes rightly that according to this line Platonising philosophy was dominant at the time Aristotle wrote this passage and therefore that the chapter is early. Frede observes that Aristotle is speaking of all the Platonists and he explains the mallon of lines 27–28 saying that they rank universals as being substances “more strictly speaking or as having more of a title to being regarded as substances”, or that the principles are “more substantial” than the things that depend on them. This is Platonic language, because Aristotle does not admit degrees of substantiality, but it is possible that he attributed this conception to the Platonists. Frede moreover explains the “logical inquiry” of Platonists as an “abstract method of inquiry”. Now, it is highly evident that Aristotle does not share the conception of the Platonists, following which the genera are substances and consequently the principles of substances, but in this passage the mention of their logical way of inquiring not necessarily has the pejorative sense of an empty way of inquiring, to which Aristotle himself assimilates it (E. E. I 8.1217b21). Here it is simply an explanation of the reason why they hold up their thesis, i.e. the research of the definitions (logoi), the en tois logois skepsis mentioned in Met. Alpha (6.987b31–32). There are passages where Aristotle opposes the logical way of searching (logiko¯s) to the physical one (phusiko¯s), preferring clearly the latter (Phys. III 5.204b4, 10; De gen. et corr. I 2.316a11). Here in Lambda 1 he implicitly opposes the logical way of inquiring of the Platonists to the physical – without mentioning this name – of the Presocratics and probably preferring the latter. But it must not be forgotten that the “logical” and the “physical” character of discourses were, together with the “ethical” one, the three types of characters which sentences or problems could have for the Academy, and consequently the three parts of philosophy, as results from Xenokrates fr. 1 Heinze, and that Aristotle himself in his Topics (I 14.105b20–29) shared this classification, that would have been canonical in the Hellenistic period. If in Met. Lambda Aristotle does not follow the logical way of inquiry, and if we exclude that he is following the ethical one, because his research is a theo¯ria

The Program of Metaphysics Lambda (chapter 1)

75

about substance, it remains that he follows the physical way of inquiring, so that his theo¯ria reveals itself as a physical one.

V. “There are three kinds of substance – one that is sensible (of which one subdivision is eternal and another is perishable, and which all recognize, as comprising e.g. plants and animals) – of this we must grasp the elements, whether one or many; another which is immovable, and this certain thinkers assert to be separated (kho¯riste¯n), some dividing it in two, others combining the Forms and the objects of mathematics into one class, and others limiting it only to the objects of mathematics.” (1069a30–36) I disagree in part with the ROTA, which speaks correctly of “kinds of substances” instead of “substances”, but interprets the first sentence as an affirmation of existence made by Aristotle, while I think that the philosopher is simply considering three possible kinds of substance, or three kinds whose existence was admitted by the preceding thinkers, but which, at least for the eternal substances and for immovable ones, needs to be demonstrated. Besides, I think that the word kho¯riste¯n does not simply mean “capable of existing apart”, like the kho¯riston of the line 24, because all substances, not only the immovable ones, are capable of existing apart; it means rather “separated from matter”, i.e. immaterial, or “separate from the sensible substance”. The text of the division of the kinds of substance is problematic, because all the manuscripts and Bekker have on line 32 he¯ d’aidios, that Ross and Jaeger suppress as redundant, but marking a strong separation with what precedes. Everybody admits that the substance whose existence is recognized by all is only the sensible perishable substance, i.e. plants and animals, but the text of the manuscripts and Bekker could mean that we must grasp the elements only of the eternal substance, while the text of Ross and Jaeger means that we must grasp the elements of all sensible substance, both perishable and eternal. Alexander of Aphrodisias already had the same problem, because he says (fr. 4b F) that there are two manuscripts, the one saying that we must discern the elements of both the sensible substances, i.e. of the perishable and of the eternal one, while the other says that we must discern the elements only of the eternal substance. He declares that the first manuscript is better, because the aim of Aristotle is the discussion of the whole sensible substance, including all the things which are in the universe.

76

Enrico Berti

Themistius says that both the sensible substances, perishable and eternal, are accepted by everybody, because they are visible. As for eternal substance, he means the heavenly bodies: he must think that Aristotle is speaking only of the existence of these substances, because the eternity of the celestial bodies was not accepted by everybody, but only by Aristotle. Concerning the text, he seems to have the manuscript preferred by Alexander. Pseudo-Alexander, on the contrary, seems to read a manuscript in which the mention of the eternal substance is only on line 32, because he affirms that it is not of the eternal substance that we must grasp the elements (as his text would seem to suggest), but only of the sensible one, which is predicated both of the corruptible and the incorruptible substance. As results from the context, for “elements” he means only the material ones, i.e. the four traditional elements for the corruptible substances and the “fifth body” for the heavenly one. In this way pseudo-Alexander is opposing – as R. Salis has noted – the eternal substance to the sensible, incorruptible substance, giving the impression of identifying the eternal substance with the immobile one, perhaps influenced by what Aristotle will say in Lambda 6, 1071b4–5.15 In a preceding fragment about the same passage (fr. 4a F) Alexander said that the demonstration of the principles of beings, whatever they are, belongs to the metaphysician, although they are used by the natural philosopher, who does not explain them, but merely postulates them, for the immovable substance is the principle and the cause of natural things and it is that which Aristotle is now discussing above all. This affirmation raises a big problem for Averroes, who observes that the words of Alexander contain some obscurity, because natural philosophy explains the existence of the principles of perishable substance at the beginning of the Physics, but it explains also the existence of the principles of eternal things (the heaven, made by ether, which Averroes calls “the fifth body”) at the end of the VIII book of the Physics. Averroes continues saying that Ibn Sina too, i.e. Avicenna, thinks that it belongs to the first philosopher to explain the existence of the principles of sensible substance, whether eternal or not, while the natural philosopher only postulates the existence of nature. But for Averroes this is an error, produced by the obscurity of the words of Alexander. The true solution of the problem, for Averroes, is that physics explains the existence of the principles as principles of the movable substance, whereas the metaphysics inquires into them as principles of substance qua substance, not of the movable substance (1425, 75 Genequand). For this reason Averroes prefers the text of the second manuscript, for the primary aim of Aristotle, in his opinion, is the explanation of the eternal substance. 15

Salis (see note 10), 69.

The Program of Metaphysics Lambda (chapter 1)

77

Among the modern commentators Bonitz observes that the substances of which everybody, for Aristotle, admits the existence, are not only the corruptible substances, but also the eternal one, as results from Metaphysics Eta 1.1042a7. For this reason he associates himself to the commentary of Themistius, but he notes that the relative pronoun on line 32 which should refer to both the sensible substances does not, since its position seems to refer only to the eternal one. Ross, as we said, eliminates from the text the second mention of the eternal substance (line 32) and, concerning the interpretation of Bonitz about the sensible substances accepted by everybody, observes that the eternity of the heavenly bodies was not universally accepted. In any case he thinks that the discussion of all the sensible substances, both perishable and eternal, is made in chapters 2–5 of the book Lambda. Tricot and Reale follow Ross. Frede dedicates the whole appendix of his essay to the discussion of the text. Here he observes that two manuscripts of the family of Ab, i.e. M and C, have only one mention of the eternal substance, on line 32, and so seem to justify the interpretation of pseudo-Alexander. In his opinion Aristotle does not mean to say that we have to determine the elements only of eternal substance, for in following chapters he proceeds to determine the elements for sensible substance quite generally. Concerning the substances accepted by everybody, Frede agrees with Bonitz contra Ross, observing that the relevant point is not whether it was universally agreed that celestial bodies are eternal, but whether they were treated as substances. In conclusion, following Frede, the division of the three kinds of substance is introduced by Aristotle to capture all candidates for “substancehood”. The first two kinds, i.e. sensible perishable and sensible eternal substances, are accepted by everybody and their elements or principles are the object of the inquiry of chapters 2–5. The third kind, i.e. immovable substances, are those admitted by the Platonists, but Aristotle has in mind also the unchanging substances demonstrated by himself in Phys. VIII and later in Lambda 6, so he refers to Platonists in part for confirmation of the assumption that there are such separate unchanging substances. The division, inside the sensible substance, between perishable and eternal substance is due, following Frede, to the fact that eternal substances are the crucial link between the incorruptible and the corruptible; so – he says – Aristotle must think that the postulation of such substances is necessary to explain how eternal substances can be the principles of corruptible things. I agree with Frede, observing only that this passage of chapter 1 announces the research of the principles of sensible substance, corruptible and eternal, in chapters 2–5 (the principles of eternal substance are mentioned in chapter 2.1069b25–26), but it says nothing about the existence of the eternal sub-

78

Enrico Berti

stance, which will be demonstrated in chapter 6, as a necessary link to the demonstration of the existence of the immobile one. Concerning the identity of the philosophers who accept the immovable substance as separated, Alexander (fr. 5 F.) says that the first view (two substances, Forms and mathematical objects) is that of Plato, the second one (one substance, identifying Forms and mathematical objects) is that of some disciples of Plato, and the third (only mathematical objects) is that of the Pythagoreans. For Pseudo-Alexander the first view is that of Plato, the second one of some Pythagoreans and the third of other Pythagoreans. Averroes seems to accept the identification proposed by Alexander, and Thomas Aquinas also speaks of the “Platonici” and the “Pythagorici”. Now, we know from other testimonies (cf. the editions of their fragments by Heinze and Lang) that the second view is that of Xenocrates, whom Aristotle never mentions by name in the Metaphysics, and the third one is that of Speusippus, whom Aristotle mentions, but without explicitly attributing this view to him. Both could be considered Pythagoreans, because they attributed to numbers the function of principles of sensible things, and because they were probably influenced, as – I think – the old Plato, by the Pythagoreans, but both belonged to the Platonic Academy. This means that the immovable substance of which Aristotle speaks in our passage is not the substance of which he will demonstrate the existence in the sequel of book Lambda, but the immobile substance admitted by the Platonists, i.e., by the philosophers of the Academy.

VI. “The former two kinds of substance are the subject of physics (for they imply movement), but the third belongs to another science, if there is no principle common to it and to the other things (ei me¯demia autois arkhe¯ koine¯)” (1069a36–b 2). The text of Aristotle contained in the commentary of Alexander (fr. 6 F.), as is reported in Arab, seems to have epei, instead of ei, because Freudenthal translates “da sie auch nicht ein gemeinsames Prinzip haben”, and Genequand, “for they not even have a common principle”. But M. Crubellier – quoted by Frede, p. 73, n. 4 – points out that the other Arabic translation has ei, and that there is no indication in Averroes that Alexander read epei. Nevertheless Alexander and all the commentators, both ancient and modern, agree that there is no common principle between movable and unmovable substance, and therefore physics has the movable substance as its object and the first philosophy, i.e. metaphysics, has the unmovable one. Moreover, Al-

The Program of Metaphysics Lambda (chapter 1)

79

exander thinks that physics takes over the principles of movable substances from first philosophy, raising the doubts, as we will see, of Averroes. Themistius, in fact, affirms that the two kinds of substance have nothing in common, neither principles nor other things, so that Brague in commenting his paraphrase speaks of a “totale équivocité entre la substance mouvante et la substance immobile”. Pseudo-Alexander affirms that the unmovable substance, which he calls “intelligible and divine” (noe¯te¯ kai theia), is the object of the philosophy “exposed in this book”, i.e. first philosophy, but he reads ei, because he explains that here ei is equal to epeide¯, adding that he thinks (oimai) that this is the sense (dianoia) of the words of Aristotle. Averroes observes that the affirmation made by Alexander, i.e. that physics takes over its principles from metaphysics, if it is taken at face value, is wrong, because physics searches itself for its principles, including the unmoved mover, as results from Phys. VIII. Nevertheless Averroes also thinks that there is no common principle between movable and unmovable substance, and consequently that the study of the latter belongs to metaphysics. Afterwards Averroes adds that the affirmation of Alexander must not be taken literally, because physics explains the material and the efficient cause of movable substance, while metaphysics explains its formal and final cause. Therefore, following Averroes, there is a strict closeness, or a continuity, between the two sciences, because their principles are not different in their being, but only in the way in which they are considered. Alexander – he says – was taken too literally by Ibn Sina, who consequently was wrong. An original and interesting interpretation of this passage was given by Thomas Aquinas, following which, even if there is something common to the two kinds of substance, the movable and the unmovable, the consideration of both belongs to the science which considers the common element, i.e. metaphysics. Consequently physics considers the movable substance qua movable, while metaphysics considers the movable and the unmovable substance in what they communicate, i.e. in being beings and substances (in hoc quod sunt entia et substantiae). As the science of being and substance is the science of being qua being proposed in Met. Gamma 1, the sense of Thomas’ interpretation is that the other science mentioned by Aristotle in our passage is not directly the “theology”, as pseudo-Alexander seems to think, but the science of being as being, which is also “theological”, as Aristotle says in Met. Epsilon 1. The modern commentators are unanimous in interpreting the ei of the conditional sentence as an epei. Bonitz translates it as “da”; Ross affirms that the science which differs from physics is metaphysics and that it is exposed in chapters 6–10 of Lambda; Tricot translates “puisqu’elle n’a aucun principe commun” and declares that this science is the metaphysics and cites pseudoAlexander as a confirmation; Reale translates the ei as “dal momento che”,

80

Enrico Berti

but identifies the different science with that indicated by Aristotle in E 1 and K 7, following Thomas Aquinas in this interpretation. Elders writes: “Because both ousiai have no principles in common, they are the object of different sciences”. Only an Italian translator, Carlo Viano, translates “se non c’è nessun principio che sia comune a tutte queste sostanze”.16 Frede is the first scholar who discusses widely the problem posed by the conditional sentence “if (ei) there is no principle common to it and to the other things” and asks himself whether this condition is satisfied or not.17 First of all he observes that manuscripts C and M read kine¯seo¯s instead of koine¯, but he prefers to maintain the text of the other manuscripts. Afterwards he argues that from chapters 6–10 of Lambda it results that the first unmoved mover is in a sense a principle common to the movable and the unmovable substances. It is evident, in fact, that the first unmoved mover is the principle of the movement of the first heaven and, by means of this latter, of the movement of the Sun (incorruptible movable substance), which is the principle of all the changes of the corruptible substances on the earth. But the first unmoved mover seems to be the principle also of the other unmoved movers, for instance being the object of their thought and so their final cause. In this way the condition posed by the conditional sentence would not be satisfied and we ought to accept – following Frede – the “unwelcome conclusion” that the physics would be the science of all the substances. But – continues Frede – the conditional sentence is ambiguous and leaves open the possibility that there is a discipline other than physics which studies both sensible and unchanging substance. To this possibility Frede mentions an opinion expressed by Crubellier, following which the whole sentence which precedes the conditional may constitute a twofold apodosis, i.e. it may mean that, if there is no common principle, physics deals with sensible substances and another science, which is theology, deals with unchanging substances; but if there is a common principle, neither physics deals with sensible substances, nor theology deals with unchanging substances, and 16 17

Aristotele, La metafisica, a cura di C. A. Viano, Torino 1974. The problem was briefly raised also by Ch. Kahn, who observed in a footnote that the condition of the non existence of a common principle remains unfulfilled, because the Prime Mover is precisely this common principle (cf. Kahn, C. H. 1985: The Place of the Prime Mover in Aristotle’s Teleology, in: A. Gotthelf (ed.), Aristotle on Nature and Living Things, Bristol–Pittsburgh, 311–338, esp. 319, n. 10). On the contrary D. Devereux has noted that the principle which the different genera of substance seem to have in common, i.e. the Prime Mover, is not actually a principle of suprasensible substance. Nevertheless Devereux admits that there is some overlap between the subject matters of physics and first philosophy (Devereux, D. 1988: The Relationship between Theophrastus’ Metaphysics and Aristotle’s Metaphysics Lambda, in: W. Fortenbaugh/R. W. Sharples (eds.), Theophrastean Studies, New Brunswik–Oxford, 167–188, esp. 176–177).

The Program of Metaphysics Lambda (chapter 1)

81

there is another science, which is neither physics nor theology, which deals with all kinds of substances. This science is – concludes Frede – “the global theory of what is”, i.e. the science of being as being and first of all of the substance, mentioned by Aristotle in Met. E 1 and partially developed in Met. Z and H. It is the interpretation – it seems to me – of Thomas Aquinas and Reale. This conclusion has provoked the reaction of P. Donini, who is an admirer of Frede and recognizes in him the merit of having discussed the question more than any other scholar.18 For Donini there is also a principle common to sensible and unchanging substance, which is the first unmoved mover, not because is it thought by the other unmoved movers (what Aristotle never says), but because it is the first and in this way it is principle to the others. Besides – observes Donini – at the end of the book Lambda, against Speusippus, who admitted different principles for different kinds of substances, Aristotle solemnly declares that the ruler of the universe must be only one. This means that the condition posed by the conditional sentence is not satisfied and consequently physics is the science which deals with all the kinds of substance. The possibility of admitting a twofold apodosis is excluded, following Donini, by the de¯ which introduces the whole period, by the parenthesis which explains why physics deals with the sensible substance and by the particles men-de which oppose the first sentence to the apodosis of the conditional. The thesis that physics is the science of all the substances belongs – following Donini – to a period of Aristotle’s evolution when he still shared the Academic conception of physics as the unique theoretical science, distinguished from logic and ethics, as results from the Topics, the Protrepticus and the book Alpha elatton of the Metaphysics. To this regard Donini declares that he agrees with my hypothesis, following which the book Alpha elatton belongs to a period in which Aristotle did not yet distinguish physics from metaphysics.19 In a second period – following Donini – Aristotle opposed physics to metaphysics conceived only as the science of the unchanging substances (Phys. I–II, De cael., De part. an. I–II), and finally, in the last period, Aristotle conceived metaphysics as the universal science (Met. Γ, E, Z). Therefore Lambda belongs to a different period from Gamma and Epsilon and must not be interpreted in the light of the latter two. 18

19

Donini, P. 2002: Il libro Lambda della Metafisica e la nascita della filosofia prima, in: Rivista di storia della filosofia, 57, 181–199 (repr. in Id., Commentary and Tradition 2011, 17–35). Berti, E. 1983: La fonction de Metaph. Alpha elatton dans la philosophie d’Aristote, in: P. Moraux/J. Wiesner (eds.), Zweifelhaftes im Corpus Aristotelicum. Studien zu einigen Dubia. Akten des 9. Symposium Aristotelicum, Berlin–New York, 260–294.

82

Enrico Berti

In my contribution of 2003 I observed that the unmovable substance of which Aristotle is speaking in our passage is not the substance whose existence he will demonstrate in the second part of the book Lambda, but it is the substance accepted by the Academicians, i.e. the Forms and the mathematical objects, and consequently the principles of this substance are those admitted by the Academicians, i.e. the One and the Indefinite Dyad, or the One and the Many. These principles were refused by Aristotle, but for the Academicians (surely for Plato, probably for Xenocrates too, not for Speusippus) they were common to the unmovable and to the movable substances, so that for them there was a unique science, which studied the principles of all things, as Aristotle testimonies for instance in Met. A 9.992a18–993a10.20 For Aristotle the functions of the One and of the Indefinite Dyad were assumed respectively by the form and the matter (see Lambda 2–3), which are by analogy the same principles of all sensible substances, both corruptible and eternal (see Lambda 4–5); and Aristotle added to them, as efficient cause, the first unmovable mover, which is numerically (not by analogy) the same principle for all the movable substances (cfr. Lambda 4, 1070b34–35; 5,1071a35–36; 7, 1072b13–14). With Frede and Donini I supposed that the first unmovable mover is a principle also for the unmovable substances, because in Lambda 8, 1073a23– 24, Aristotle says that it is “the principle and the first of beings” (he¯ … arkhe¯ kai to proton to¯n onto¯n), and in Lambda 10 he criticises the multiplicity of principles accepted by Speusippus and affirms the necessity of only one ruler (1075b37–1076a4). This holds even if Aristotle does not explain how, i.e. in which way, the first unmoved mover is the principle of the other unmovable movers. Consequently for the author of the book Lambda, so as for the Academicians, there was a unique science which studied the principles of all things, a science which for the Academicians was the dialectic, conceived in the Platonic sense as the synoptic science (see Republic VII), while for Aristotle it was the research for the principles of substance, mentioned at the beginning of ch. 1. The change introduced by Aristotle in comparison to the Academicians was the distinction between the logical, i.e. dialectical in the Aristotelian sense, and the physical method of the research, reminded also, as we have seen, in Lambda 1, 1069a28, so that the science of the principles for the Academicians follows the logical method and for Aristotle follows the physical one.

20

On this passage see my article, La critica di Aristotele alla scienza universale in Metaph. A 9, (Berti 2009: La critica di Aristotele alla scienza universale in Metaph. A 9, in: L. Cardullo (ed.), Il libro Alpha della Metafisica di Aristotele tra storiografia e teoria, Catania, 133– 154).

The Program of Metaphysics Lambda (chapter 1)

83

We find the same conception of a unique science of the principles, the Aristotelian Prinzipienlehre, in book Alpha elatton, where the “theo¯ria about the truth”, or about the causes (ch. 1), i.e. the research for the principles, or for the primary causes (ch. 2), is identified with physics (ch. 3.995a18). This conception, obviously, is very different from that which we find in Metaphysics E 1, where philosophy, or the sciences (episte¯mai), or “all thought” (pasa dianoia), are divided into theoretical, practical and poietical sciences, and among the theoretical ones physics is clearly distinguished from first philosophy, i.e. from metaphysics, which is a universal (katholou) science, because it searches for the principles of being qua being, and is also a “theological” (theologike¯) one, because it concerns the causes of the divine things (ta theia), i.e. of the heavenly bodies, which are unmovable substances and are themselves divine. But it is the same conception which we find in a work of the young Aristotle, the Protrepticus, where philosophy (to philosophein) is divided into the sciences “about just and convenient things” (peri to¯n dikaio¯n kai to¯n sumpheronto¯n), i.e. ethics, or practical philosophy, and the sciences “about nature and the other truth” (peri phuseo¯s kai te¯s alls ale¯theias) (fr. 5b Ross, B 31–32 Düring), i.e. a unity of physics and metaphysics. This is the reason why, like Donini, I tend to suppose that Met. Lambda and Met. Alpha elatton (where we find the same division of philosophy into theoretical and practical, without mention of the poietical one, cfr. 993b20– 21, as in the Protrepticus) are earlier than Met. Epsilon (and Gamma, and Zeta), and reflect a conception of philosophy as a Prinzipienlehre, still near to the Academic conception of philosophy, but which proceeds with a physical, not a logical or dialectical, method. Afterwards this conception was substituted by the more Aristotelian conception of philosophy as divided into three kinds of science, theoretical, practical and poietical, and of the theoretical one as divided into mathematics, physics and metaphysics, where the latter is the science of the principles of being qua being.21 A new contribution to this discussion is now put forward by D. Lefebvre, who dedicated a whole article to our passage22. He rightly observed that the autois in line 1069b2 can be interpreted not as a neuter which, following the

21

22

The chronological posteriority of the tripartite division of the sciences in comparison to the bipartite one has already been noted by Manison, A. 1958: Philosophie première, philosophie seconde et métaphysique chez Aristote, in: Revue philosophique de Louvain, 56, 165– 221. Lefebvre, D. 2012: La question de l’unité d’une science des substances: Interprétations de Métaphysique, Λ, 1, 1969A36–B2, in: M. Bonelli (ed.), Physique et métaphysique chez Aristote, Paris, 113–132. The author has kindly sent me a copy of this article, which I saw after the Colloquium in Bonn (November, 28 th–December, 1st, 2010).

84

Enrico Berti

traditional interpretation, refers to the kinds of substance, because in this case the feminine, like ekeinai in line 1069a36, would be more natural. It should be interpreted as masculine, referring to the tines of line 1069a34, i.e. the Academicians, and in this case the translation of the conditional sentence should be: “if for them [i.e. the Academicians] there is no common principle [to both kinds of substance]”. The meaning of the whole passage should be, therefore, that, if the Academicians do not admit a common principle, they are obliged to divide their science of principles into two different sciences, physics for the movable substances and another science for the unmovable. This interpretation, in my opinion, is very attractive, because it is coherent with the doxographical character of the whole chapter. But we can observe that, among the Academicians, only Speusippus did not admit common principles for all things, while Plato (and probably Xenocrates too) did. Perhaps Aristotle is thinking that the principles admitted by Plato were insufficient to explain the movable substances, because they were not principles of movement. In any case it seems that the division of the research on substances into two different sciences is considered by Aristotle as a negative consequence which he attributes to the Academicians, but which he wants to avoid. In conclusion I think that Aristotle admits a unique science of substances, which researches the principles of the movable as well as the unmovable substances. This science is neither physics nor the theological science mentioned in Met. E 1 but a “Prinzipienlehre” based not on a logical method, as the “Prinzipienlehre” of Plato, but on a physical, i.e. a causal, method. The whole book Lambda is the exposition of this science, therefore it is not divided into two parts, a physical (cc. 2–5) and a theological one (cc. 6–10), as many interpreters (Jaeger, Aubenque and others) suppose. This science studies before, i.e. in chapters 2–5, the principles of sensible substances (both corruptible and eternal), identifying them with matter and form, i.e. the material and the formal cause, which are the same for all things only by analogy, and after, in chapters 6–7, it establishes the existence of an unmovable substance, conceiving it as the first efficient cause of both the eternal and the corruptible substances. In ch. 8 Aristotle determines the number of unmovable substances, showing that there is a first unmoved mover, which in some unspecified way is the principle of the others and is one not by analogy, but in number. In ch. 9 he determines the nature of this principle as thinking of thinking. The final cause, in my interpretation, is mentioned only in chapter 10, where Aristotle speaks of the good (the second, not the first good) as the order of the universe, following which each thing has its own end. If this is true, chapter 1 states the program of the whole book Lambda, a program which opposes the Academic, or more precisely the Platonic,

The Program of Metaphysics Lambda (chapter 1)

85

“Prinzipienlehre” – following which the principles are the same in number for all things (the One and the indefinite Dyad), but which does not explain their movement, abandoning the latter to physics – and presents a new, properly Aristotelian, “Prinzipienlehre”, based on the distinction of the four kinds of causes, according to which the material, the formal and the final causes, as emerges from the rest of the book, are the same only by analogy, while the first efficient cause is the same in number for all things. Met. Alpha elatton might be an introduction to this research. It would be interesting to test the hypothesis of Jaeger, following which also the book N, which contains the more extended criticism to the Academic “Prinzipienlehre” and presents many affinities with Lambda, could belong to the same research, that in this way should be the true “Ur-metaphysik” of Aristotle.

Bibliography Alessandro di Afrodisia, Commentario alla Metafisica di Aristotele, a cura di G. Movia, Milano 2007. Aristotele, La metafisica, a cura di C. A. Viano, Torino 1974. Berti, E.: La fonction de Metaph. Alpha elatton dans la philosophie d’Aristote, in: P. Moraux/J. Wiesner (eds.), Zweifelhaftes im Corpus aristotelicum. Studien zu einigen Dubia. Akten des 9. Symposium Aristotelicum, Berlin–New York 1983, 260–294. Berti, E.: Il libro Lambda della Metafisica di Aristotele: Tra fisica e metafisica, in: G. Damschen/R. Enskat/A. Vigo (eds.), Plato und Aristoteles – sub ratione veritatis. Festschrift für Wolfgang Wieland zum 70. Geburtstag, Göttingen 2003, 177–193. Berti, E.: La critica di Aristotele alla scienza universale in Metaph. A 9, in: L. Cardullo (ed.), Il libro Alpha della Metafisica di Aristotele tra storiografia e teoria, Catania 2009, 177–193. Burnyeat, M. F.: A Map of Metaphysics Zeta, Pittsburg 2001. Devereux, D.: The Relationship between Theophrastus’ Metaphysics and Aristotle’s Metaphysics Lambda, in: W. Fortenbaugh/R. W. Sharples (eds.), Theophrastean Studies, New Brunswik–Oxford 1988, 167–188. Donini, P.: Il libro Lambda della Metafisica e la nascita della filosofia prima, in: Rivista di storia della filosofia, 57 2002, 181–199. Elders, L. (ed.): Aristotle’s Theology, A Commentary on Book Λ of Metaphysics, Assen 1972. Fazzo, S.: L’esordio del libro Lambda della Metafisica, in: Rivista di filosofia neoscolastica, 110 2008, 159–181. Frede, M.: Metaphysics Λ 1, in: M. Frede/D. Charles (eds.), Aristotle’s Metaphysics Lambda. Symposium Aristotelicum, Oxford 2000, 53–80.

86

Enrico Berti

Freudenthal, J./Fränkel, S.: Die durch Averroes erhaltenen Fragmente Alexanders zur Metaphysik des Aristoteles: Berlin 1885. Ibn Ruschd’s Metaphysics, A Translation with Introduction of Ibn Ruschd’s Commentary on Aristotle’s Metaphysics, Book Lam, Ch. Genequand, Leiden 1986. Kahn, C. H.: The Place of the Prime Mover in Aristotle’s Teleology, in: A. Gotthelf (ed.), Aristotle on Nature and Living Things, Bristol–Pittsburgh 1985, 183–205. Lefebvre, D.: La question de l’unité d’une science des substances: Interprétations de Métaphysique, Λ, 1, 1969A36–B2, in: M. Bonelli (ed.), Physique et métaphysique chez Aristote, Paris 2012, 113–132. Manison, A.: Philosophie première, philosophie seconde et métaphysique chez Aristote, in: Revue philosophique de Louvain, 56 1958, 165–221. Themistius: Paraphrase de la Métaphysique d’Aristote (livre Lambda) traduit de l’hébreu et de l’arabe, introduction, notes et indices par R. Brague, Paris 1999. Thomae Aquinatis: In duodecim libros Metaphysicorum Aristotelis expositio. Ed. R. M. Spiazzi, Turin–Rome 1964. Salis, R.: Il commento di pseudo-Alessandro al libro Λ della Metafisica di Aristotele, Soveria Mannelli 2005.

The Principles of Sensible Substance in Metaphysics Λ 2–5 CHRISTOF RAPP

In the first lines of book Lambda of the Metaphysics, Aristotle announces that he wants to seek the principles and causes of ousia. After the division of three types of ousia in chapter 1, the following chapters immediately start inquiring into the principles and causes of sensible ousia, and since being a sensible ousia amounts to being a changeable ousia, the subsequent investigation is about changeable ousia and its peculiar principles and causes. The final sentence of chapter 5 of book Lambda concludes by saying that it has now been stated (i) what the principles of sensible things are, (ii) how many they are and (iii) in what sense they are the same and in what sense different (1071b1–2). The corresponding attempt to determine the number of principles of sensible and changeable ousia (corresponding to questions (ii) and (iii)) is the topic of the following paper. For this purpose, I will primarily draw on chapters 4 and 5, in the course of which Aristotle undertakes several attempts to distinguish senses in which some of the identified principles are the same and senses in which they are not. The first sentence of chapter Λ 4 makes an important statement that will be unfolded and differentiated throughout the following two chapters. This is the claim that, in a sense, the causes and principles are different for different things, but that in another sense, in particular if one speaks universally or analogically, they are the same (1070b31–3). This statement is obviously meant to open the discussion of question (iii) as mentioned at the end of chapter Λ 5, namely the question in what sense the principles are the same and in what sense they are not. It therefore presupposes that it is already clear by this point what types or kinds of principles we are speaking of. This refers us back to the discussion of chapters 2 and 3, which are actually meant to determine the kind or type of principles of changeable substance. For this reason, we have to start off by discussing selected passages from chapters Λ 2 and 3.

88

Christof Rapp

I. What does it mean to be the principle of a changeable ousia? At the beginning of chapter Λ 2, the discussion of principles of ousia zooms in on sensible ousia. This is clearly meant as a reference back to the trichotomy from chapter 1, where Aristotle had distinguished between two types of sensible ousia on the one side – one of them eternal and one perishable – and one type of immovable ousia on the other (Λ 1.1069a30–6). This trichotomy includes a remarkable anacoluthon, as it does not juxtapose sensible and non-sensible or movable and non-movable, but sensible and immovable ousia. This might be taken as a first indication of the close connection that Aristotle seems to see between a substance being sensible and a substance being movable or changeable. There is, e.g., no attempt in this context to argue for the exclusion of the possibility of sensible, but immovable substances.1 The subject matter of the following chapters Λ 2–5 is, hence, confined to the principles of sensible ousia and, right from the beginning of chapter Λ 2, sensible ousia is taken to be changeable (μεταβλητή). Famously, chapter Λ 6 marks the transition to immovable ousia and the discussion of the unmoved movers, which are taken to be of this latter sort of ousia. Here, at the transition from the discussion of movable and changeable to the discussion of immovable ousiai, Aristotle recalls that he had originally distinguished three types of ousia, but now summarizes the two sensible types of ousia as the two ‘natural’ ones (αἱ φυσικαί: 1071b3) and opposes them to the immovable type of ousia that is going to be the topic of the remaining chapters 6–9. Similarly, at the end of Λ 1, Aristotle emphasizes that the two former types of ousia, the two sensible ones, are the subject of the study of nature, as they are connected with movement (μετὰ κινήσεως γάρ: 1069b1). The intimate connection between being changeable and being ‘natural’ or being ‘the subject of the study of nature’ does not come as a surprise for the reader of Aristotle’s Physics. Even before the second book of the Physics reveals that nature (φύσις) is to be understood as an inherent principle of change and rest, it is taken for granted that the study of nature has to be concerned with the explanation of change and motion and, correspondingly, with the explanation of changeable and moveable things. Already in Physics I 2, for example, Aristotle is quite explicit about his position that immovable beings, as they are advocated most notably by Eleatic and

1

Charles, D. 2000: Metaphysics Λ 2: Matter and Change, in: M. Frede/D. Charles (eds.), Aristotle’s Metaphysics Lambda. Symposium Aristotelicum, Oxford, 83, wonders about Aristotle’s silence on this point and tries to provide the missing premises: Sensible substances possess matter, and what possesses matter must be subject to change, i.e. one or the other sort of change.

The Principles of Sensible Substance in Metaphysics Λ 2–5

89

Academic philosophers, cannot fall within the ambit of the study of nature (184b25–7). Applying all this to the discussion of chapter Λ 2 it is fair to say that, although book Λ is dealing with substance and not, as the Physics, with motion and change, the inquiry into the principles of sensible, natural and, hence, changeable ousia, must engage in the study of motion and change to the extent that changeable ousiai themselves are essentially characterized by the fact that they are subject to (at least some kind) of change.2 This also seems to be the reason why Aristotle feels entitled to derive the principles of changeable ousiai from the analysis of change, as he does in the first section of Λ 2 (1069b3–9), for here he smoothly moves from the assertion that sensible ousia is changeable (μεταβλητή) to stating the principles of change (μεταβολή). This proximity between the genuinely physical approach and the metaphysical sub-project of Λ 2–5 might raise the concern of how, after all, both projects can be kept separate – given that in many other contexts Aristotle insists on the difference between second and first philosophy, between physics and the project of the Metaphysics. Two quick responses to this concern are available: Firstly, it is nowhere clearer than in book Λ of the Metaphysics that, for the first-philosophical project as a whole, the inquiry into sensible substance is not the whole story, but must be extended to an inquiry into immovable substance, whereas physics or the study of nature is necessarily restricted to natural and, hence, changeable beings. Secondly, Aristotle’s methodological approach to identifying the principles of changeable ousia by way of analysing the ingredients of change is certainly not meant to deny the following difference: It is one thing to ask how (and in virtue of which principles) certain sensible substances come to be and cease to exist or how (and in virtue of which principles) all sensible substances manage to undergo change and to change their position, but it might be quite another thing to ask what the principles and causes of the being of the same substances are, once it has come into being.3 Even so, it might turn out that

2

3

In addition, the kind of sensible substances Aristotle seems to be most interested in are not only subject to change, but are crucially characterized by the fact that they come to be and cease to exist. Of course, this is only true of the sub-class of perishable substances and not of the second sub-class of sensible-but-eternal substances; the peculiar features of this second sub-class of ousia, which is inhabited by eternally moving celestial bodies, is not a major concern of chapters Λ 1–5: Only when the discussion eventually turns to different types of matter that correspond to different kinds of change (1069b24–6), does Aristotle mention the peculiar type of matter that is required for ousiai that are capable of moving, but not of coming to be or ceasing to exist. In Physics I 7.191a19–22, Aristotle points to a division of labour concerning this matter: Here he says that, although the number of principles (of change and generation, or of changeable, natural substances) has been established, it might not be clear what substance is, thus apparently leaving the discussion of ousia to another discipline.

90

Christof Rapp

the being of sensible and changeable substance might be due to the same or the same sort of principles and causes through which these sensible and changeable substances have come into being in the first place. In other words: Although it might be peculiar of the first-philosophical inquiry into the principles of changeable substance that it wishes to reveal the principles and causes of the being (and the unity) of such a substance, it might turn out that, these principles, being the principles of changeable substances, are the same as the principles of change that these substances qua being changeable typically undergo.4

II. Importing the principles from Physics I (chapter Λ 2) From the beginning of Λ 2 onward, Aristotle tackles the task of determining the principles of sensible and changeable ousia, which he intends to do by stating how exactly change (μεταβολή) can come about. Here is the very condensed passage that provides the essential ingredients of such an account: “Now if change proceeds from opposites or from intermediate points, and not from all opposites (for the voice is not-white), but from the contrary, there must be something underlying which changes into the contrary state; for the contraries do not change. Further, something persists, but the contrary does not persist; there is, then, some third thing besides the contraries, viz. the matter.” (1069b3–9, Revised Oxford Translation) Obviously, Aristotle takes as his starting-point a model that echoes a discussion from Physics I; in fact, it even seems that Λ 2 and, in particular, the passage quoted above provide something like a summary of the main results of the first book of the Physics. Since we speak of sensible ousia, and being sensible, as we said, amounts to being changeable (1069b3), Aristotle immediately refers to the structural analysis of change (μεταβολή)5 according to

4

5

A similar shift from the principles of change to the principles of changeable things can also be found in Physics I: Although in some passages Aristotle seems to analyse generation and its principles as such, some of the more resultative passages clearly speak of the causes and principles of the natural beings and the principles from which they have come into being, etc. (see Phys. I 7.190b17–9). In the corresponding chapters of Physics I Aristotle speaks of γένεσις (generation) and γίγνεσθαι (coming to be) and not of change. (At the beginning of chapter 7, for example, he uses γένεσις in its broadest sense, in order to cover both substantial generation and nonsubstantial change; even though he distinguishes in the course of this chapter between the substantial and the non-substantial case.) Presumably, this terminological difference is due

The Principles of Sensible Substance in Metaphysics Λ 2–5

91

which each occurrence of change requires, firstly, a pair of contraries between which the change takes place and, secondly, some third thing that underlies the change; and this third something is identified with matter. This quick result is achieved after only six or seven lines. After the discussion in the main chunk of chapter Λ 2, which is mainly concerned with different types of change and the idea of different types of matter corresponding to those different types of change, Aristotle confirms unequivocally that the quick result from the beginning of the chapter was actually meant to contribute to what he was looking for, i.e. the principles and causes of sensible and changeable ousia, for Aristotle concludes the chapter by saying: “The causes and the principles, then, are three, two being the pair of contraries of which one is formula and form and the other is privation (λόγος καὶ εἶδος τὸ δὲ στέρησις), and the third being the matter.” (1069b32–4, Revised Oxford Translation) We are thus provided with a triple of principles or types of principles. This is, indeed, a painlessly reached result; as we will see, it will not remain the only one. So for the purpose of further reference, let us start ascribing numbers to the different results: Res 1: The causes and principles of sensible ousia are three: privation, form & matter. The introduction of Res 1 deserves a full commentary6, but in the present context we have to confine ourselves to the following, rather sketchy, remarks: In spite of several obscurities concerning the details, it is clear that passage 1069b3–9 straightforwardly aims at the introduction of a triple of principles: the two contraries and the underlying something or matter. As for the two contraries, Aristotle does not pause here to argue that they are interrelated as the possession and the loss or lack of a quality or form (as he does in Physics I 5, in particular in the sections following 188b8); it is only in the summary in passage 1069b32–4 that he analyses the contraries into a form-like and a privation-like ingredient. This somehow unwarranted transition seems to confirm the impression that Aristotle consciously introduces the form-privation-substratum-scheme just as a reminder of a theory that he has argued for somewhere else, e.g. in Physics I. Just as in Physics I 5, which, in general, tries to make a case for the contraries as principles, Aristotle is careful to include the possibility of a change that comes from or leads to an

6

to the fact that in Metaphysics Λ 1–5 he wishes to include the sensible eternal substances and that it would be odd to imply that they could come to be. Such as e.g. the commentary given by Charles (see note 1).

92

Christof Rapp

intermediate state on a contrary scale.7 In order to narrow down the kind of opposites relevant to the contraries, he gives in our passage the example of the not-white, which, at first glance, only seems to exclude contradictions from the realm of relevant opposites. However, Aristotle comments on this example by saying that voice is not-white. Certainly, voice is an instance of a being that falls into the complementary class of not-white-beings. But voice is also an example of a subject that is not at all capable of becoming either white or not-white. One might therefore wonder whether the example of the voice that happens to be not-white is not only meant as an example of contradictory opposites, but rather as a very much abbreviated hint at the rule that change neither comes from something at random nor results into something at random, but is only between contrary qualities and their intermediates. In a similar context, to be sure, the example of not-white is used in Physics I 5: “For how could white come from musical, unless musical happened to be an attribute of the not-white or of the black? On the contrary, white comes from not-white – and this not from any not-white, but from black or some intermediate” (188a35–b1). Apart from the contraries, Aristotle has to establish the need for something underlying. In our passage 1069b3–9 the corresponding argument is again very much abbreviated and it is tempting, once again, to read it as a summary or reminder of an argument given elsewhere. Indeed, Physics I 6 gives certain preliminary arguments for thinking that the principles are more than two; perhaps the most promising is that contraries do not directly act on each other, but on some third thing, e.g. rarity does not directly alter density, but both of them make some third thing to be either dense or rare (189a22–6). In the Physics the case for the third underlying something is not settled before chapter I 7, where Aristotle is most keen to emphasize that there is something underlying which is of a double nature: in a way it is something remaining that undergoes the change (which is finally identified as matter), but in another way it is characterized by the absence of the quality that will be acquired in the course of the change. Our passage in Λ 2 could also be read along these lines: In line b6 Aristotle first concludes that there must be something underlying that changes into the contrary state; by way of a reason, he adds, that the contraries themselves do not undergo the change. He probably means that it is not being-white that undergoes the change to being-black (possibly related to Phys. 189a22–6). In lines b7–8 Aristotle points out that on the one hand there is something that persists, while on the other hand none of the contraries persists8 (being-white does 7 8

See e.g. Phys. I 5.188b6 and 9. Phys. I 7.190a17–9 puts it in almost the same words: “One part survives, the other does not: what is not an opposite survives …”

The Principles of Sensible Substance in Metaphysics Λ 2–5

93

not persist if something starting from being white comes to be black). Once it is excluded that it is one of the contraries that persists, Aristotle is in a position to conclude that there must be some third thing besides the two contraries, i.e. the matter. In the light of Physics I 7 this could be taken to mean that we can understand matter as a genuine principle of change or changeable substance, once we have differentiated between what persists and what does not persist and once we have subtracted the non-persisting, privative dimension of what changes or comes to be. Finally, a word on privation (στέρησις). Saying that matter and form are the principles from which a sensible compound substance comes to be and out of which it is constituted sounds sufficiently familiar to the reader of Aristotle’s physical and metaphysical treatises. Much less familiar is the remark that privation too is such a principle of sensible substance, especially if we do not confine ourselves to the principles of change and generation, but speak about the principles of changeable beings – since, after all, privation seems to import a sense of non-being, as it is designed to capture what something is not. Indeed, one might wonder how it is that privation, referring to non-being, could ever be a principle of ousia, i.e. of primary being. The answer to this puzzle seems to be connected with the thought that privation is, in a sense, not an independent principle, but always comes together with the corresponding possession of a form or with the material substratum by signifying a state the substratum is in. And this again seems to be the philosophical background for why Aristotle, after having introduced the two contraries and the substratum in Physics I keeps wavering between saying that there are two and saying that there are three principles. Arithmetically speaking, it does not provide major difficulties to see that two contraries plus one substratum make a total of three principles. However, it seems to be due to the peculiar nature of privation that Aristotle is reluctant to straightforwardly fix the number of principles as three. For on the one hand it is not the privation of a form as such that accounts for coming to be and change, but the alternation between lack and possession, or the absence and presence of a form or quality that does the explanatory job. And one might be inclined to count this entire alternating mechanism as only one principle that in addition to the substratum explains the coming to be and the being of changeable ousia (and if one had to decide whether to take form or privation as the primary ingredient of this mechanism, one would certainly choose form, and not privation, as it is the presence and absence of form that is apt to explain what happens in the course of change and generation).9 On the other hand 9

This is more or less what Aristotle himself says in Phys. I 7.191a5–7: “Yet in another way of putting it this [that the contraries are two, scil.] is not necessary, as one of the contraries will serve to effect the change by its successive absence and presence.”

94

Christof Rapp

privation is also dependent on the substratum that undergoes the change or the coming to be; for, if we say that something comes to be from its privation, we do not mean that it comes to be from sheer non-being (this is the point that, according to Aristotle’s analysis in Physics I, was one of the main sources of the confusion for earlier thinkers), but that, for example, the musical man comes to be from the unmusical man and the bronze sphere from the unshaped bronze – i.e. from a state (in fact, the privative state) in which the substratum happened to be before it undergoes the change. The substratum, being one in number, turns out be two in form or two in being:10 namely the persisting material substratum and the privative state that is contrary to the state into which the substratum is changed. From this perspective, the privation is one mode of being of the complex substratum (i.e. the unmusical man as opposed to man or the unshaped bronze as opposed to mere bronze) alongside the substratum in the material sense.

III. Adding the external principles and causes (chapter Λ 3) The beginning of chapter Λ 3 takes up two of the three principles just mentioned, matter and form, and argues that neither matter nor form comes to be (1069b35–6). Again, Aristotle refers to something like a general structural analysis of change: Whatever changes, he says, changes as something into something (εἴς τι) under the influence (ὑπό τινος) of something (1069b36– 1070a1). That which is changed is matter; that into which it changes is form; hence, matter and form are not themselves generated.11 As in chapter Λ 2, Aristotle introduces these ideas with little argumentative effort, and again it seems that he is relying on discussions to be found elsewhere in his œuvre. In this particular case, chapters Z 7–9 of the Metaphysics suggest themselves as a parallel text that could provide the background that seems to be missing in the present context.12 The newly introduced matrix of change – i.e. that there is always something that is changed, something else into which it is changed, and a third thing by which or under the influence of which things are changed – introdu-

10 11 12

Compare for a similar formulation in Phys. I: 190a15–7. On the relation of this thought to aporia #8 from Metaphysics B 4, 999b6–14 see Menn, forthcoming. It is with good reason, hence, that Judson, L 2000: Formlessness and the Priority of Form: Metaphysics Zeta 7–9 and Lambda 3, in: M. Frede/D. Charles (eds.), Aristotle’s Metaphysics Lambda. Symposium Aristotelicum, Oxford, structures his comments on Metaphysics Λ 3 as a comparison with chapters Z 7–9.

The Principles of Sensible Substance in Metaphysics Λ 2–5

95

ces a new player in addition to the three stated in Res 1, and Aristotle hastens to specify this new player as the first mover (1070a1). If we bracket, for the time being, the question of why Aristotle picks out the first one and whether ‘first’ is either the first in a series of movers or only the proximate or immediate, then it seems clear that, for reasons that are closely connected with the conditions of generation and change in the realm of sensible ousia, the mover must be rated among the causes and principles of sensible ousia. This, at least, is something that chapter Λ 4, beyond any doubt, will take for granted, so that we can infer: Res 2: The mover is also among the principles of sensible ousia. It therefore seems, at least for the time being, that we have to reckon with a quadruple rather than a triple of principles of sensible ousia. In addition to the introduction of the mover, chapter Λ 3 provides a couple of interesting claims, the precise role of which within the project of identifying the principles of sensible ousia is not entirely clear from the outset, but seems connected with the introduction of the moving cause in one way or another. Above all, it seems, there are two ideas in chapter Λ 3 that will become crucial for the overall account of principles of sensible ousia, namely the synonymy principle and the distinction between pre-existing and simultaneously existing causes. The discussion of the synonymy principle immediately follows the debate about the impossibility of generation of matter and form. Its connection to this previous discussion is anything but clear; the new topic is introduced by a simple “μετὰ ταῦτα” (1070a4). Each ousia, Aristotle states, is generated from something synonymous, i.e. something that shares its name (1070a4–5). Res 3: Each ousia is generated from something synonymous. This rule can be proved for both natural and artificial beings; and, since luck and spontaneity are nothing but privations of nature and art (and hence do not constitute independent cases), the synonymy principle can be stated as a general and exhaustive rule. Still, it applies to natural and artificial things in different ways. Art, Aristotle claims, is a principle of movement in something other than the thing moved, while nature is a principle in the thing itself (1070a7–8). The notions of “in something other” (ἐν ἄλλῳ) and “in the thing itself” (ἐν αὐτῷ) are somehow awkward, given that Aristotle illustrates the latter, natural, case with the example of ‘man begets man’. Naturally, this is not meant to say that the begetting man is in the man that is to be generated; it rather says that there is something of the same kind or something with the same name that serves as a principle and cause for the coming-to-be of a man. In the case of artificial beings, by contrast, the synonymous principle must be thought to be in something else, i.e. in a sub-

96

Christof Rapp

stance of another kind. If one takes, for example, health to be among the causes of the artificially produced health of a particular being, it is the health in the soul of the doctor that is the synonymous principle or cause of the artificially brought about instance of health in a particular patient. In both the natural and the artificial case the synonymous principle seems to be connected with a preceding cause, i.e. a cause that pre-exists the generated thing – in the one case the father, by virtue of transmitting his biological form of a human being is such a preceding and synonymous cause, while in the other case the doctor (who is not himself synonymous with the brought about health), by virtue of imposing the (synonymous) form of health on the bodily condition of the patient plays an analogous role.13 In one of the subsequent passages of Λ 3, the moving cause is presented as the paradigmatic external sort of principle. Seen from this perspective, the introduction of the synonymy principle could be motivated by the appearance of the moving cause. It might also be seen as a preparatory step for what follows: In identifying and counting the principles of sensible substances, the synonymy principle might help to sort out the variety of principles and even to reduce the number of pertinent principles. There is at least one more idea that we find in chapter 3 and that will become pertinent for the discussion of chapters 4–5; that is the distinction between pre-existing and simultaneously existing causes. More precisely, Aristotle classifies moving causes as pre-existing, and mentions causes in the sense of λόγος as an example of the simultaneously existing cause (1070a21–2): Res 4: The moving cause exists as preceding the effects, while other causes (like the λόγος) exist simultaneous with their effects. Examples of causes existing simultaneously with their effects are: the health that exists whenever and as long as a particular man is healthy, and the shape of a bronze sphere that exists at the same time as the particular bronze sphere. Obviously, Aristotle thinks exclusively of what he calls elsewhere the formal cause or just the λόγος (1070a29) or the ousia in the sense of λόγος. In another sense, matter would also be a candidate for a principle or cause that exists simultaneously with the particular ousia it is a principle or cause of; however, matter as a principle from which an ousia is generated might be an ambivalent case, because some portion of matter must have pre-existed

13

For the analysis of this latter case see also the parallel passage in Z 7.1032b11–4: “Therefore it follows that in a sense health comes from health and house from house, that with matter from that without matter, for the medical art and the building art are the form of health and of the house.” In Λ 4, Aristotle will straightforwardly identify the moving cause with the medical art (see 1070b28).

The Principles of Sensible Substance in Metaphysics Λ 2–5

97

the compound ousia, and this ambiguity might be the reason why Aristotle is reluctant to use it as an exemplary case for simultaneously existing principles. At any rate, Res 4 seems to present itself as an exhaustive distinction that allows an unambiguous classification of all possible candidates. In order to work that intuition out, one should perhaps add that the external causes as such not only have to pre-exist the compound ousia, but also will not enter the generated entity as a sort of ingredient (as is true of matter). A similar distinction will be introduced in chapter 4 below: In 1070b22– 5 Aristotle distinguishes between principles that are present in a thing and principles that are external to it; only the former ones can be called ‘elements’, whereas the latter ones are illustrated by moving causes. Res 4a: Beside the element-like, inherent principles, there are other, external principles, in particular, the moving causes. Both results, Res 4 and Res 4a, show that for Aristotle the introduction of the moving cause is not just the casual addition of another type of principle; rather, the moving cause requires us to broaden our understanding of how an ousia can be related to its principles and causes. In Λ 2 and in the corresponding chapters of Physics I, moving causes were not mentioned at all, which might be due to their being external and not internal or constituting principles. Indeed, one might be inclined to expect that the inquiry into the principles and causes of an ousia requires an answer that restricts itself to constituting, inherent principles. As soon as we start, however, to broaden our perspective just a little and begin to regard the particular sensible substance as part of a system of natural changes and generation, we have to acknowledge that a causal-explanatory profile of substances that excludes preceding causes would be incomplete.

IV. Principles are in some sense the same (chapter Λ 4) Chapter 4 opens with a quite general statement about the sameness of principles and causes: “The causes and the principles are different for different things (ἄλλα ἄλλων) in a sense (ὥς), but in another sense, namely if one speaks universally and analogically (καθόλου … καὶ κατ᾽ ἀναλογίαν), they are the same.” (1070a31–3) It is relatively clear by now what kinds of principles and causes Aristotle is speaking of. Naturally, we think of the triple of causes mentioned in Res 1, and the quadruple of causes he envisages, once we consider Res 1 and Res 2

98

Christof Rapp

together. We have to take into account that, at least, some causes are synonymous with their effects and that, in a way, each ousia comes from something synonymous; that of the synonymous principles and causes, some behave as in the case of natural generation, while others behave just like in the artificial case, that according to Res 4 some causes pre-exist their effects, while others exist simultaneously with them; and finally that according to Res 4a only the inherent, simultaneously existing principles are elements, whereas the preceding, external ones are not. Obviously, this opening statement is only the first move and has to be fleshed out in what follows. Before Aristotle starts to unfold this statement, he first raises a difficulty or aporia, the discussion of which occupies the first third of chapter Λ 4, i.e. lines 1070a33 to b10. The difficulty is said to consist in the question of whether a substance and a relative (or other items of non-substantial categories) could have the same principles. Apparently, the question is introduced at this stage of the discussion in order to comment on the formulation that, in a sense, all things can have the same principles. The wording, in which this aporia is given is as follows: “For one might raise the question whether the principles and elements are different or the same for substances and for relatives, and similarly in the case of each of the categories.” (1070a33–5, Revised Oxford Translation) Aristotle speaks of ‘principles and elements’, but in the following lines he will actually restrict the meaning of ‘principle’ to element-like principles. This observation is in line with the fact that, at the end of the aporia, he concludes the discussion by saying that not all things can have the same elements (1070b8–10). In the course of the discussion of this aporia Aristotle gives two arguments why substances and relatives or entities from other non-substantial categories cannot have the same elements. Although the overall direction of these two arguments seems to be clear, they are quite difficult to reconstruct in detail.14 The first argument says that if they had the same elements, they would have to be ἐκ τῶν αὐτῶν (1070a36): either from the same things or consisting of the same things – but what would these same things be? The common element of substance and the other categories, I take it, which, as an element, would have to be prior and, hence, distinct from what it is the element of, must be some category, some general entity over and above sub-

14

For an in-depth analysis of this passage, see Crubellier, M. 2000: Metaphysics Λ 4, in: M. Frede/D. Charles (eds.), Aristotle’s Metaphysics Lambda. Symposium Aristotelicum, Oxford, 145–8.

The Principles of Sensible Substance in Metaphysics Λ 2–5

99

stance and non-substantial categories. But, to be sure, there is nothing like this, nothing common, no κοινόν, besides the substance and the other categories. The only remaining alternative would be that substance is an element of the relative or the relative of the substance, but this does not even seem to be worth considering. The second argument (1070b4–9) deals with material as well as with intelligible elements; it seems to be struggling with one of the presuppositions that was made in the first objection, namely the assumed difference between elements and what they are the elements of. That material elements, like e.g. the letters a and b, are different from what is made of them, like the syllabletoken ba, might be conceded, but, one might object, this may work differently in the case of intelligible elements. And the two intelligible elements that are actually under consideration, i.e. ‘being’ and ‘one’, τὸ ὂν and τὸ ἕν, might be seen as candidates for the status of elements of everything, since they apply to everything that is and that exists as one something. Unfortunately, Aristotle’s reaction to this possibility is quite elliptical. One possible suggestion would be the following: It seems that, for Aristotle, τὸ ὂν and τὸ ἕν cannot be proper elements just because each and every compound can be said to be ὂν or ἕν and because it is the whole compound that is ὂν and ἕν, whereas genuine elements are different from the whole. The aporetic passage ends with the statement that not all things can have the same elements. Whether this result should be stated as a positive and independent result or whether it is only meant as an upshot of the previous aporia is difficult to tell. A preliminary reason for thinking that this result might not be Aristotle’s last word on the issue of substances and their relation to non-substantial beings is given by Res 11 in chapter Λ 5, since there he will argue for the opposite conclusion that the causes of all things are the same, in that substances, or strictly speaking the causes of substances (Res 11*), are the causes of everything else. Since Aristotle marks the transition from the aporetic passage to the unpacking of his own account by the remark “or rather, as we say” (1070b10), we should be wary of taking the result of the aporia as part of Aristotle’s own position on the sameness of principles. Whereas the aporia concludes with the upshot that not all things can have the same elements, Aristotle starts unfolding his own account with his differentiating claim that they are in one sense the same, but in another sense not the same, thus invoking the opening statement of chapter Λ 4, which seems to serve as a program for what follows, i.e. the program of distinguishing between senses in which principles can be taken to be the same and senses in which they cannot. – Still, one could wonder which of the concerns that are expressed in the course of the aporia would have been shared by Aristotle. It is likely, to begin with, that Aristotle would use such an opportunity to make a case against the intelligible elements, τὸ ὂν and τὸ ἕν. It would also

100

Christof Rapp

be plausible, from his point of view, that substances and non-substances cannot have the same principles alike, as the priority of substance and the posteriority of a non-substance must be reflected in different relations to their common principles; for example, the direct or proximate principles of a substance can only be indirect or remote principles of the dependent non-substance. Finally, it is clear that in the course of the aporia Aristotle tends to restrict the question of the sameness of principles somehow artificially to the sameness of element-like principles, so that even if it is the case that the undifferentiated result from 1070b9–10 – i.e. that not all things can have literally the same elements – was not unwarranted, this result would not hold for all kinds of principles, but only for the element-like principles. Let us move on now to the explicit articulation of Aristotle’s own point of view. Having repeated in 1070b10–1 the opening statement from Λ 4, that in a sense the principles are, while in another sense they are not the same, Aristotle starts illustrating his general point by a few examples: Take the element-like principles of (certain) perceptible bodies and say that they are hot, cold (the one is form, the other privation), plus a sort of matter that can take on these attributes, then it is true that all these things have the same elements or principles, namely hot, cold and the respective matter, although the elements and principles of different things are different; so this is one way to show that for some things the elements and principles are the same, but not for all. Ex 1: hot, cold

+ appropriate matter for (certain) perceptible bodies

The example chosen has troubled commentators a lot. The problem is that we would expect the text to give examples of ousia, since the elements/principles of ousiai have themselves to be ousiai in a way, while hot and cold do not obviously meet this criterion. However, the example seems to be deliberately chosen with a view to the triple of principles from Res 1, as hot and cold are more apt to exemplify the type of physical change that occurs between a quality/form and its privation. Another problem with this example is that it does not make sufficiently clear how broad or narrow the realm in which those principles are the same must be understood. At least, Aristotle does not seem to be eager to restrict this realm to something like the standard notion of a species; in light of the examples that will immediately follow, we cannot be entirely wrong if we think of the generality of something like genera. Many problems concerning Ex 1 will remain, but the tentative result of this example seems to be the following: Res 5: There is a sense in which principles/elements are the same for things taken from the same genus/realm, i.e. in that they have the same type of form, privation and matter as principles (where ‘ge-

The Principles of Sensible Substance in Metaphysics Λ 2–5

101

nus/realm’ and ‘type’ are to be understood in accordance with Ex 1). The sense in which principles/elements are the same, as described in Res 5, must be seen against the background that in some other sense they are not the same. Indeed, Aristotle says: “These things then have the same elements and principles, but different things have different elements; and if we put the matter thus, all things have not the same elements, but analogically they have, i.e. one might say that there are three principles – the form, the privation, and the matter. But each of these is different for each class; e.g. in colour they are white, black, and surface. Again, there is light, darkness, and air; and out of these are produced day and night.” (1070b16–21, Revised Oxford Translation) In this quotation we get two further examples in the light of which Ex 1 must be read, and they confirm the suspicion that the frame in which principles can be said to be the same is something like genera: Ex 2: colour: white, black, surface Ex 3: day & night: light, darkness, air These two examples are introduced in order to exemplify the statement that, what form, privation and matter is, will be different in different genera (περὶ ἕκαστον γένος), so that ‘colour’ and ‘day and night’ might be taken as the realm or genus, within which these principles might be the same, while different genera are associated with different principles. Let us wrap up these ideas in the following statement: Res 6: There is a sense in which principles/elements are different, in that form, privation and matter do not mean the same in different genera (where ‘genera’ is to be understood in accordance with Ex 1, Ex 2, and Ex 3). In addition to this, the quoted passage 1070b16–21 equips us with a further sense in which principles can be said to be the same, and this is the sense in which we say – in accordance with Res 1 – e.g. that there are exactly three principles, namely form, privation and matter. For Aristotle, saying that something is one by analogy always implies two quite different aspects: On the one hand, analogy provides us with a genuine principle of collection and unity; it is one of the privileged ordering principles when dealing with transgeneric questions. It is fully legitimate to subsume things from different genera under one notion or heading whenever they are analogically related. On the other hand, using analogical terms Aristotle stresses that they do not amount

102

Christof Rapp

to the stricter forms of unity that are guaranteed, e.g., by species-terms, since what it means to be analogically this or that is bound to be different in different cases. In this sense, the next result that can be derived from the given quotation refers at the same time to a sort of unity and to the peculiar difference that is intrinsically connected with the analogical type of unity: Res 7: The sense in which the previously mentioned principles (i.e. form, privation, matter) are just three is the analogical one. Whenever we collect items under an analogical unity, we must be prepared to explain the respects in which those analogically related items are different in different cases. This also applies to the present case, in which we speak of analogically identical principles and causes. In fact, the different instances of privation and form might look so differently, that it requires a veritable effort to recognize that they are in a sense the same, i.e. through being all instances of privation and form. In Physics I 5, Aristotle actually points out that privation and form are in one sense the same and in another sense different. Due to the almost bewildering variety of contraries that have been suggested by pre-Aristotelian thinkers, people even failed to see, as Aristotle points out in 188b30–189a2, that all these different accounts ultimately converge in using a pair of form-like and privation-like items as principles. For example, it is not obvious that the cold (from which the warm can be derived), the unmusical (from which the musical emerges), the untunedness (which can turn into attunement) and the shapelessness of a pile of bricks and stones (which can be turned into a well-ordered house) are all instances of what Aristotle calls ‘privation’ and are all likewise related to their contraries, which Aristotle would describe as ‘form’. The sense in which all principles can be, according to Res 7, the same, can only be expressed by the notions of privation, form and matter: they are the same, precisely in that they all function in their field as privation, form and matter respectively. In addition to what we said about the analogical character of privation and form, there is a section in Physics I 7, which, after the establishment of the substratum as a third and independent principle of generation, attempts to show that the substratum and, hence, matter too can only be grasped by analogy (191a7–8), for the underlying nature is to substance just as the bronze is to the statue, the wood to the bed, and the matter and anything formless before receiving form to anything which has already received the form. In this respect Res 7 seems to be in full agreement with Physics I about the status and identity of analogically conceived principles, such as privation, matter and form. These are unambiguous and important results; however, we have not yet reached the end of Λ 4. The next move in the argument is gratifyingly consistent with what we know from chapters Λ 2 and 3 and, in particular with the

The Principles of Sensible Substance in Metaphysics Λ 2–5

103

transition from Res 1 to Res 2. Just as these chapters first introduced the triple of privation, form and matter, and only then added the mover as a fourth type of principle, chapter Λ 4 first applies the analogical meaning of sameness to the triple of privation, form and matter, and only then mentions the moving cause. Elaborating on Res 4 and the distinction between simultaneously existing and preceding principles, Aristotle opens the final chunk of Λ 4 by explicitly stating what we have already mentioned as Res 4a. Having introduced the moving cause as a fourth player into the peculiar project of chapter Λ 4, Aristotle now seems prepared to extend the account of analogical concepts he had already given for the triple solution in Res 7 to the quadruple account: Res 8: Analogically speaking, there are three elements (see Res 7) and four principles (see Res 2), and just as the three element-like principles privation, form and matter are different in different genera, so are the external principles. The quadruple scheme is exemplified by two cases, taken from the field of artefacts, i.e. Ex 4 and Ex 5, where medical art is the external mover that has to deal with the triple of health, disease and body, and the building art is the external mover dealing with the specific pair of order and disorder to which the material of building, e.g. bricks is susceptible. Ex 4: Health, disease, body

+ the medical art

Ex 5: Form, (a corresponding) disorder, bricks

+ the building art

A further example, Ex 6, remains a fragment, but is obviously meant to remind us of the peculiar character of the synonymous mover in the generation of natural things, as it was explained in Res 3. Ex 6: (certain natural things …)

+ man15

If we consider the natural, men-begets-men case (in light of the synonymyprinciple introduced in Res 3) and construe the artificial cases in Ex 4 and Ex 5 as implying that the moving cause, e.g., the medical art or the building art, is not terribly different from the form, e.g. health or the form of the house, but rather, on the contrary, “somehow the same”, as Aristotle says (1070b33), we are faced with the phenomenon of coinciding principles, which might affect the question of how many principles there are: Res 9: Two of the principles mentioned (in Res 8) may coincide in species, i.e. form and the moving cause.

15

Admittedly, the transmission of these words in line 1071b31 seems unreliable.

104

Christof Rapp

Drawing on this peculiar coincidence, it can now be said that, in one sense there are three principles, while in another sense there are four: Res 9a: (With the coincidence assumption): There are three principles (analogically speaking): form = moving cause, privation, matter. Res 9b: (Without the coincidence assumption): There are four principles (analogically speaking): form, privation, matter, moving cause. Having accommodated the moving cause into the context of chapter Λ 4, Aristotle surprisingly refers to another moving cause in addition to the moving causes mentioned in Ex 4, Ex 5 and Ex 6. He very abruptly closes the chapter by saying that “besides these there is that which as first of all things moves all things” (τὸ ὡς πρῶτον πάντων κινοῦν πάντα). Res 10: In addition to the principles mentioned so far, there is a principle that moves everything and is itself the first of everything.16 This line raises a bundle of exegetical problems: What is the sense of ‘first’? Why on earth is it introduced in the final sentence of the chapter? Why does Aristotle not further comment on this idea? Some of these questions can be settled in the light of similar and similarly surprising remarks in the subsequent chapter Λ 5. For the time being, we can confine ourselves to saying that here, as opposed to the previous occurrences of supposedly ‘first’ movers (1070a1, 1070b27), the first mover does not seem to be the immediate or proximate one17; for it is striking that in the present passage the mover is not addressed as ‘first mover’, but as the ‘first of everything’; furthermore it is said to move everything, which is certainly not true of the proximate movers we have considered so far. Hence, it is not implausible that Res 10 is meant to prepare the stage for the kind of mover we will encounter in the second half of book Λ, but why is it, then, that Aristotle is so brief about this peculiar cause? Perhaps the simplest explanation would be that in the present context, where Aristotle speaks of moving causes and, in particular, of synonymous moving causes, and where he gives, as it were, a causal profile of sensible ousia he just wants to remind us of the fact, that the list of pertinent principles and causes of sensible ousia is not complete.

16 17

This is, indeed, rather a statement than a result; it might be directed against Speusippus’ refusal to acknowledge one single principle of all levels of reality, see Menn, forthcoming. In particular in 1070b27, where Aristotle wishes to show that the mover too is different in different cases, it seems most likely that the parlance of a ‘first mover’ is meant to pick out the proximate mover; for it is the mover that is close to what it moves that must be different in different cases, not the remoter or even remotest mover.

The Principles of Sensible Substance in Metaphysics Λ 2–5

105

Notwithstanding the obscurities that are connected with the final sentence of chapter Λ 4, it seems possible to derive two corollaries concerning Res 10, which will shed further light on the character of this first something that it said to move everything. First of all, it is remarkable that this particular mover is introduced in contradistinction to the synonymous moving causes that have been mentioned just before. In addition, the synonymy requirement would make little sense for something that is thought to move everything: with which of all things should it be synonymous? Hence, we can infer that … Res 10a: … the first mover from Res 10 is a non-synonymous mover. Furthermore, the proximate movers are – according to Res 8 – different in different cases; this again only makes sense if the movers are specific to a limited realm. The same requirement cannot hold for a mover that is said to move everything. If this mover is one and the same for everything it moves, its sameness is not only analogical. Hence, we have unearthed at least one kind of principle that is the same for all, and not only in an analogical sense: Res 10b: The first mover from Res 10 is not different in different cases. It is the same principle in a numerical (not just analogical) sense. At any rate, the puzzles surrounding the somewhat unrelated introduction of the first mover in Res 10 should not distract us from the significant and unambiguous results that have been reached in the course of this chapter, viz. that the triple of privation, form and matter or the quadruple of privation, form, matter and mover can be taken as the same principles only in an analogical sense.

V. Potential and actual, general and particular principles (chapter Λ 5) Chapter Λ 5 continues the project of the previous chapter, i.e. the project of distinguishing senses in which the principles are said to be the same, and senses in which they are not the same. However, chapter Λ 5 is not an easy read: it “consists of a set of fairly unpolished notes”18, in some passages the Greek is difficult to read, and it is far from clear how the different strands of arguments within this chapter are mutually related. Still, it is possible to iden-

18

Code, A. 2000: Metaphysics Λ 4, in: M. Frede/D. Charles (eds.), Aristotle’s Metaphysics Lambda. Symposium Aristotelicum, Oxford, 161.

106

Christof Rapp

tify the three most conspicuous topics of chapter Λ 5: (1) In the course of this chapter, the argument that the principles and causes of all things are the same, because nothing can exist without substances, gains some prominence. The thesis is stated in the first paragraph of the chapter (1070b36–1071a3), is taken up again in section 1071a24–9 and is finally highlighted in the summary of the chapter (1071a29 –b2). (2) The most original contribution of chapter Λ 5 to the leading question of the sameness of principles presumably lies in the introduction of potentiality and actuality: some of the principles we have encountered so far can be classified as either potential or actual beings and this again yields another sense in which principles can be said to be the same. The corresponding argument is unfolded in section 1071a3–17. (3) In section 1071a17–24, Aristotle points out that general formulas, such as “man begets man” must not mislead us into thinking that there are general principles and causes, e.g. a universal man; it is rather the particular man, such as Peleus, that becomes the cause of another particular man, such as Achilles. (1) The chapter starts off with the claim that some things exist independently and others not, and that ousiai belong to the former class. From this assumption the argument jumps rather quickly to the conclusion that the causes of all things are the same, because the other, non-substantial, entities, such as properties and movements, would not exist without ousiai, substances (1070b36–1071a3). It seems as if the argument wants to show that all nonsubstantial things have the same causes, namely substances, as they are dependent for their existence on them. Res 11: Ousia is the cause of everything else. This is an important claim; it seems to reconsider concerns that were first raised in the aporia at the beginning of Λ 4. Nevertheless, it does not seem to be directly pertinent for the project announced in Λ 1 and 2, i.e. the inquiry into the principles of changeable ousia. It is rather a contribution to the broader project of clarifying the sameness of principles and causes of all beings – substantial and non-substantial ones. Indeed, the final sentence of Λ 5 will conclude by saying that we have now stated what the principles of sensible beings are, how many they are and in what sense they are the same and in what sense not (1071b1–2), without explicitly restricting the field of sensible beings to substances. Similarly, the opening statement of chapter Λ 4 did not include an explicit restriction to substances: the causes and the principles of all beings are in one sense different for different beings and in another sense the same for all. Towards the very end of Λ 5, in lines 1071a29–30, Aristotle speaks quite unconcernedly of an inquiry into the principles and elements of substances, relatives and qualities. It is, hence, possible, that in the course of chapters Λ 2–5 the initial project was implicitly extended to the

The Principles of Sensible Substance in Metaphysics Λ 2–5

107

discussion of causes and principles in the sensible realm of reality – without restricting it to the causes and principles of sensible substance. Still, there is an ambiguity concerning the introduction of Res 11,19 and due to this ambiguity some version of Res 11 may indeed be connected with the causes and principles of substances, for when Aristotle takes up the topic of the dependency of non-substantial beings again – first in section 1071a24– 9 and later in the concluding remarks in lines 1071a34–5, it is presupposed that the principles and causes of ousia are at stake rather than ousia as such: Res 11*: The causes of ousia are the causes of all beings. In a way, this is a better and a more pertinent result than Res 11, for the message of Res 11 that substance is the cause of all (non-substantial) beings does not bring us any further in addressing the question of whether and how the principles are the same: true, it allows us to say that the principles of all non-substantial beings are the same, namely substance; but this would reveal the fact that there are, as a matter of fact, many substances and that all of them might be principles and causes of non-substantial beings. In principle, Aristotle could have added here that the principles of non-substantial beings are in a sense one, because it is also true to say that ‘substance’ is their principle, but in another sense they are not the same, as different non-substantial beings have different substances as principles. In fact, however, Aristotle does not offer any such hint in the first section of Λ 4, and hence it seems more plausible to think that, from the very beginning, he was out for something like Res 11*, since this latter result, as opposed to Res 11, directly refers the reader to some of the previously established results: if the causes of all (other) beings can be traced back to the causes of substances and if the sameness of principles and causes of substances have already been considered (as, e.g., in Λ 4), then the consideration of principles and causes of nonsubstantial beings will not raise new problems concerning their sameness, since concerning the principles and causes of substances (which turn out to be the principles and causes of all other beings too) we have already said in which sense they are the same and in which sense they are not. In the concluding lines 1071a34–5 at the end of the chapter Aristotle clearly indicates that his actual aim throughout has been something like Res 11*: The causes of substances are the causes of all beings, since, if we remove the former, we would remove the latter too. (2) The next major achievement of chapter Λ 5 is a new, additional way to unify the principles of sensible beings. The occurrence of matter in the previ-

19

On this ambiguity and on the significance of this claim see also Code (see note 15), 161–6.

108

Christof Rapp

ous chapters had already provided the opportunity to remark that matter is something in potentiality; now, in chapter Λ 5, the new development is that Aristotle explicitly uses the pair of potentiality and actuality as a unifying heading under which several principles can be collected: Res 12: There is another sense (in addition to the sense mentioned in Res 7 or 8), in which principles are the same in an analogical way: Privation, form and matter are analogically the same in that these principles are mutually related as potential and actual beings. Since the notions of potential and actual being were explicitly introduced as analogical terms in book Metaphysics Θ, it goes without saying that if we, as it were, unify several principles as being either potential or actual, thus classifying all principles as either potential or actual, we must also be prepared to distinguish what being potential and being actual mean in different cases. And Aristotle warns us that these are not only different for different things but also apply in different ways to them. “And in yet another way, analogically identical things are principles, i.e. actuality and potency; but these also are not only different for different things but also apply in different senses to them. For in some cases the same thing exists at one time actually and at another potentially, e.g. wine or flesh or man does so. (And these too fall under the above-named causes. For the form exists actually, if it can exist apart, and so does the complex of form and matter, and the privation, e.g. darkness or the diseased; but the matter exists potentially; for this is that which can become both the actual things.)” (1071a3–11, Revised Oxford Translation) They are ‘different in different cases’, I take it, in that ‘being actual’ applies to the form, the matter form-compound and the privation, while being potential applies to matter. Hence, potentiality and actuality is an analogous way to collect principles, which are themselves analogously constituted. It is noteworthy, that this new sense of how principles can be seen as the same relies on the results of the previous chapters – while the passage just quoted in particular builds on Res 7, in which privation, form and matter (but not yet movers) were presented as analogically unified principles. More troublesome, however, is Aristotle’s remark that these notions do not only mean different things in different cases, but also apply in different ways to them. Aristotle explains what he means in the subsequent, partly enigmatic, passage: “But in another way ‘in actuality’ and ‘in potentiality’ differ in the case of things of which the matter is not the same, of which the form is not

The Principles of Sensible Substance in Metaphysics Λ 2–5

109

the same but different, just as the cause of man is both its elements, fire and earth, as matter, and the distinctive form, and furthermore some other external thing, i.e. its father, and beyond these the sun and its oblique course, which are neither matter, nor form nor privation, nor the same in form but rather are movers.” (1071a11–7, translation taken from Code 2000) Here is a tentative suggestion as to how this passage could be understood:20 One way in which an actual being is related to a potential one, is the relation of my form (actual) to the portion of matter (potential) from which I was generated or of which I consist. Another way, however, in which an actual being is related to a potential one, is how my father (actual) is related to the matter which is a human being (me) potentially. In this latter case it obviously holds that the matter (i.e. the portion of matter that constitutes my father and the matter from which I was generated) is not the same. And in yet another way, when not only the matter, but also the form of the actual being is different from mine, a non-synonymous mover, e.g. the sun, is related to something potential. If something like this is by and large correct, then it seems that the present passage is primarily meant to introduce and to redescribe the moving cause in the new terms of actuality and potentiality, after the previous passage (1071a3–11) had confined itself to the triple of privation, form and matter. This would be a clear parallel to Aristotle’s procedure in chapter Λ 4, where he applied the analogy-reading of the sameness of principles first to the triple of privation, form and matter (resulting in Res 7), and only then to the quadruple of privation, form, matter and mover (resulting in Res 8). The philosophically salient point of this application is that the moving causes are actual beings that are related to some potential being. The unrelated occurrence of the sun in this passage is just as surprising as the occurrence of the first mover that popped up at the end of chapter Λ 4. Indeed, there seem to be philosophical parallels between the two of them: As we tried to highlight by corollary Res 10a, the prime mover in Λ 4 was probably mentioned in order to contrast a non-synonymous type of mover with the previously treated synonymous ones. Similarly, in our passage, the sun is explicitly juxtaposed to the father-case on account of its being a mover with a different form. Finally and analogically to corollary Res 10b, if the sun could be said to be a cause of all natural beings in the sublunary world, then it would be another instance of a principle that is the same in a literal (non-analogous) sense.

20

The following suggestion is inspired by the way Code (see note 15), 172–3, reads the passage, though it might be different in details.

110

Christof Rapp

We could, hence, wrap up this discussion by flagging the following results: Res 13:

In addition to what has been stated in Res 12, principles are the same in that moving causes are related as something actual to certain potential beings.

Res 13a: There are actually existing movers of the same form, as e.g. the father. Res 13b: There are actually existing movers of a different form, as e.g. the sun. Again, one could get the impression that the reference to such a non-synonymous mover has the function to reserve, as it were, a blank space for other principles, in particular moving principles, that are, in a way, among the principles of sensible ousia, but whose detailed discussion does not fall within the narrow scope of chapters Λ 2–5. The establishment of a non-synonymous mover that can be said to be a principle of all beings in a non-analogical sense would be an important springboard for the argument of chapters Λ 6– 7. It is, hence, a bit surprising that Aristotle does not try to highlight the significance of this result in the present passage. However, in the final summary of chapter Λ 5 there is an erratic hint (1071a35–6) of the claim that the first being is something in actuality (ἔτι τὸ πρῶτον ἐντελεχείᾳ). But what does τὸ πρῶτον in this statement refer to? And where exactly has this been established? Also, this result somehow seems to run counter to another remark from Λ 5, namely that of the first principles of all beings the one is actual and the other potential (1071a18–9): for would it not be fallacious to isolate the actual part from this pair and to present it alone as the first principle? A possible (but quite speculative) way out would be this: The actuality of principles has not been introduced prior to our current passage. Hence, the alleged result stated in 1071a35–6 is likely to refer to something that has been established in this passage. Still, none of the principles in actuality that are mentioned in our passage is explicitly said to be first or primary. In Res 10, at the end of Λ 4, by contrast, we find an unambiguous reference to a principle that is the first of all beings. Just like the sun in our passage, it was introduced as a non-synonymous moving cause. The non-synonymous moving cause in our passage was proven to be a principle in actuality. If this is supposed to be true of all non-synonymous moving causes, we could conclude that the primary principle in Res 10 must be an actual principle too. And such a result, indeed, would be worthy mentioning in the summary at the end of Λ 5. In general, it is a bit unfortunate that Aristotle does not elaborate on the impact of Res 12 and 13, but rather seems to hasten to the next differentiation. However, the section that immediately follows in the text (i.e. the sec-

The Principles of Sensible Substance in Metaphysics Λ 2–5

111

tion starting at line 1071a17) seems to include a brief and casually mentioned corollary of Res 12 and 13: If all principles mentioned so far, i.e. privation, form, matter and mover – can be classified as either actual or potential principles and if all the mentioned types of principles are required to account for change and changeable substances, then it follows indeed that, in the most general account of principles we can give, the principles are just two, namely something that is in actuality and something that is in potentiality. Of course, such a formulation would conceal, as Aristotle said, the fact that these are different for different things (i.e. they include form-like, mover-like etc. principles, which again are themselves different in different areas) and also apply in different ways to them (i.e. the relation of actual and potential principles within one substance as opposed to the relation of actual and potential between numerically different beings). At any rate, something like this seems to be meant when Aristotle states in line 1071a18–9 that for all things there are first principles, one of which is a this that is actually primary and one that is potentially (πάντων δὴ πρῶται ἀρχαὶ τὸ ἐνεργείᾳ πρῶτον τοδὶ καὶ ἄλλο ὃ δυνάμει). (3) In the famous passage that starts at line 1071a17 (‘famous’ because of its role in the notorious discussion about the individuality of form; it is anything but clear, however, whether this passage is significant for the project of Metaphysics Z), Aristotle nails down the simple, but weighty truth, that some causes can be expressed in a universal way (and hence are the same), while some others cannot (hence they are different). He adds the much debated claim we mentioned in the previous section namely that for all things there are πρῶται ἀρχαὶ (primary or proximate principles? or just ‘decisive’?), one of which is a ‘this’ (τοδί) that is actually primary and one that is potentially. However, the universal or general beings (τὰ καθόλου) are not of that kind, for of particular things the principle is a particular thing: Res 14: Some causes can be expressed in a universal way, some cannot. The universals themselves are not the causes, but the particular is a principle of a particular thing. In a way, this seems to be a quite basic concern and ambiguity, and one might wonder why it is only mentioned at such a late stage of the inquiry. Clearly, if we use general terms, such as ‘man’, it might seem that many numerically different principles and causes falling under this description are one and the same, although, as a matter of fact, the general man, i.e. what is common to all particular men, cannot be a cause in the same way as the particular man is. – This seems to be the core message of this passage; it allows, however, of quite different readings: At one extreme, one could read the passage as saying that both ways of dealing with principles and causes

112

Christof Rapp

are equally sound and legitimate. This is what the first sentence of this passage seems to suggest: “Further, one must observe that some causes can be expressed in universal terms, and some cannot.” And, indeed, in Aristotle’s theory of scientific demonstration statements of this sort, referring to generally expressed causes and explanations, do play a crucial role. At the opposed extreme we could understand the passage to mean that all the general formulas are misleading, for they suggest a sameness that does not really exist: it is only a general name that we use as a façon de parler, but there is no way to escape the simple truth that there is nothing like a general man who has ever managed to reproduce himself biologically and, thus, to become a cause and principle in the relevant sense. There are some remarks pointing in this direction: for example, the concluding passage 1071a29–b2 is opened by the remark that the principles and elements are the same, as long as they are homonymously formulated (πολλαχῶς γε λεγομένων), but turn out to be different, once we have distinguished (διαιρεθέντων) the different meanings. Section 1071a24–9, however, might be taken to suggest a different picture: If we look at things belonging to different fields or genera, it is quite clear that their principles and causes are different (they are different to the extent that one might fail to see anything common between them). If, at this level of generality, there is a sense in which principles are the same, it must be the analogical one, which leaves us with quite general terms, such as matter, form, privation, mover (this corresponds to Res 6). When we now move, as it were, one or two levels downwards and look at things within one genus or even within one species, the outlook becomes quite different: In accordance with Res 5 we find that beings within the same genus or realm will use the same type of form, privation and matter as principles. And when we look, more specifically, at things within the same species, we encounter an obvious sense in which the principles are the same for all members of the species, as they all have the same names. Hence, we will no longer seek for ways of unifying and collecting the diverse principles; rather we have to change the perspective of the inquiry and start seeking, conversely, for ways in which these principles are different. And it will turn out, then, that they are “not different in form/species, but in the sense that the causes of different individuals are different (οὐκ εἴδει ἀλλ᾽ ὅτι τῶν καθ᾽ ἕκαστον ἄλλο)” (1071a27–8). What this picture of different levels of generality suggests for the purpose and upshot of the entire endeavour might be something along the following lines: It is not Aristotle’s ambition to dispel the myth of general principles and causes and to make the nominalist point that only particulars can be proper causes. Nor does he aspire to reduce the number and diversity of causes in such a way that he will end up dealing exclusively with these general and analogous notions, such as matter, privation and form. We rather have to

The Principles of Sensible Substance in Metaphysics Λ 2–5

113

take him at his word, when he says at the beginning of Λ 4 that the principles and causes are in one sense the same, and in another way different. This is the agenda of Λ 4–5, namely to articulate these different senses and to spell out how they are the same and how they are not. If we take seriously Aristotle’s tenet in Metaphysics α 2, that an infinity of principles would preclude the possibility of knowledge, then it is, for epistemic reasons, indispensable to bring the multitude of particular principles and causes under more general headings; still, when dealing with these more general terms and headings, we must not delude ourselves into thinking that these general terms themselves have the causal power that only an actual particular being can have.

VI. Conclusions and afterthoughts Chapter Λ 2 sets out to assess the principles of sensible and, hence, changeable ousia. For this purpose, it introduces and reaffirms the three principles of generation or of natural, changeable beings that the reader of Aristotle knows quite well from the first book of the Physics, i.e. privation, form and matter (Res 1). To this set of three principles chapter Λ 3 adds a fourth principle, the moving cause (Res 2); reacting to the introduction of the moving cause the chapter also distinguishes between inherent, simultaneously existing principles on the one side and pre-existing principles on the other (Res 4) and introduces the synonymy principle (Res 3). By the end of chapter Λ 3 all pertinent types of principles seem to be on the table, so that Aristotle can now move on to his next sub-project, i.e. the project of distinguishing in which sense the principles of all beings are the same, and in which sense they are not. From the opening sentence of chapter Λ 4 and from the final sentence of Λ 5 one can conclude that this sub-project is meant to cover the whole of chapters Λ 4–5. Chapter Λ 4 promotes two main ideas: First of all, there is a sense in which the principles of beings within one genus or limited realm are the same (Res 5), i.e. in that all of these beings are brought about by the same types of form, privation and matter. Later on, in chapter Λ 5, however, we are told that Res 5 does not contain the whole story, but must be complemented by something like Res 14, according to which the universals are not themselves the principles and causes, but the particulars that fall under them. Secondly, chapter Λ 4 puts forward the idea that principles and causes of things belonging to different genera are not the same, but are different in different cases (Res 6); what we mean, then, by saying (as in Res 1 and Res 2) that the principles of (all) sensible substances (comprising substances of different genera) are privation, form and matter, thus implying that they are the same for

114

Christof Rapp

all these diverse substances, boils down to saying that in all these cases we find principles playing an analogous role, i.e. the role of privation, form and matter. These numerically and specifically different principles are only analogously related; terms like ‘privation’, ‘form’ and ‘matter’ are designed to bring out the analogies between all these different cases. Due to this peculiar nature their semantics must not be confused with that of, say, a species term: they do not signify a common nature, but allow that things are different in different cases. The weak kind of sameness they imply is, hence, just an analogical unity. However weak this analogical unity might be, it is still important for getting a philosophical understanding of the general nature of changeable beings. The merely analogical sameness of principles of beings in different genera is first stated for the triple of privation, form and matter (Res 7) and is then extended to the quadruple of privation, form, matter and mover (Res 8). In some cases the moving cause and the formal cause are synonymous; in this particular sense two of the four principles in the quadruple may coincide (Res 9). However, there are also movers that are not synonymous and are not different in different fields: this is most notably true of the principle that moves everything and is the first of all beings (Res 10, 10a, 10b). Chapter Λ 5 adds the pair of actuality and potentiality to the envisaged account of the sameness of principles. The various principles are analogically the same in that they are related as actual beings are to potential ones. With regard to the triple of privation, form and matter it turns out that all other principles are actual, while matter is only potential (Res 12), with respect to the quadruple of privation, form, matter and mover it turns out that the mover is a principle in the actual sense (Res 13) and, once again, that there are two types of movers, the synonymous (Res 13a) and the non-synonymous one (Res 13b). Chapters Λ 4 and 5 spend some time dealing with the question of whether substances and non-substantial beings can have the same principles or not. While the earlier chapter seems to deal with this question in a rather aporetic fashion, the latter chapter clearly states that the principles and causes of substances are the causes of all other beings too (Res 11*). Aristotle’s engagement with this kind of question indicates that between chapters Λ 2–3 and Λ 4–5 the focus of the inquiry has been shifted from the question of which are the principles of sensible substances to the question of whether the principles of all beings in the sensible realm (substantial and non-substantial beings alike) are the same or not. The main results of chapters Λ 2 and 3 – that there are four types of principles, privation, form, matter and mover and that the forms co-exist – are introduced without much ado. Quite likely, Aristotle here draws on discussions in certain other treatises: such as Physics I 5–7, Metaphysics Z 7–9

The Principles of Sensible Substance in Metaphysics Λ 2–5

115

or any other pertinent treatise. Thus, his main ambition seems to lie in the sub-project of Λ 4–5, the question of whether and how the principles of all (sensible) beings are the same. This latter sub-project is, first of all, crucial for understanding the result of the earlier chapters, namely that the principles of sensible substances are the mentioned three or four. Without the differentiated result of Λ 4–5 that these principles are in one sense one and in another sense different, one could either make the mistake of thinking that the principles are infinitely many (and that, hence, knowledge becomes precarious, if not impossible) or, else, go wrong in thinking that the principles of many or all things have to be general entities (similar concerns are expressed in the aporiai of Metaphysics B, most notably in the ninth and the fifteenth aporia). This very dilemma is avoided, however, if one understands, as Aristotle suggests, the principles at the highest level of generality as just analogically unifying principles. At the same time the sub-project of Λ 4–5 is pertinent for the second part of book Λ. The unmovable substance that is the topic of the second part of book Λ will be established as a principle on which the heaven and nature depend (Λ 7, 1072b13–4). If this is meant to imply that this unmovable substance, being one and the same, is the principle of everything, the question of whether and how the principles of all beings can be the same is an important preparatory step for the transition to the second part of book Λ. This observation reminds us that Λ 2–5 is not just a self-contained treatise on sensible substance, but also and not just incidentally the first part of the most ambitious project of tackling the invisible and immovable substance and the first and highest principle. With regard to this more ambitious project, Λ 2–5 provides some significant results. First of all, it reveals a sense in which the principles of sensible substances cannot be the same principles of all things in a robust sense (and, hence, cannot provide the first principle tout court): privation, form and matter are only analogously the same; they collect a manifold of principles that are different in different cases and, therefore, cannot be the first principle. Second, the inquiry of Λ 2–5 makes a special case for moving causes, of which the synonymous variant is similarly different in different cases and, hence, fails the test for the one first principle. Still, the inquiry also sets the stage for another variant of moving cause, which is not synonymous and, hence, not different in different cases. In at least one passage, at the end of chapter Λ 4, the first of all beings is proleptically identified with such a non-synonymous type of mover. Apparently, the argument of Λ 6 refers back to such a state of discussion when it takes up the thread by inquiring into the origin of movement. Finally, it is the sub-project of Λ 4–5 that introduces the pair of actuality and potentiality into the discussion. We learn that of the first principles, the one is in actuality and the other is in potentiality and that the moving cause must be something that exists in actuality (which solves a puzzle

116

Christof Rapp

that had been posed in the fourteenth aporia from Metaphysics B). The subsequent chapter Λ 6, whether it uses these ideas as immediate premises or not, argues along the same lines when it shows that the first principle of movement must be something that does not only happen to be actual, but is actual by its essence. – On the whole, this seems to be sufficient evidence to state that the project of Λ 2–5, though officially dedicated to the sensible substances, is also intended to shed light on the nature of the first mover, which itself will not be found among the sensible substances anymore. The treatise of Λ 2–5 is, hence, not a self-contained study of sensible substances; the parts in which the endeavour of these chapters is most ambitious and most original, i.e. the sections in chapter Λ 4–5, quite clearly anticipate the discussion of the first unmovable principle and, thus, Aristotle’s theology. This is, among other things, why it is misleading to picture the study of Λ 2–5 as a project that is competing with the study of Metaphysics ZH, which is also dedicated to sensible ousia. It is true that ZH several times refer to a study of eternal ousia, but these are references to a different, complementary project and not to a project that is part of or directly prepared for by the discussion in ZH. The project of ZH culminates from chapter Z 17 onward in an exposition of what and how sensible ousia is and does not serve the search for a higher kind of principle. Moreover, Λ 2–5 on the one side and ZH on the other are not competing in the sense that one of them might have superseded the other. The main results from Λ 2–5 (i.e. the results concerning the sameness of principles) are not even touched upon in ZH. Where, on the one hand, there is common ground between Λ 2–5 and ZH (this is, to my mind, most notably Z 7–9 and book H), we do not have competing versions of the same theory, but rather a set of assertions in the one treatise (Λ 2–5) that are scrutinized and backed up in the other treatise (for example, the assumptions that in Λ 2–5 are made about form as the actual and matter as the potential part are only presupposed in Λ 2–5, but more carefully justified in Metaphysics H). Where, on the other hand, the two treatises are completely different (concerning, e.g., book Z’s occupation with essence and definition in Z 4–6, 10–11, unity of definition in Z 12, the universal and Platonic forms in Z 13–16, etc.) we have, again, no indication that Aristotle is after a competing or incompatible theory of ousia, but we get discussions of general theorems concerning ousia which the brief and goal-directed treatise of Λ 2–5 simply has no reason to engage with.21

21

For comments and suggestions I would like to thank Andreas Anagnostopoulos, David Charles, Antonio Ferro, Stephen Menn and Christian Pfeiffer.

The Principles of Sensible Substance in Metaphysics Λ 2–5

117

Bibliography Charles, D.: Metaphysics Λ 2: Matter and Change, in: M. Frede/D. Charles (eds.), Aristotle’s Metaphysics Lambda. Symposium Aristotelicum, Oxford 2000, 81–110. Code, A.: Metaphysics Λ 5, in: M. Frede/D. Charles (eds.), Aristotle’s Metaphysics Lambda. Symposium Aristotelicum, Oxford 2000, 161– 179. Crubellier, M.: Metaphysics Λ 4, in: M. Frede/D. Charles (eds.), Aristotle’s Metaphysics Lambda. Symposium Aristotelicum, Oxford 2000, 137– 160. Judson, L: Formlessness and the Priority of Form: Metaphysics Zeta 7–9 and Lambda 3, in: M. Frede/D. Charles (eds.), Aristotle’s Metaphysics Lambda. Symposium Aristotelicum, Oxford 2000, 111–139. Menn, S.: The Aim and Argument of Aristotle’s Metaphysics, IIIβ1: Introduction to Λ, and Λ 1–5, forthcoming.

What the Form Has to Be and What It Needs not Be (Metaphysics, Λ 3) MICHEL CRuBELLIER

Chapter Three is not much to look at. In his 1996 SA paper,1 Lindsay Judson called it the ‘ugly duckling’ of Book Λ. ‘Ugly’ is perhaps too harsh a word, but to most readers it will certainly appear plain and unattractive, since it looks like a loosely connected list of theses of which we have extensive, thorough expositions and smart demonstrations at other places in the Corpus.2 This pejorative assessment is in line with the widespread opinion that the first part of Book Λ is a sort of elementary syllabus of Peripatetic natural philosophy, with no other merit than that of opening the way to the grand metaphysics of the second part. I would like to show that in fact this chapter is better designed than it may appear at first sight, and that it develops in its own way an argument that has no exact equivalent in other parts of the Metaphysics. It is often said 3 that Λ 3 is a résumé or a ‘short version’ of Z 7–9. I would like to point to one important difference. Z 7–9 appears to have been inserted later on into the main text of Book Z, and retains the character of a self-contained essay on the so-called ‘principle of synonymy’ (on which see section 2);4 on the contrary, despite its laconic and sometimes sudden moves, the text of Chapter 3 fits quite well into the general argument of the first part of Book Λ. Chapter 1 leads to the conclusion that the planned inquiry into the first principles and causes of reality in general should begin with a study of what we would call ‘material’ objects – Aristotle describes them as ‘sensible’ and ‘subject to change’. This part of the inquiry is carried on till the end of

1

2

3 4

Judson, L. 2000: ‘Formlessness and the Priority of Form: Metaphysics Zeta 7–9 and Lambda 3’, in: M. Frede/D. Charles (eds.), Aristotle’s Metaphysics Lambda, Symposium Aristotelicum, Oxford, 111–135, 111. The closest parallels are Metaphysics Z 7–8 (and, to some extent, 9) and H 3, 1043a14– 23. There are interesting links with the 8 th aporia of Book B (B 1, 995b31–36 and B 4, 999a24–b24). For instance by Burnyeat (Burnyeat, M. F. 2001: A Map of Metaphysics Zeta, Pittsburgh), 35 n. 65. Burnyeat (see note 3), 34.

120

Michel Crubellier

chapter 5, where a transitional sentence clearly introduces another kind of substance, namely the motionless. Chapters 2 and 3 give a formal description of the processes of generation of sensible objects. What is said of matter in Λ 2 seems to gather the main results of Physics I, 6–9, but a striking fact is that these chapters do not presuppose previous theses or definitions. Aristotle seems to avoid formal technical terminology, or at least to vary intentionally the phrases he uses to mention or describe matter and form. The notion of matter is introduced first (Λ 2), then the form (Λ 3). These chapters belong to the dialectical style of inquiry that Aristotle himself calls ‘logical’. The questions they deal with could be expressed thus: – ‘Under what conditions, or circumstances, are we entitled to speak of change?’ – ‘Under what conditions can we claim that this process of change is intelligible, i.e. susceptible of being explained?’ These questions are a priori. Someone might feel inclined to call them ‘physical’, but they are physical only insofar as change is the general feature that characterizes the realm of ‘nature’. But the answers Aristotle brings to these questions do not require experience-based knowledge of natural facts. Rather, they pertain to the kind of abstract argument that we may find in the Physics, for instance in the first chapter of Book V, in which Aristotle defines change just by the fact that a given subject is described at time t1 by a predicate P and at time t2 by a predicate Q, without postulating anything about an ‘influence’ or ‘action’ exerted by something on something else. Of course, that does not mean that he thinks there is no such thing as a real causal influence or a real process. There is an important difference, in that respect, between Chapter 2 and Chapter 3 of Book Λ. Chapter 2 considers only what is going on within the thing that changes, while Chapter 3 introduces the notion of a cause. To put it in other words, Chapter 2 examines what changing is, while Chapter 3 points out that there must be something to explain the fact that something changes (at a certain time) and the result of that change. Another difference is that while Chapter 2 considers all possible kinds of change,5 the scope of Chapter 3 is restricted to the generation of a substance. A third important feature of these chapters is that they study change using a simple abstract model, i.e. they consider the case of one (otherwise

5

At lines 1069b9–13, Aristotle enumerates the four standard types of change: generation/ corruption, growth/diminution, alteration and change of place.

What the Form Has to Be and What It Needs not Be

121

unspecified) object changing from P to Q (in Chap. 3, alternatively, the generation of one otherwise unspecified 6 substance), without any further assumption as to the relation between the causes of this singular case and the causes and principles of any other possible objects. Such relations are taken up in the next two chapters. Chapter 4 discusses the question: ‘do all things have the same principles and causes, or are these different for different objects?’ It is discussed in the standard form of an Aristotelian aporia: first the cons (1070a35–b10) and then the pros (1070b10–35). It is to be noticed that the negative part of the answer is more definite, while, on the positive side, Aristotle offers nothing more than a plausible interpretation of the phrase ‘the same causes and principles for all things’, which rests on the notion of ‘analogical identity’, which is the weakest possible form of identity. (For our present concerns, it is also worth noticing that this analogy rests on the description of a standard minimal explanation pattern, which recapitulates the main results of Chapters 2 and 3). Chapter 5 brings this discussion to an end by stating as precisely as possible in which ways one can say that all (sensible) things have, and in which ways they have not, the same causes and principles. One important point made in that conclusion is that it would be impossible to claim that certain kinds of items have principles absolutely different from those of other kinds, while it would be equally impossible to say that there are things that have exactly the same causes (even objects which belong to the same species do not have exactly the same causes). In the following pages, I propose a reading of the text that takes as its starting-point the thesis that something is generated ‘from something that has the same name’. I assume that Aristotle takes this sentence as problematic and asks whether one can find a philosophically precise and consistent interpretation of it. That would present a significant contrast to Z 7–9, if Burnyeat is right (as I think he is) in saying that Z 7–9 was originally written to establish or vindicate the ‘principle of synonymy’. I will also put forward the hypothesis that this text might have been written as a sort of answer to a well-known passage of Plato’s Phaedo, the so-called ‘intellectual autobiography of Socrates’. But this historical point is not my main concern here and I hope that the analysis I offer of the general argument can stand even if this connection with the Phaedo is rejected.

6

In sections 4 to 6, that is, in the course of the discussion, Aristotle will introduce some specifications and distinctions that are required for his argument.

122

Michel Crubellier

I. (1a) After that: neither the matter nor the form (I mean the latest ones) is generated. (1b) For everything that changes is something, changes through the agency of something and tends to become something. The agent is the first mover; what changes is the matter; and what that tends to be, is the form. (1c) So there will be an infinite regress if not only the bronze is made round, but the round shape, or bronze are made as well. In fact, that has to stop. ‘Neither the matter nor the form is generated’: of course Aristotle did know that bronze – the example of matter that he adduces for that claim – is in fact generated. Since it is an alloy, it can exist only as the result of a refined technical process. So this sentence cannot mean that the matter for the production of a given object is absolutely ungenerated (for then he would have chosen the worst possible example), but rather that the generation of this particular compound of matter and form does not include the generation of its matter or its form, since that would lead to an infinite regress. For the production of bronze would involve another matter (copper and tin) and another form (say, some definite proportion of each one). Symmetrically, the generation of a ‘round’ shape would have its own matter (continuous bi- or tri-dimensional extension, maybe) and its own form.7 If the same rule were to hold for these second-order matters and forms, any process of generation would involve infinitely many sub-processes, which would result in the impossibility of any knowledge of the natural world (and maybe in the impossibility of the complete realization of the process of generation itself). Thus Aristotle’s claim here is restricted, as he himself says, to the proximate matter and form (as Ross clearly explains,8 this is what ta eschata, ‘the latest’, must mean here). Although the motto anangke¯ ste¯nai is generally associated with the notion that there must always be a point at which a series of causal explanations has to stop (and thus with the doctrine of the Prime Mover), in this context it is a part of a counterfactual. Aristotle’s positive thesis is that a process of generation can be described (and is best described) under the form of an ‘atomic’ pattern, in which just one substratum takes just one definite form under the action of one first mover – i.e. the proximate ones.9 7 8 9

According to Z 8, 1033b1–16, this form of the form could be the abstract notion that every point of the ‘round’ shape has to be at the same distance of a central point. Ross W. D. 1924: Aristotle’s Metaphysics, a revised text with introduction and commentary, Oxford, vol. II, 354. In fact, the case of the proximate first mover is different and slightly more complicated than those of the proximate matter and form: it is ‘first’ also in that it may act through

What the Form Has to Be and What It Needs not Be

123

In other words, the first mover is a cause in that it possesses the form that it brings about in the substratum, and this explanation is self-sufficient. The physician’s medical knowledge fully explains that, and how, the patient recovered health. There is no need for other causes. The causes that would explain the doctor himself and his knowledge (his parents or his teachers at the school of medicine) are not involved in the present process of recovering. Is this only a case of conventional schematization, something like a pedagogical device? I think there is more to it than that. First, Aristotle thinks that the atomic scheme he describes in this chapter and in the following one has a reality of its own, i.e. that the substratum (which constitutes the proximate matter), the form, the moving causes, and the simple relations that exist between them, are elements of the actual process of change and are as real as the sensible particulars. Second, he thinks that change cannot be explained without referring to some non-changing items, as he explains in the second part of the 8 th aporia in Book B: ‘If there is not something eternal, there cannot be generation either. For what is generated must be something, and that from which it is generated too; and in the end these terms must be ungenerated, if it is true that )the regress* must come to a stop and nothing can come to be out of the nonexistent, for no process of change can be infinite, but all )changes* have an endpoint (…). If this is impossible, there must exist something beside the compound, i.e. the structure and the form’ (999b5–16).10

II. (2) After that: every substance comes to be from something which has the same name with the same meaning;11 for things that are by nature are substances, and the rest as well. For things come to be either through art or by nature or by chance or spontaneously; now, art is a principle located in something else and nature a principle located in something which is the same (a man begets a man); and the other )two* causes are privations of these.

10

11

some intermediary steps (using frictions, for instance, or prescribing some drugs, in order to restore health). I assume without further justification that for Aristotle this part of the development of the eighth aporia contains some truth. At least he deems it worthy of consideration, and it seems to me that it is quite similar to the hypotheses that he uses as a guiding thread in our chapter. This somewhat expansive translation of ἐκ συνωνύμου was devised to follow as closely as possible the definition of synonymy in the Categories (1a6–12).

124

Michel Crubellier

In spite of the use of ek (‘from’), the phrase ‘comes to be from something’, etc., does not refer, or at least not specifically, to the material cause (which is ordinarily referred to by ἐκ + genitive). Ross says that it refers here, exceptionally, to the making or moving cause. This interpretation is motivated by the end of the chapter, when Aristotle insists on the synonymy between the moving cause and its product, and thus Ross is basically correct. But I would like to point out that Aristotle would not have expressed it that way if he simply meant the moving cause. He would most likely have said: ‘hupo suno¯numou’.12 So it seems to me that he had a reason to remain deliberately vague or undecided at this point in his argument and that this reason is that he was taking issue with Plato. The thesis that a physical process can be safely explained by referring to something with the same name is crucial to the positive part of Socrates’ autobiographical story in the Phaedo: ‘I thought it was safe to answer to myself, and to anyone else as well, that what is beautiful is beautiful because of the beautiful’.13 Aristotle, following a customary dialectical behaviour of his, begins by accepting the Platonic assertion under its general or external form, then tries to give it a precise meaning and content. That is why, at this point in his argument, he retains only the general notion that the cause of generation must be something which has the same intelligible content as the thing generated. In Λ 3, however, this principle seems to be qualified, since it is stated for substances only. That admits of a stronger and a weaker interpretation. In the stronger version, it means that it does not hold for non-substances, while in the weaker one, Aristotle does not need to state it for other kinds of objects, since, from the beginning of Book Λ, it has been specified that the inquiry bears on the principles and causes of substances. An argument for the stronger version is that, in the following examples, the products of nature or art seem to be conceived of as unified wholes, with a well-defined form that is the aim to which the production process tends. An argument for the weak interpretation is that, in Aristotle’s view at least, items pertaining to other categories are also generated by something synonymous (for instance, heat by something that is hot, and so on),14 and the same seems to be true in the Phaedo, where Socrates mentions such predicates as ‘big’, ‘two’, and so on. The truth probably lies somewhere in between, that is, the generation of substances is probably taken as a paradigmatic case.

12 13 14

See Z 7, 1032a24 for a similar phrase. Phd, 100e. In Chap. 4, 1070b10–12, after having announced that he will speak of the elements of sensible bodies, Aristotle considers the generation of ‘the warm’.

What the Form Has to Be and What It Needs not Be

125

The claim that the cause must be synonymous with the thing caused is supported by an explanation, the exact meaning of which does not appear immediately. It is given in two sentences introduced by γὰρ (‘for’): ‘for things that are by nature are substances, and the rest as well. For things come to be either through art or by nature or by chance or spontaneously’, and so on. According to regular Greek grammar, the first sentence means (1) ‘that things that are by nature’ are substances, and (2) that )all?* the other things )are substances* as well. Here lie two difficulties. First, the second part seems to render the first one irrelevant, since if everything whatever is a substance, what is the point of mentioning that natural beings, in particular, are substances? Second, there is no immediate link between the notion that natural beings are substances and the claim that everything comes to be from something which has the same name. To bridge the gap, one could begin with the reasonable guess that natural beings are introduced here as a prime example of something that comes to be from something synonymous. That Aristotle is thinking of the reproduction of living beings is unquestionable, since the well-known adage ‘a human being begets a human being’ is invoked immediately after, and is repeated at the end of the chapter (a27–28). Then ‘art’, ‘luck’ and chance’ are introduced in such a way that they each appear as a kind of variant of natural generation. This is obvious for technical production, which is said to differ from the natural one by the fact that art, as a principle of generation, acts on ‘something else’, while the principle of natural generation acts from within something which is ‘the same’. This seems to be a well-known Aristotelian tenet: nature is an inner principle for a goal-directed change, while art ‘imitates nature’, acting from outside. But in the present case the distinction is not that clear. Other kinds of natural changes (for instance, growth or the fulfilment of vital functions) have a principle which is internal to the living body into which the process takes place, while the generation of a new living being does require an agent which is ontologically distinct from the being that is generated (in Aristotle’s zoology, the father). Nevertheless, this agent may be called ‘the same’ in a weak sense. There is greater difficulty in explaining ‘luck’ and ‘chance’. They are characterized as ‘privations’ of the first two types of causes. To understand this, one must consider that something that happens by luck is the same sort of result as we would have chosen and endeavored to obtain by a conscious goal-directed activity: it is like a product of art, except that no art has been used in order to bring it about. Similarly, one could describe the product of luck as something which has developed spontaneously, just like natural productions, except that there was no definite agent imparting its own form

126

Michel Crubellier

to the product (well known examples: a patient returning to health because he had to run and that restored the correct amount of natural warmth into his organism; or the spontaneous generation of mice out of a bunch of old rags). So, clearly, in these cases, the generation of some given object does not depend on ‘something that has the same name’. It seems strange and paradoxical to support the claim that ‘every substance comes from something synonymous’ by mentioning that there are cases in which no such cause is involved. Indeed, for any person educated in the standards of modern scientific methodology, these are counterexamples, which would amount to a refutation of this thesis; but Aristotle appears to argue in a different fashion. In themselves, luck and chance events cannot be completely explained, they are not completely intelligible. But they may at least be described as generationprocesses (as opposed to what would have been the case if the man’s running or the bunch of rags had produced only chaotic or disordered effects); and such a description can only be made (and understood) by reference to the standard models of natural or technical productions. Here, the exception straightforwardly proves the rule. We are now in a position to understand how the argument goes in the second sentence introduced by γάρ. Since, in every situation of which I can say that ‘something is generated’, I am able to do so either because it is a case of natural generation or because it resembles natural generation up to a certain point, we have to look at natural generations to understand what the generation of something really is (i.e. to recognize what is intelligible in it). Now, let us return to the laconic sentence τὰ γὰρ φύσει οὐσίαι, καὶ τἆλλα. The second part of it (‘and the rest as well’) may be read as a concise anticipation of the following sentence, that is, the extension from natural beings to other cases of generation. One could paraphrase: ‘now, what is true of generation of natural beings will also hold, up to a certain point, for other cases’

III. (3) There are three substances: matter, which is a certain ‘this’ in the sense that it appears )so* (for all things that exist by contact and not through natural continuity, are a matter and a substratum); nature is a certain ‘this’ towards which )the process of change* is directed, i.e. the possession of something; and thirdly there is the particular substance which is made of these, like Socrates, for instance, or Callias.

What the Form Has to Be and What It Needs not Be

127

The distinction between three types of substances, or the three meanings of ‘substance’, is another well-known Aristotelian thesis.15 But what we have here is not just a pedantic reminder of a piece of doctrine. An interesting difference appears immediately in what is said about matter. When he considers the possible grounds for calling matter a substance, Aristotle usually mentions (for instance in Z 3, B 4) the fact that ousia is generally acknowledged to be a substratum and that matter is nothing else than a substratum (and ungenerated as such); or, more precisely, as he does in Chapter 2, he argues that matter is potentially the objects that can be produced out of it. Here, he considers the possibility of calling matter a ‘this’. The same fact is mentioned for the form (although it is called ‘nature’) in the next sentence. It seems to me that τόδε, in this context, would be best understood as a substitute for the name shared by the thing produced and the entity that is the cause of its generation (e.g. human being). The notion that the matter of a human being might be validly said to be a human being (or this human being) is clearly problematic to Aristotle himself. Compare what is said here with the matter-of-fact way in which the next sentence assumes that ‘nature’, in the sense of the final goal of generation, is the human being (and, still more significantly, in the case of the compound of matter and form, the fact that Socrates or Callias is a human being simply goes without saying). What does Aristotle mean by saying that matter ‘is a this because it appears )so*’? Matter is the human being insofar as it is what we perceive when we see (touch, hear, and so on) this human being. This assertion is expressed in a positive form; that is, Aristotle does admit that saying that what I see is a human being is undoubtedly a sound way of speaking. However, the following sentence immediately lessens the force of this admission, since it defines matter as something which lacks real unity, since it is what it is ‘by contact and not through natural continuity’. That amounts to saying that this is not the real human being, just as Aristotle says in other contexts that a corpse is not an animal, or that a hand separated from the living body is no more a hand – although we are used to calling them by the same names.16 To sum up: section 3 considers successively three possible referents for the name that indicates a substance. While making this review, Aristotle has in mind the thesis that the cause of the generation of something has to be something with the same name. Of these three notions of substance, he has dismissed the matter, as corresponding to a loose way of speaking. Thus we are left with two items, the

15 16

De an. II 1, 412a6–9. This is a famous case of homonymy, as opposed to synonymy.

128

Michel Crubellier

‘nature’ and the compound. The compound is clearly the thing generated, so that the problem is to know whether, and under what precise form, the – so far undefined – ‘nature’ could provide a causal explanation for its generation.

IV. (4a) In some cases, for sure, the ‘this’ does not exist beside the compound substance, for instance the form of a house (unless it be the art), and there is neither generation nor corruption of these; in fact the house without matter, or health )without matter*, and everything that depends of an art, are and cease to be in a different way. (4b) Now, if ever that may happen, that will be the case for natural objects. (4c) That is why Plato was not so wrong when he said that all the things that are by nature are Forms, provided there might be Forms also of such natural things as fire, flesh or a head: for all these things are matter, and the latter is )the matter* of what is most properly a substance. The question whether one can conceive of a ‘this’ distinct from the particular compound (παρὰ τὴν συνθετὴν οὐσίαν) is linked to the question asked at the end of the previous section, i.e. the causal role of ‘nature’. That does not mean that the affirmative or negative answer to this question will immediately entail a solution to the problem of causal explanation. Undoubtedly, the cause must be distinct from what it explains; but there are different ways of being distinct, and Aristotle had a keen eye for such things. Therefore, a correct Aristotelian solution must include the determination of the kind of distinction that exists between something and its cause. Socrates’ ‘naive but safe’ solution, in the Phaedo, rested on the assumption that there exists ‘a beautiful in itself and by itself, and a good, a large, and all the rest’.17 The Platonic Forms provide a simple and radical interpretation of the thesis that everything is generated from something that has the same name. For every ‘this’ or ‘that’ which can be known and understood in the sensible or natural world, there must exist an object that is a ‘this’ in an absolute sense, which implies that it is nothing else than a ‘this’, and thus must exist apart from all the sensible, changing objects that are ‘this’ only partly, relatively and temporarily. As it is well known, and as Plato himself recognized, this solution raises in turn serious difficulties as to the sort of things that such Forms can be, the delimitation of the ‘thises’ that may be

17

Phd, 100b (Socrates goes on: ‘if you agree and grant me that point, I hope I’ll be able to discover and explain to you that the soul is immortal’).

What the Form Has to Be and What It Needs not Be

129

given such a status, and their relation to the concrete particular things that ‘participate’ or ‘partake’ in them. As a disciple of Plato, a reader of the dialogues and a participant in the discussions about Forms in the ancient Academy, Aristotle was well aware of these difficulties. The argument in Chapter 3 is mainly directed at finding a plausible interpretation of the thesis ‘every substance comes to be from something which has the same name’ that could be immune to the difficulties met by his master. The question: ‘in which cases does a this exist apart from the concrete sensible particulars?’ seems to have played a role in the discussions about Forms in the Academy. It is attested in the Parmenides, and Aristotle also alludes elsewhere to limitations put to the realm of Forms by Plato or the Platonists.18 He also raises the question in dialectical contexts, as a part of the general problem of Forms. The elements of the answer that he gives here are far from clear. He alludes to two classes of generation processes, natural and artificial. This picture might become more complicated if we take into account the ambiguous case of health, which in itself is a most natural state of affairs, but happens to be produced by an art, medicine, which was created and is practiced for that specific end. In a characteristic Aristotelian fashion, he gives no clearcut answer to his own question, although he suggests that it is more likely to be the case with natural beings than with artefacts, and that this is relevant for the topic of explanation through forms. That he wishes to avoid a sharp opposition between the two cases, can be seen by the fact that he qualifies both sides of his answer: 1. the form of artefacts does not exist apart from the sensible particulars, ‘unless it be the art’ (a14–15, confirmed at the last line of the chapter by the claim that ‘the art of medicine is the account of health’, on which see below); 2. ‘if ever it may happen’ that the ‘this’ be something apart from the compound substances, ‘that will be the case for natural objects’; moreover, that eventuality is submitted to the (problematic) condition that there be Forms for some objects (fire, flesh, a head) for which that seems less plausible (a19–20). But these qualifications will not cancel the fact that Aristotle considers that, in any case, the existence of a separate form suits natural beings better than artefacts. This claim is set out quite directly at lines a13–14 as if it did not

18

Parm. 130b–e and Met. A 9, 991b6–7, quoted below in section 6.

130

Michel Crubellier

require any argument. This is not due to the general terseness of our chapter, since it is introduced in exactly the same way in other contexts.19 Nevertheless, it seems to me that this point might surprise at least a modern reader. It seems easy to apprehend the form of a house as something distinct from the house itself, e.g. by imagining the plan of the house, drawn on a sheet of paper, or the ‘mental image’ in the mind of the architect or master mason. In fact, Aristotle himself points at something like that by saying ‘unless it be the art’, and there are some passages in the Corpus that seem to support this view.20 Here, too, matters are perhaps more complicated, insofar as ‘the art (of building)’ refers to a generic potentiality (the ability to build houses), whereas the ‘compound substance’ must be a particular house. But, be that as it may, it seems that in the case of natural beings, we have no such items as a plan or a mental image available to help us conceive the notion of a ‘nature’ distinct from the corresponding particular. Thus, Aristotle’s assumption that this should be more easily admitted for natural beings does not seem to rest on psychological grounds. Why, then, did he make it? I think we may gather some indications from what follows, mainly from the mention that things like ‘the house without matter’ do not undergo processes of generation or corruption. At first sight, this remark too may seem very surprising: if they are not liable to generation and corruption, does not that mean that such entities must be entirely apart from the sensible house, and more generally from the realm of changing sensible things? But the text specifies that they ‘are and are not in a different way’, which means, first, that they come and cease to be. The ‘different way’ in which they come and cease to be, although it is not spelled out here, is described in H 3, 1043b14–16 as ‘passing away without a process of destruction and coming to be without a generation’. Their coming-to-be is not a generation because there is no such thing as a substratum which would exist and change continuously throughout the process. They come to be accidentally, so to speak, when the parts which constitute them are put in contact. Therefore, they are less constant and less rooted in reality than beings that are subject to real generation and corruption. On the contrary, the ‘nature’ of natural beings, since it is more continuous and self-contained, is a better candidate for existing ‘in itself and by itself’, distinct from the particular natural beings. This criterion is basically Platonic. 19 20

Met. H 3, 1043b19–21; B 4, 999b18–20; and (in a Platonic or Academic context) A 9, 991b5–7. Z 7, 1032b1–14; Λ 9, 1075a1–3. – At H 4, 1044a29–31, Aristotle points out that this is linked to a specific condition of artificial productions, namely a certain indeterminacy of matter in relation to the form; but this condition is confined within narrow limits, since ‘one will not make a saw out of wool’.

What the Form Has to Be and What It Needs not Be

131

‘That is why’ Aristotle continues, ‘Plato was not so wrong when he said that all the things that are by nature are Forms’. Notice that this is the first positive mention of the form or Form. (At line a14, Aristotle mentioned ‘the form of a house’, but that could refer to the general shape and structure of the house; moreover, this mention occurred in a counterfactual hypothesis. Even here, the use of this term may be due to the Platonic context.) I leave aside the questions related to the historical value of this testimony about Plato’s doctrines. The restriction introduced at the end of the section, in the case of ‘things like fire, flesh or a head’ seems to me unmistakably Aristotelian, and thus is not a part of the Platonic testimony. For it refers to two distinct pieces of Aristotelian philosophy, which appear to conflict on that particular point. According to the doctrine of the Categories, the parts of a substance are substances, and thus a head is a substance (the criterium here seems to be grammatical: a head is designated by a noun, and heads may be counted). On the other hand, the examples that are adduced here form a series, in which each term is ‘matter’, i.e. a part of the matter of the following one, according to Aristotle’s biology; fire is an elementary body and one of the components of flesh; flesh is one of the tissues (an ‘homoeomeric part’) of which the ‘anhomoeomeric parts’, i.e. the organs of the living being, are made. From this point of view, none of these things, although they are undoubtedly natural, is a substance. Another way of expressing the same idea would be to point out that, although each of these things is the result of a teleological process, none of them is the final end of this process. The final end is the generation of the complete animal. (At line a20, I take ἡ τελευταία to refer to κεφαλή, ‘a head’, which is feminine in Greek, as well as ὕλη, ‘matter’, and I translate it accordingly as ‘the latter’).21

V. (5a) So then, the moving causes are causes on the condition that they have come to be previously, (5b) while those which are causes in the sense of ‘what accounts for )something*’ exist together )with the thing caused*. For when a man is healthy, then health exists, and the shape of a bronze sphere exists simultaneously with the bronze sphere. (5c) Whether something lasts longer, is a question to scrutinize. In some cases

21

The meaning of the sentence would not really be different if it was referred to ὕλη and treated as equivalent to τὰ ἔσχατα (‘the latest’) at line 1069b36. It would just be changed into a more abstract and universal proposition: ‘)in any case whatever* the latest matter is the matter of what is most properly a substance’.

132

Michel Crubellier

nothing stands in the way, e.g. if soul is such a thing – not the whole of it, but intelligence, for it is probably impossible that the whole of the soul )could last longer*. The next sequence considers the possible temporal relations between different kinds of causes and the facts or objects that they are supposed to explain: must the cause exist before the sensible particular thing it explains (a21), or together with it (a21–24), or could it even be the case that it should last longer than it (a24–26)? What is the point of asking these questions here? As I have mentioned in my introductory remarks, Chapter 3 takes a step forward by bringing in the notion of cause, which was not taken into account in Chapter 2, and the notion of a cause implies a definite position in time. Aristotle was aware of this point, which he discussed in detail in Posterior Analytics II, 12. (One could say that causality is precisely the hinge between logic and natural philosophy, and that it gives access to natural philosophy because it implies temporal determinations). He also seems to have considered that this was a weak point in Plato’s conception of a cause. As we will see in section 6, the facts established in the present section will result in the rejection of Plato’s eternal Forms. Maybe these general remarks will dispose the reader to take a more indulgent view of the outset of this section. For, nonetheless, it begins in a rather abrupt fashion: a) οὖν (‘So then …’ at line a21) is normally used to tie up the thread of the argument after a digression.22 Now, lines a18–20 may certainly be seen as a digression, but – to which point of the preceding argument are we supposed to fasten the thread? b) lines a21–22 are taken to refer to two of the notorious ‘four causes’, and this might be viewed as another instance of the ‘syllabus-like’ style of the chapter. However, they have not been mentioned, nor alluded to, in the previous lines, neither do there appear any of those phrases used to introduce some well-known piece of doctrine (not even something like the οὐσίαι δὲ τρεῖς of line a9). Aristotle does not tell us why he will consider only two of the four standard causes, and we have to supply a fair amount of the argument by ourselves. 22

An alternative solution would be to treat μὲν οὖν as a combination, thus meaning ‘at least’ or ‘for sure’ (as it is most probably the case at the beginnng of 4a, line a13) That might fit with the reference to a well-known piece of doctrine. But the parallelism τὰ μὲν (…) κινοῦντα / τὰ δ’ὡς ὁ λόγος seems unescapable; and even translating by ‘for sure’ would not make things really better, since the reference to the four causes is far from obvious, as we will see.

What the Form Has to Be and What It Needs not Be

133

We may tacitly admit that the ‘material’ cause has been ruled out, since it was said that matter is not really a ‘this’.23 The ‘movers’ of line a21 are most certainly instances of the ‘starting-point of change’ or ‘moving causes’. The phrase ‘those which are causes as the account’ suggests the ‘formal cause’, which is often called logos (for instance at a29). Now, why does Aristotle overlook the final cause here? In section 3, the mention of ‘nature’ as ‘a certain this towards which the process of change is directed’ would seem to refer to the final cause; but immediately after that, we read that a sensible particular like Socrates or Callias is ‘made of these’, i.e. matter and the ‘nature’. Thus, the overall picture we get is something like the one in Physics II chap. 7: matter remains in the background, the formal and final causes are identified, while the moving or making cause is a sort of counterpart of them: ‘three of these causes often come down to one, for the what it is and the for the sake of what are one and the same, and the starting-point of the process is of the same species as these, for a human being begets a human being’.24 So, these lines are not just a digression. By specifying the temporal connections that might exist between a cause and what is caused by it, Aristotle is able to individuate two kinds of cause, each one being characterized by a distinct temporal connection, and to give a definite content to the generic notion of ‘something with the same name’. These results will serve as a basis for his argument against Platonic Forms. Before moving on to this last section of the chapter, let us have a look at the answers that are given at lines a21–26. They are not difficult in themselves. That the moving cause must ‘have come to be previously’ is not inconsistent with Aristotle’s general thesis25 that the effect is simultaneous with its cause, since it says only that the process of generation of the cause must precede its acting as a cause. That the formal cause exists together with the particular thing or fact it explains, is self-evident: otherwise it would not be its form. But it must be pointed out that the text, in this passage, does not claim anything more than that: i.e., Aristotle does not say that there is nothing more in health, or in the spherical form, than the state of this particular man or the shape of that particular piece of bronze.

23

24 25

However, insofar as it is really a cause, matter too should be synonymous with the thing it explains. This is true in a sense, since matter, as Aristotle said in chapter 2, is always a matter for a certain thing X: ‘Now, if )matter* is something potentially, nevertheless that will not be anything whatever, but a different thing )has to come* from a different matter’ (1069b28–29). Therefore, this matter is (potentially) a this. But it must be pointed out that this kind of ‘synonymy’ is quite different from the one alluded to at lines a9–11 (matter ‘appears to be a this’). Phys. II 7, 198a24–27. Anal. Post. II 12; Phys. III 1, 201b5–7.

134

Michel Crubellier

The mention of the soul and the question asked about its possible immortality in section 5c seem more surprising. One might be tempted to explain them by a sort of very formal concern for symmetry or exhaustivity: since we have considered the existence of the cause before its effect and at the same time, we should perhaps consider the possibility of its existence after it.26 In fact, the answer seems rather elusive, and, unlike the answers in sections 5a and 5b, it does not result in the individuation of one specific type of cause. To remain consistent with the preceding answer, the ‘cause’ that would last longer than its effect should be the form of the sensible particular, of which the soul of a living being is a prime example. Now, if we try to spell out Aristotle’s answer by referring to his theory of intellect, we will find out that this ‘part of the soul’ is able to exist after the disappearance of the living body precisely because it is not the form of any part of the body whatsoever. The soul as a whole could not last any longer.27

VI. (6) Then it is clear that there is no reason, just on the basis of what has been said here, to hold that there must be Ideas: for a human begets a human, )i.e.* a particular human being begets this one here; and the same holds for the products of art, since the art of medicine is the account of health. The translation tries to retain the ambiguity of ‘just on the basis of what has been said here’ (διά γε ταῦτα a27).28 This may mean either that what Aristotle has put forward is enough to establish that there is no need for Platonic Ideas, or that the Platonic arguments he has reported are not enough to establish that one must posit Ideas. The second possibility may seem rather unlikely, since nowhere does Aristotle explicitly say he is reporting an argument for the existence of the Ideas. Yet there are parallel uses of dia ge tauta in Book N, in the discussion about the real or substantial existence of numbers.29 Although the text of lines 1090a17–20 is difficult,30 the second

26 27 28 29 30

One could also see here another reminiscence of the Phaedo, if it is true that the chapter was meant to revisit Socrates’ defence of the doctrine of Forms. De an. II 1, 413a3–7. The same phrase occurs in a parallel passage of Z 8, 1033b29. Met. N 2, 1090a9; N 3, 1090a20. These two occurrences of the phrase διά γε ταῦτα belong to the same context (the traditional division between chapters 2 and 3 is misleading). The text of the manuscripts is most certainly corrupt. I follow Christ’s emendation (i.e. the suppression of λαμβάνειν at lines a17–18).

What the Form Has to Be and What It Needs not Be

135

occurrence is illuminating: ‘Those who posit the Ideas and say that they are numbers, try in a way, at least, by referring to the operation of extracting a particular )X* out of a multiplicity, to say how and why, in each case, there is a unique )X*; but since what they say is neither necessary nor possible, there is no need to say, at least for these reasons (διά γε ταῦτα), that numbers exist’. If this is accepted as a parallel, it would confirm my hypothesis that the mention of ‘something with the same name’ was meant to hint at the Phaedo. Be that as it may, the thesis that one can do without Plato’s Forms rests on the fact that ‘a human begets a human’. The specification ‘a particular human being begets this one here’ is called for by the fact that the formula ‘an F produces an F’ could be used to describe the causality of Ideas as well: for Socrates, Beauty-itself, which is in a sense a beautiful thing, is the cause of beauty in the particular beautiful things. On the other hand, Socrates considered that a moving cause could provide only an accidental or arbitrary explanation (see Phaedo 96e–97b, 98b–e) since it had no necessary intelligible connection with the fact that it is supposed to explain. On the contrary, Aristotle claims that the principle of synonymy can be satisfied at the level of a particular moving cause; i.e., that the logos, the intelligible connection, does exist between the particulars themselves. We are now in a position to understand better the way the argument goes in Chapter 3 taken as a whole. Aristotle is trying to find a plausible interpretation for the principle of synonymy; in order to do so, he circumscribes the field of his inquiry in two ways: First, by considering the thing generated within its immediate neighbourhood, so to speak, i.e. its relations to its immediate matter and form and its immediate producer. Second, by asking for the temporal connections between a cause and the facts or objects it explains. These two specifications enable him to individuate three distinct types of cause (namely the material, the formal and the moving cause) and to establish that since the real moving cause must be able to bestow the form upon the generated thing, it must first have it in some definite way. One consequence of this way of arguing is that the formal cause seems to be left in the background. In a sense, this is a general feature of Book Λ, which does not dwell much on the nature and status of the formal cause (while these topics are pursued in the central books). In the last lines of Chapter 3, Aristotle even seems to blur the difference between natural forms and the forms of artefacts, since ‘the art of medicine’ (ἡ ἰατρικὴ τέχνη, in which the explicit mention of techne¯ points out to the specific activity of the physician) is ‘the logos of health’. A similar impression emerges from an additional argument that Aristotle occasionally adds to the sentences ‘a human begets a human’, in Metaphysics A 9 or Generation and Corruption II 9. There (with an explicit reference to the Phaedo in both cases) he suggests

136

Michel Crubellier

a sort of experiment in order to show that the Ideas, even if they existed, would be neither sufficient nor necessary to explain natural processes: the eternal existence of health, as the norm of human physical life, has never been able to cure an illness, unless there be a physician in a position to practice his art on this particular patient. Symmetrically, ‘there are many things that are produced, like a house or a ring, of which we believe that there are no Forms. So that it is clear that it is possible for other things to be or come to be by the agency of causes such as the ones we have mentioned’.31 If one gives the last sentence its full force, that would entail that the causes of natural generation need not be more real or eternal than the causes of artefacts. That would certainly be an exaggerated version of Aristotle’s anti-Platonism. For him, there are two important differences between the form of artefacts and the forms of natural beings: (1) the forms of natural beings are complex and self-contained wholes, which have more reality than their parts,32 while the forms of artefacts can be broken down into simpler structures. The efficiency of technology rests precisely on that possibility. Thus, (2) the production of an artefact requires a deliberating agent, while in the case of a natural being, the cause of its production is nothing else than the form itself.

Appendix The Greek Text (After Silvia Fazzo’s edition,33 with some minor changes. Contrary to Ross, Fazzo avoids parentheses and punctuates lightly.) [69b35]

(1a) Μετὰ ταῦτα ὅτι οὐ γίγνεται οὔτε ἡ ὕλη οὔτε τὸ εἶδος, λέγω δὲ τὰ ἔσχατα. (1b) πᾶν γὰρ μεταβάλλει τὶ καὶ ὑπὸ [70a] τινὸς καὶ εἴς τι· ὑφ’ οὗ μὲν, τοῦ πρώτου κινοῦντος· ὃ δέ, ἡ ὕλη· εἰς ὃ δέ, τὸ εἷδος. (1c) εἰς ἄπειρον οὖν εἶσιν, εἱ μὴ μόνον ὁ χαλκὸς γίγνεται στρογγύλος, ἀλλὰ καὶ τὸ στρογγύλον ἢ ὁ χαλκός· ἀνάγκη δὴ στῆναι· (2) μετὰ ταῦτα ὅτι ἑκάστη [5] ἐκ συνώνυμου γίγνεται οὐσία· τὰ γὰρ φύσει οὐσίαι, καὶ

31 32 33

A 9, 991b3–9; there ‘we’ means ‘we, disciples of Plato’. See also Generation and Corruption II 9, 335b7–24. As is clear from the dictum that a hand separated from the living body is ‘a hand’ only in an equivocal sense. Fazzo, S 2012: Il libro Lambda della Metafisica di Aristotele, Napoli.

What the Form Has to Be and What It Needs not Be

137

τἆλλα. ἢ γὰρ τέχνη ἢ φύσει γίγνεται ἢ τύχῃ ἢ τῷ αὐτομάτῳ. ἡ μὲν οὖν τέχνη ἀρχὴ ἐν ἄλλῳ, ἡ δὲ φύσις ἀρχὴ ἐν αὐτῷ, ἄνθρωπος γὰρ ἄνθρωπος γεννᾷ. αἱ δὲ λοιπαὶ αἰτίαι στερήσεις τούτων. (3) οὐσίαι δὲ τρεῖς, ἡ μὲν ὕλη [10] τόδε τι οὖσα τῷ φαίνεσθαι, ὅσα γὰρ ἀφῇ καὶ μὴ συμφύσει, ὕλη καὶ ὑποκείμενον· ἡ δὲ φύσις τόδε τι εἰς ἣν καὶ ἕξις τις· ἔτι τρίτη ἡ ἐκ τούτων ἡ καθ’ ἕκαστα, οἷον Σωκράτης ἢ Καλλίας. (4a) ἐπὶ μὲν οὖν τινῶν τὸ τόδε οὐκ ἔστι παρὰ τὴν συνθετὴν οὐσίαν, οἷον οἰκίας τὸ εἶδος, εἰ [15] μὴ ἡ τέχνη, οὐδ’ ἔστι γένεσις καὶ φθορὰ τούτων, ἀλλ’ ἄλλον τρόπον εἰσὶ καὶ οὐκ εἰσὶν οἰκία ἡ ἄνευ ὕλης καὶ ὐγιεία καὶ πᾶν τὸ κατὰ τέχνην· (4b) ἀλλ’ εἴπερ, ἐπι τών φύσει· (4c) διὸ δὴ οὐ κακῶς Πλάτων ἔφη ὅτι εἴδη ἔστιν ὁπόσα φύσει, εἴπερ ἔστιν εἴδη ἄλλα τούτων, οἷον πῦρ σὰρξ κεφαλή [20] ἅπαντα γὰρ ὕλη ἐστί, καὶ τῆς μάλιστα οὐσίας ἡ τελευταία. (5a) τὰ μὲν οὖν κινοῦντα αἴτια ὡς προγεγενημένα ὄντα, (5b) τὰ δ’ ὠς ὀ λόγος ἅμα. ὅτε γὰρ ὑγιαίνει ὁ ἅνθρωπος, τότε καὶ ἡ ὑγιεία ἔστιν, καὶ τὸ σχῆμα τῆς χαλκῆς σφαίρας ἅμα καὶ ἡ χαλκῆ σφαῖρα. (5c) εἰ δὲ καὶ ὕστερόν τι ὑπομένει, σκεπτέον· [25] ἐπ’ ἐνίων γὰρ οὐδὲν κωλύει, οἷον εἰ ἡ ψυχὴ τοιοῦτον, μὴ πᾶσα ἀλλ’ ὁ νοῦς· πᾶσαν γὰρ ἀδύνατον ἴσως. (6) φανερὸν δὴ ὅτι οὐδὲν δεὶ διά γε ταῦτ’ εἶναι τὰς ἰδέας· ἄνθρωπος γὰρ ἄνθρωπον γεννᾷ, ὁ καθ’ ἕκαστον τὸν τινά· ὁμοίως δὲ καὶ ἐπὶ τῶν τεχνῶν· ἡ γὰρ ἰατρικὴ τέχνη ὁ λόγος τῆς ὑγιείας [30] ἐστίν.

Bibliography Burnyeat, M. F.: A Map of Metaphysics Zeta, Pittsburg 2001. Fazzo, S.: Il libro Lambda della Metafisica di Aristotele, Napoli 2012. Judson, L.: Formlessness and the Priority of Form: Metaphysics Zeta 7–9 and Lambda 3, in: M. Frede/D. Charles (eds.), Aristotle’s Metaphysics Lambda. Symposium Aristotelicum, Oxford 2000. Ross, W. D.: Aristotle’s Metaphysics. A Revised Text with Introduction and Commentary, Oxford 1924.

Individuals, Form, Movement: From Lambda to Z–H MARCO ZINGANO

I would like to explore a possible scenario, aiming to show that it is a likely scenario. The possible scenario is this. According to the Categories, the ontological pieces of the world are individuals, there called pro¯tai ousiai, first substances, in which all the accidents inhere and of which all species or genera are said. Species and genera display a Janus-face nature. On the one hand, they are substances, for they are expressly said to be second substances in this treatise; on the other, they are treated as a sort of quality, poion ti, since they mark off what first substances properly are. In contrast to this, the reader of the central books of Metaphysics finds a different account: forms, which correspond to the species of the Categories, have now ontological priority and are consequently posited as pro¯tai ousiai, whereas genus is banned from the realm of substance. Individuals are indeed substances, but they are no longer first substances. Thus, there is a shift, and it does seem to be significant, as one moves from individuals towards forms as first substances. I will take it for granted that there is an important discontinuity between Categories and Metaphysics. This is of course controversial, for one may attempt to harmonize the two texts, stressing instead their continuity. But I would like to entertain here the idea that the discontinuity-scenario is not only possible, but also plausible. My strategy will be to show that persistent puzzles in Lambda are better understood if read in line with the scenario I propose. Let me lay my cards squarely on the table and present my thesis. Book Lambda stands half-way between the two positions represented by Categories, on the one hand, and the central books of Metaphysics, on the other. For Lambda still accords ontological priority to individuals, as the Categories does, but in so far as Lambda looks for a connection between the realms of sensible and non-sensible substances by means of a common cause of movement, it differs significantly from the Categories and announces what will be pursued in Metaphysics Z–H. To show this, let me first highlight an important convergence between Lambda and Categories. As is well known, Categories puts individuals in a central position; if we keep the manuscripts’ reading and refuse Simplicius’

140

Marco Zingano

proposal of expunging a line in chapter 5 for reason of (alleged) dittography (2b6), then in the course of just 14 Bekker lines we meet three times the assertion that all other things are either said of the individuals as substrata or are in these same substrata. Book Lambda shares with Categories the same privileged ontological status for individuals. But there are other tenets which are peculiar to Categories. Being in a substratum corresponds to accidents; being said of a substratum corresponds to eide¯ or gene¯. Species and genera are kept in a sort of ontological limbo. They are deuterai ousiai, and thus they are substances, but they do not satisfy all the criteria of a substance; in particular they do not satisfy the tode ti requirement. Every substance signifies a tode ti, as one reads in Categories (3b10). This is immediately true for first substances, or individuals. As for species and genera, the so-called second substances in this treatise, they at first glance signify also a tode ti, thanks to the form of the name, e.g. man or animal. This is not true, however. For they signify rather a certain quality, since they qualify the proper substances or individuals. They are not just any quality, though, but a special sort of quality, insofar as they mark off what a first substance is. Hence, they waver between these two categories – substance, as second substances, and quality, as a quality of a sort –, remaining in what I would call a metaphysical limbo: more than a simple quality, less than a proper substance. The chapters of Categories exhibit a double structure. In the first part, it is said which and how many items belong to the term under consideration. For instance, in chapter 5, devoted to substance, there are two sorts of substance, first and second substances, the latter being further divided into species and genus. In the second part of each chapter are described the main traits or characteristics that any item must satisfy in order to belong to the category at issue. These traits are very abstract, like being in another thing or not being in something else, having a contrary or having no contrary at all, being liable to degrees or not being liable to more and less, and so on. To qualify for a certain category, an item has to satisfy the whole set of conditions, or at least most of them. However, at the same time, there is one specific trait, which is seen as idion to that group or category, whose satisfaction is necessary and sufficient to qualify as a member of a given category. Conversely, its non-satisfaction prevents an item from belonging to it. These two criteria are not the same, but the Categories do not address this issue. It operates with both criteria, recognizing no problem with this duality.1 Concerning substances, the idion trait is said to be a liability to receive contraries

1

One explanation may be this: they are not vying for the same task, but are doing different jobs. By the general set of traits it is determined that they belong to the common group of being, whereas the idion-trait puts each being in a particular category.

Individuals, Form, Movement: From Lambda to Z–H

141

(4a10–11). This is taken as what is most distinctive of substance, for in no other case could one bring forward anything which would be able to receive contraries (proposition and opinion being dismissed as properly liable to receive contraries in 4a21–b18). What does not satisfy this specific trait cannot be considered as substance. Now, sensible substances notoriously satisfy this condition. But also notoriously non-sensible substances cannot satisfy it, for they cannot be otherwise, as they are unchanging and hence unable to receive contraries. Nonsensible substances are thus ruled out by the idion-trait. This means that Categories restricts itself to examining sensible substances alone, leaving aside all non-sensible substances. It investigates sensible substances in isolation from other substances. And this is a significant divergence between this treatise and Book Lambda. For Lambda clearly has the ambition of, and to some degree succeeds in establishing a connection between these two realms. The conditional in Lambda 1 1069b1–2 ei me¯demia autois arche¯ koine¯ (“if there is no principle common to them”) which closes the general introduction to the inquiry into substance that will be pursued in this book, will eventually prove not to be satisfied, as the argument shows that there is a non-sensible substance such that ek toiaute¯s ara arche¯s e¯rte¯tai ho ouranos kai he¯ phusis, to quote a well-known sentence of 7, 1072b13–14: “the heavens and the world of nature are suspended from such a principle”. It is not easy to explain the isolated treatment of sensible substances in the Categories. It could hardly be the fact that Aristotle denied existence to any substance other than sensible substances, for not only are there documents attesting that he typically admitted the existence of non-sensible substances, but also Lambda 1 shows that their existence is taken for granted, the only question being what their nature is. Ousiai de treis, says Aristotle in Lambda 1; one of them is the non-sensible substance, and there is no doubt about its existence. Now, it may be tempting to suspect in the Categories a sort of Speusippean influence in broad lines, given that one has different domains treated in isolation from one another. Such suspicion must remain, however, only a guess, and the strongly negative attitude Aristotle shows against what he calls the Speusippean episodic ontology warns us to be cautious on this matter.2 In any case, one does find in the Categories an isolated ontology of sensible substances, whereas Book Lambda endeavours to establish that there is sunaphe¯ tis kai hoion koino¯nia pros alle¯la tois te noe¯tois kai toı¯s te¯s phuseo¯s, “a sort of connection and as it were a common ground between objects of reason and the things of nature”, so that they are not

2

But see Krämer, H.-J. 1973: Aristoteles und die akademische Eidoslehre, in: Archiv für Geschichte der Philosophie, 55, 119–187.

142

Marco Zingano

disconnected ho¯sper hekatera kecho¯rismena, “as if each was separated”, to quote twice from Theophrast’s Metaphysics (2, 4a9–12; see also Lambda 10, 1075a16 panta de suntetaktai po¯s, “all things are ordered together somehow”), a treatise which seems to be very close to Lambda. Book Lambda employs a twofold strategy to demonstrate this connection. First, it looks for the common principles of sensible substances. This is done in three main respects. (i) Substance has priority regarding all other categories because of its natural priority: if it is destroyed, the others are destroyed, but an item from the other categories can be destroyed without that substance being destroyed. This is the old Platonic principle of sunanairesis, applied here to the relationship between substance and the other categories. The upshot is that the principles of substance will be in some sense principles of all things, because substances have natural priority over the other categories. But one still has to show what these principles are. A beginning of an answer to this question is (ii) the energeia – dunamis pair, which plays an important role in the argument of Lambda 4–5. I will come back to it in a moment. At the conclusion of Lambda 4–5, it is said (iii) that the principles of all things are the same by analogy, to¯i analogon (1071a26), and in connection with this is expressly mentioned the tripartite scheme of matter, form and privation, to which is added a fourth item, the mover. So far, Book Lambda shows that the common principles stemming from the tripartition matter / privation / form can only have a thin ontological consistency, if any at all. The tripartition matter / form / privation is central to understanding change in the Physics. The gist of its philosophical significance is this. Philosophers used to explain change as if it occurred directly from one contrary to another; they were unable to see that a third item was necessary, something that persists during change, namely matter. Here, this scheme is applied in order to reach a slightly different outcome. The idea here in Lambda is that, although this is my matter, and that is your matter, or, in the same vein, this is my form, and that is your form, there are matter and form as general principles – but only by analogy. The important thing to note here is that such tripartition is not intended to ascribe ontological priority to form and strip individuals of it – quite the contrary, it is dependent upon individuals as primary beings, and presupposes them as such. Later, a second tripartite division will be conflated with it, to wit, the tripartition of matter / form / compound, but again, this second tripartition is not meant to take the place of individuals in their ontological priority, but is dependent on it, as the first tripartition already was. But now comes the second part of this twofold strategy. It is argued that the fourth item, the mover, is an individual and that, at the same time, it plays the role of the common principle for all substances. If it succeeds in

Individuals, Form, Movement: From Lambda to Z–H

143

both parts, Book Lambda will thus have shown that there is a principle connecting all things together, for this principle satisfies the ontological requirement for being a proper substance, as it is an individual (non-sensible) item, and is simultaneously the ultimate principle of the movement of all sensible substances. Chapters 6–10 are thus devoted to demonstrating this principle and explaining how it can be extended to all non-sensible substances. Here I am interested only in its role as the common principle underlying the movement of all sensible substances, which is argued for in chapters 2– 5 of Book Lambda. Before going on, I would like to make two conceptual clarifications. Analogies can be rather expansive, since they point only to a common role played by items that may have very different natures. Typically an analogy refers to a relation of the sort: as A is to B, C is to D, whatever A, B, C or D are. In Lambda, universality is construed by means of analogy, and this is true not only of relations such as fatherhood, and other similarly expansive general terms, but also, and more importantly, of species and related notions. As a consequence, species (and related notions like form) have no ontological thickness. This fits quite well the ontology of individuals argued for in Categories. By contrast, in the central books of Metaphysics, form is ontologically thick. The main argument of Z–H is aimed at establishing ontological priority to form, which is the cause of X being what it is. In the central books, hence, a to¯i eidei relation vindicates the ontological thickness of forms as causes, whereas the same relation in Lambda has its ontological commitment if not barred, at least lessened by its analogical structure. Take for instance deciduous trees: the idea in Lambda would be that their particular causes make them behave in such a way that they all shed their leaves in autumn. This can be seen as a quite modern concept of cause. However, in Z–H, the trees lose their leaves not because particular causes make each of them behave in this way, but because they all share the same form – the form of deciduous tree – which as a cause makes them behave in this way. This is a claim not only for realism, but, more importantly, for essentialism.3 3

In so far as form is construed as principle or cause, ontological thickness or thinness accounts for the main difference between Lambda and Z–H. In Lambda, causes and principles are particular and dependent on the individuals in which they inhere, whereas in Z–H they transcend individuals and vindicate ontological priority. (See Frede, M./Charles, D. (eds.) 2000: Aristotle’s Metaphysics Lambda. Symposium Aristotelicum, Oxford, 26, for the opposite position.) Form will be thus predicated of some individuals, the ones that belong to the same species. In Z 13, though, Aristotle contends that no universal can be a substance, and this contention poses the most serious threat to the form-ontology of Z–H. I will not address this issue here. For a clear formulation of this problem as a dilemma to Aristotle, and a pessimistic opinion about any solution, see Lesher, J. 1971: Aristotle on Form, Substance, and Universals: a Dilemma, in: Phronesis, 16, 169–178.

144

Marco Zingano

My second remark is a reminder that Aristotle, in Gamma and Z–H, distinguishes very carefully between two connected but distinct questions. One question is how to get a unified theory of being, given that being is not a genus. The solution is provided by the doctrine of the pros hen legesthai relation of the other categories towards substance. As is well known, such doctrine gives a central place to substance, which becomes pro¯to¯s on, prior being. Now, in Lambda, substance has also priority over the other categories; however, this is not a pros hen legesthai priority but a natural priority in accordance with the sunanairesis principle, that old Platonic principle. As a matter of fact, the pros hen legesthai relation is conspicuous by its absence from Lambda.4 It is important to note that pros hen legesthai and sunanairetike¯ relations are very different. The latter relation imposes a natural depen4

One does find in Lambda the phrase pollacho¯s legeomena, which is directly connected to pros hen legesthai in the central books, but has no such connection here. As a matter of fact, this expression is quite obscure in Lambda. It appears twice in the concluding lines of Lambda 5. Aristotle there provides a summary of his position, namely, that the causes are in one sense different, but in another sense, “if one speaks universally and analogically” (4 1070a32), they are the same for all. At the end of Lambda 5, resuming the question of whether causes and principles are different or the same for all the categories, he says that de¯lon hoti pollacho¯s ge (mss.: te) legomeno¯n estin hekastou (5 1071a31). Ross unpacks the sentence in his commentary as saying that de¯lon hoti pollacho¯s ge legomeno¯n to¯n stoicheio¯n tauta esti ta stoicheia hekastou, and interprets: “so as long as the names of the elements are used ambiguously, i.e. so long as we say ‘matter, form, privation, mover’ without specifying the particular matter etc., we may say everything has the same elements”. This is possible, and even likely, as immediately afterward one encounters a contrast to the fact that, when these senses are distinguished, the elements are not the same, but different. The problem is that, in this case, the same idea will reappear expressly restated in the next lines, but now as an exception to the case when the elements and causes are distinguished and have thus different causes. I think it is better to take the sentence as meaning something like this: “)concerning the categories*, when the elements and causes are said in different senses, that is, in different categories, there are (causes and elements) to each one (that is, to each category), which, when they are distinguished (now taken one by one), are different, except in some cases (and then come the three cases already mentioned above)”. The other passage with pollacho¯s legetai comes some lines below, in 1071a37, right at the end of Lambda 5. It is not clear either, but I think that, if one sticks to the reading I proposed, one arrives at a meaning which is similar to the meaning this phrase has in many other passages in Aristotle (namely, it refers to categorical dispersion), except that here it comes with no hint of a focal meaning. I take it here to mean that the things first in act are different, that is, the contraries that are not said to be kinds (which include genera above the species, as well as species, all of them being spoken analogically) nor in many senses (= in the highest kinds or categories; see Crubellier, M. 2000: Metaphysics Λ 4, in: M. Frede/ D. Charles (eds.), Aristotle’s Metaphysics Lambda. Symposium Aristotelicum, Oxford, 137 for this reading). In the household-simile at Lambda 10, Aristotle writes that pros men gar hen hapanta suntetaktai (“for all are ordered together to one end”, 1075a18–19), but this is not a case of pros hen legomenon: not every relation that depends upon one and the same thing is a pros hen legesthai relation.

Individuals, Form, Movement: From Lambda to Z–H

145

dence of the other categories upon substance, but does not imply a focal relation nor does it make room for a unified science of being. The pros hen legesthai relation implies natural dependence, and does make room for a unified science of being (see, for instance, Gamma 2 1003b12–14). The second, connected but distinct question is this: given that there are two kinds of substance, sensible and non-sensible substances, how does one relate to the other? Most of Z–H is devoted to finding an answer to this issue. Book Lambda will also offer an answer to this question, but it is – or so I will try to show – a different answer from the one provided by Z–H. Why does Lambda focus on general principles, even if only by analogy? There are several possible answers to this question, but I think the most likely scenario is this. Given that the explanation of change requires the tripartite scheme, Aristotle realizes that he can apply this very scheme in another direction, provided that one adds to it a fourth item, the mover. For on the one hand, one has the internal or enhuparchonta causes, those captured analogically by the tripartite scheme. But on the other hand, there are external causes, the efficient movers, which are principles of the change, but are not elements in the thing changed. Hence principle and element are different notions (4, 1070b23). This is already an important conceptual clarification, for, according to Aristotle, philosophers who preceded him made no systematic distinction between these two notions. Both are causes, but not every cause is internal, for a mover is a cause but is not internal. But there is more. We are still working with a proportional relation for the mover as a cause. Peleus is the father of Achilles, and this man is the father of that man; it is only by analogy that the father is a moving cause. What really moves as father is Peleus, or this man, not the father. One can conceive of a common or universal mover, but merely by analogy. A man begets a man (4, 1070b34), and one can say that anthro¯pos )arche¯* anthro¯pou katholou (5, 1071a21), “man is the principle of man universally”, but only in keeping strictly with the analogical proportion, such that there is no universal man, but only particular men. No ontological thickness is to be attributed to such entities generated by analogy. This may look like a meager result. But it is not. For Aristotle makes now a big move. The notion of a common mover, such as a father, can play no role in connecting all of nature and the heavens due to its ontological thinness. The only way to establish a connection between the realms of sensible and non-sensible substance is to posit the existence of an individual which as first of all things moves all things, to ho¯s pro¯ton panto¯n kinoun panta (4, 1070b35) – an individual which moves all things, as the father is analogically the father of every child. This is the notion of a first mover of everything, which will prove to be the first unmoved mover examined in the second half of Lambda. It is introduced abruptly here, at the end of Lambda 4. Such a

146

Marco Zingano

mover is not derived from just any universal cause; a universal cause like father, however, because it inevitably fails as a reason for connecting everything due to its ontological thinness, paves the way for the idea of a peculiar (individual) mover of everything. This mover – or these movers, as there will be one attached to each celestial sphere – is ontologically correct, in accordance with Lambda’s expectations: it is a particular cause. And it is the cause of the movement of everything. In so far as it is ontologically correct and is the cause of movement for everything, it enables Aristotle to propose a connection between all things. If there is such a connection, the universe is no longer a succession of distinct layers, but becomes a connected whole, and what the first lines of Lambda 1 still envisaged as a possibility, namely the disconnection of these two realms, is finally ruled out. A job, so to speak, has been found for the non-sensible substance(s). Let us look into the principles of sensible substances. At the end of Lambda 5, summing up his arguments, Aristotle first mentions the fourfold distinction, consisting in the hule¯, eidos, stere¯sis tripartition, to which is added the fourth item, the external mover. I will call it the privation-division. They are all universally described only analogically, thus displaying little or no ontological thickness. He then goes on to say that the principles of substances are principles of all other categories, thanks to the sunanairesis principle. Then comes the third and last way of positing principles of all sensible substances. It is the distinction between energeia and dunamis. It is referred to in line 1071a36 as to pro¯ton entelecheiai. It is not clear at first what use can be made of this distinction. It will play a major role for the determination of the nature of the non-sensible substances from Lambda 6 onwards, but, in respect to sensible substances, which are the subject matter of chapters 2 to 5, its role looks more modest. It has to do with priority, either in the sense that there must be something in actuality that suffers the change, or that the mover which causes a change in something else must be previously in actuality. In line 1071a36 to pro¯ton entelecheiai seems to convey both meanings. But another tripartite division comes in. This other tripartition is the matter / form / composite division. I will call it the form-division. We are familiar with the form-division from the Metaphysics’ analysis of substance. In the central books of Metaphysics, form is the first substance, as the ousia of something. The main ontological role is ascribed to form, and form accordingly enjoys ontological thickness. It is the cause why a particular X is what it is. Also, although it cannot but exist immanent in sensible substances, form is seen as tode ti and cho¯riston. It is the latter, cho¯riston, in so far as it can be put separately in the definitional formula (the definition of a thing exhibits it in the definiens; that is, it is separate only in logos). It is a tode ti in so far as it is continuous over time. As a matter of fact, it is more continuous than the composite, for it is neither generated nor does it suffer

Individuals, Form, Movement: From Lambda to Z–H

147

corruption, whereas the composite, which is also continuous over time, lasts only for a determinate period (it can reach eternity only as a species). On the other hand, matter, which also has no coming to be or passing away, is not continuous in a determinate way. There is thus in the central books of Metaphysics a new ontological perspective, different from the ontology of individuals in Categories. The recipient of the title of first ousia is the most evident signal of this shift: in Categories, individuals were first substances; now, in Metaphysics, form is first ousia, and it is accordingly incumbent on it, as it was in Categories incumbent on individuals, to play the major ontological role. Now, mentioning the form-division here in Lambda may be seen as a signal of continuity between this book and Z–H, for in these books the concepts of form and matter are highly operative. But some details in Lambda seem to pose an obstacle to such continuity, for some details point rather towards a discontinuity between the two texts. The first noticeable detail is that the form-division, which typically accompanies the perspective of the central books of Metaphysics, is couched in a strikingly different terminology when it appears in Lambda. One reads this in Lambda 3: “οὐσίαι δὲ τρεῖς, ἡ μὲν ὕλη τόδε τι οὖσα τῷ φαίνεσθαι (ὅσα γάρ ἐστιν ἁφῇ καὶ μὴ συμφύσει, ῦλη καὶ ὑποκείμενον), ἡ δὲ φύσις καὶ τόδε τι, εἰς ἥν, καὶ ἕξις τις· ἔτι τρίτη ἡ ἐκ τούτων ἡ καθ᾿ ἕκαστα, οἷον Σωκράτης ἢ Καλλίας.” (1070a9–13) The problems concern the first two items, matter and form. Jaeger supposes that he¯ men hule¯ tode ti ousa to¯i phainesthai is corrupted, for it is asserted here that matter is tode ti, whereas in H 1 and De anima II 1, matter is tode ti only in potentiality. Jaeger proposes accordingly to add dunamei to the phrase tode ti. Ross says that none of the attempts to amend the phrase have been successful, and proposes instead to take it as it stands and to interpret it as Averroes told us Alexander did, to wit, matter is tode ti only in appearance, that is, owing to a deceptive appearance of unity, given that it displays no real organic unity, no sumphusis. Alexander’s reading looks promising, or at least more promising than taking to¯i phainesthai as “by perception”.5 The second item, the form (he¯ de phusis tode ti eis he¯n kai hexis tis), is said by Ross to be intolerably harsh, and he proposes to place eis he¯n after hexis tis. The result is: he¯ de phusis tode ti ousa kai hexis tis eis he¯n )he¯ genesis* estin: the nature (form), which is tode ti and a (positive)

5

As it is translated in ROTA. In his commentary, Bonitz writes, in a similar vein, that “materia non re vera, sed imaginationi tantum est τόδε τι, quoniam potentiam habet τοῢ γίγνεσθαι τόδε τι” (Bonitz, H. 21992: Commentarius in Aristotelis Metaphysicam, Hildesheim, 476).

148

Marco Zingano

state to which )coming to be, generation* is directed. This is possible, and even likely, so long as one takes this passage as asserting fundamentally the same as what is said in the corresponding passages in Z–H. This is taken for granted by Ross, Jaeger and most commentators. Lindsay Judson is pretty clear about this point: “The claim that the form of a substance is itself a substance is the cornerstone of Aristotle’s mature metaphysics, and one to which almost the whole of Z and H is devoted: its appearance here, with a mere seven words of explanation, is a breath-taking piece of compression.”6 But is it really the same claim? It seems to me that the view that it is the same claim can be challenged. The first thing to notice in this direction is that, although Book Lambda has the energeia-dunamis conceptual pair and employs it in its argumentation, it does not apply it to the form-division, as it typically happens in Z–H. Jaeger and Ross did exactly this, to apply the dunamis-energeia contrast to this tripartition, and consequently they tried to equate one passage with the other. However, the fact is that such a conceptual pair, even if it is available to Aristotle in Lambda, is not applied here to this tripartite division. If one does not read the Lambda passage in the light of the Z–H passages, then there are good reasons to keep the manuscripts’ reading. The text was this (adopting Bekker’s punctuation): “οὐσίαι δὲ τρεῖς, ἡ μὲν ὕλη τόδε τι οὖσα τῷ φαίνεσθαι (ὅσα γάρ ἐστιν ἁφῇ καὶ μὴ συμφύσει, ὕλη καὶ ὑποκείμενον), ἡ δὲ φύσις τόδε τι, εἰς ἥν, καὶ ἕξις τις· ἔτι τρίτη ἡ ἐκ τούτων ἡ καθ᾿ ἕκαστα, οἷον Σωκράτης ἢ Καλλίας.” (1070a9–13) The meaning looks roughly to be this. There are three kinds of substance: matter, which is tode ti only by appearance or deceit; nature or form, which is tode ti, the towards-which, that is, a positive state )in contrast with privation*; and individuals, such as Socrates and Callias, which are composites of matter and form. What is remarkable here is the conflation of both schemes, the privation-division and the form-division. Such conflation becomes evident by the phrase eis he¯n, translated as “the towards-which”, that which gives the direction of the change in the privation-division, explained as a positive state, in contrast to privation (which is its opposite). It is its status as the point towards which the change occurs that justifies here the attribution of tode ti to form. For in a change there is something out of which

6

Judson, L. 2000: Formlessness and the Priority of Form: Metaphysics Zeta 7–9 and Lambda 3, in: M. Frede/D. Charles (eds.), Aristotle’s Metaphysics Lambda. Symposium Aristotelicum, Oxford, 130.

Individuals, Form, Movement: From Lambda to Z–H

149

something becomes something other, and in this scheme privation, matter and form are always correlated with an individual (every change is a change of an individual, whose matter persists as a property changes from privation to form, like a change from cold to hot). As a matter of fact, form in the privation-division refers to any of the four categories involved in the process of change (quantity, quality, place and generation), whereas the form-division in Z–H stresses the fact that form does not refer generically to any cause in a change, but stands specifically for the (formal) cause of X being the substance it is, this cause being itself a substance. It turns out then that, if one resists the temptation to import the Z–H reading into this passage, the conflation of form-division and privation-division becomes apparent, and, more significantly, one can also see that the form-division is dependent on, or is absorbed into the privation-division. Another case of this conflation seems to occur in 1070b10–14, a passage with which Aristotle begins his explanation of how in one sense the elements are the same, but, in another, they are different. The text is this: “Or, as we put it, in a sense they (= the elements) are and in a sense are not (the same), e.g. perhaps the elements of perceptible bodies are, as form, the hot, and in another sense the cold, which is the privation; and, as matter, that which first and of itself is potentially these; and both these (tauta = matter and form/privation) are substances and also the things composed of these (ta ek touto¯n = matter and form), of which (= the composites) these are the principles.” “ἢ ὥσπερ λέγομεν, ἔστι μὲν ὥς, ἔστι δ᾽ ὡς οὔ, οἷον ἴσως τῶν αἰσθητῶν σωμάτων ὡς μὲν εἶδος τὸ θερμὸν καὶ ἄλλον τρόπον τὸ ψυχρὸν ἡ στέρησις, ὕλη δὲ τὸ δυνάμει ταῦτα πρῶτον καθ᾽ αὑτό, οὐσίαι δὲ ταῦτά τε καὶ τὰ ἐκ τούτων, ὧν ἀρχαὶ ταῦτα).” (Lambda 4 1070b10–14) This would sound very surprising indeed, if the form-division is to be understood in terms of its meaning in Z–H. For there is no discussion in Lambda of how the principles of the composites could themselves be substances, nor is there any trace in Lambda of a distinction between ousia and ousia of X – but such a distinction is required if forms and composites are to count as substances. Moreover, if the form-division is to be taken as carrying the meaning it has in Z–H, it would be in open conflict with the analogical approach, whose main point consists in avoiding any thick ontological commitment to principles, causes and elements, whereas the form-division in Z– H is ontologically loaded with thick forms. But if one abandons the Z–H reading and refers this passage back to what has been said in Lambda 3 1070a9–13 (matter is substance only in appearance and form is the nature

150

Marco Zingano

to which change is directed), it poses no insurmountable threat for an ontology based fundamentally on individuals. For the form-division depends upon, or is absorbed into the privation-division, such that form and matter are only analogically referred to as substances, dependent obviously on the real individual substances. Such dependence can be seen here from the fact that form, in this passage, refers to two qualities, hot and cold, as it typically does in the privation-division of Physics. Another way to express the same point is to note that this passage of Lambda 4 shows that the form-division is the privation-division when the tripartition is applied not to change in general, but to the specific change of generation and corruption: in the case of coming to be and passing away, the values of the tripartite scheme are matter, form, and the composite. If this is correct, one may ask why Aristotle brought in the tode ti analysis, if the form-division is to be taken as subservient to the privation-division, as the result of the application of the privation-division to generation and corruption, with no intent to accord form any ontological thickness. I think the main reason was to distinguish between natural substances and artefacts. In the case of a house, its form is either not separable from the house itself or, if it is, it is in the mind of the builder, as he possesses the art of building. In either case, though, there is no proper generation or corruption of it, but only construction or destruction. In a treatise of sensible substances, it is important to provide criteria for distinguishing between natural products and artefacts. And this is precisely what Aristotle does in Lambda 3 1070a13– 20, a passage which immediately follows 1070a9–13 quoted above. When Aristotle couples the form-division to the privation-division, he has no ambition of altering the ontological status of forms (as he has in Z–H). His interest is more limited: he wants to restrict the status artefacts may enjoy within a general theory of sensible substances. I have postponed until now any discussion of the key-term tode ti, transliterating it instead of translating it. Now, in Categories, this phrase clearly refers to individuals, and exclusively to individuals. In Categories, first substances, like Callias and Socrates, are the only cases of tode ti, since the species and genera that are said of them are not, strictly speaking, tade tina (cf. Categories 5, 3b15). The phrase, as is well known, is used in Metaphysics not only of individuals, but also of forms. How did that become possible? Ross says this, in a note to D 8, where the form is said to be tode ti: “The form is said to be tode ti. It is more often the concrete unity of matter and form that is so described, but form is the element that gives individual character, and so the form is sometimes called tode ti.” (I 310, ad 1017b25)

Individuals, Form, Movement: From Lambda to Z–H

151

It is true that form gives character to a piece of matter, but I am not convinced that this note captures the main reason why Aristotle treats form as tode ti in the central books of Metaphysics. When Ross says that form is “sometimes” called tode ti, he probably means that form is derivatively or secondarily tode ti. But this is false: not only does Aristotle repeatedly say that form is tode ti, but such a claim is also crucial for his new argument. Lindsay Judson is right when he underlines that the claim that the form of a substance is itself a substance is the cornerstone of Aristotle’s metaphysics in the central books. Now, the main traits of being a substance are tode ti and separability, according to Z 3, 1029a29. Hence, if form is to be a substance, it has to be in some relevant sense tode ti. But is tode ti individuality, as it clearly was in Categories? If this is the case, referring to form as tode ti in Metaphysics would mean that forms are particular. This is one of Patzig and Frede’s arguments for taking forms as particular: “Aristoteles bezeichnet die Form wiederholt als ein “Dies von der Art”. Ein “Dies von der Art” aber ist nach Aristoteles ein Individuum und der Zahl nach eines (vgl. Cat. 5, 3b10–14). Folglich ist die Form ein Individuum und dem eigen, dessen Form sie ist.”7 In Categories, the expression tode ti certainly refers to individuals, and exclusively to individuals, but this is the case not because tode ti as such stands for an individual, but because individuals are the only ontological items that can possibly be envisaged as having enough thickness to persist over time. I take it that tode ti refers to something that displays a unity continuously over time.8 Individuals do this, and quite well, for they are substrata for changes. But they have only momentary stability. The crucial move made in the central books of Metaphysics is to posit that, over and above individuals, there is something which, although it is immanent to them, has perfect stability, namely, form – exactly that nature that is transmitted from individuals to other individuals in the natural process of coming to be. When this move is made, form is lifted up to the role of first ousia, a role previously reserved to individuals.9 Individuals continue to be the pieces of the furniture of the

7 8

9

Frede, M./Patzig, G. (eds.) 1988: Aristoteles, Metaphysik Z’, München, 52. If this captures the main point of the notion of tode ti, then the Timaeus becomes a very significant text in the history of this notion. For in 49D it is said that only that which has tina bebaiote¯ta (“some stability”) can be referred to as touto (“this”) and not simply as toiouton (“such”), thus making stability the proper trait of being referred to as something and not just of such a sort. Matter is ruled out as a good candidate for substance because it has only apparent unity and is not separate (see Z 3 for disclaiming that matter is substance as such; H 1 will remind us that matter is substance, but only in potentiality).

152

Marco Zingano

world, they continue to be here and there and to move around, but there is something they carry that is ontologically prior to them, namely, form, which best satisfies the requirement for stability. There is a passage in Lambda that has been taken as a moment in which Aristotle clearly commits himself to positing forms as particular. This passage is Lambda 5, 1071a27–29: “The (causes and elements) of things in the same species are different, not in species, but in the sense that the causes of different individuals are different, your matter and form and moving cause being different from mine, while in their universal formula they are the same.” “καὶ τῶν ἐν ταὐτῷ εἴδει ἕτερα, οὐκ εἴδει ἀλλ᾽ ὅτι τῶν καθ᾽ ἕκαστον ἄλλο, ἥ τε σὴ ὕλη καὶ τὸ εἶδος καὶ τὸ κινῆσαν καὶ ἡ ἐμή, τῷ καθόλου δὲ λόγῳ ταὐτά.” (Lambda 5 1071a27–29) By now, we should be familiar with the language of Lambda, a language inherited from Categories, ontologically oriented to individuals, and attributing to species and genera only analogical existence, if any at all. Some lines above, reference is made to idion eidos, the peculiar form (1071a14), the form I have, or the form you have. These forms are a set of properties that define what you are, and what I am. It is only by analogy that one gets the idea of a common species, shared by you and me. Not surprisingly, this passage deals with the status of matter and forms as such, the two components of the individual, as strictly dependent on an analogical relation with the particular matter and form of each one of us. It says thus that in their universal formula the individuals are the same. As one knows from the beginning of Lambda 4, this means that they are the same by analogy, for the causes and elements will be the same only if one speaks universally, that is, analogically (1070a32). Hence, this passage clearly pleads for particular forms, if any thickness is to be attributed to them – and how could it be otherwise, if only individuals have ontological thickness? But can this lesson of Lambda 5 – I have my own form, you have your own form – be transferred to Z–H in terms of a doctrine of particular forms? There are reasons for doubt. Lambda and Z–H deal with distinct projects; in the former, individuals occupy the central position; in the latter, form plays the major ontological role. Given Lambda’s adherence to the ontology of individuals of Categories, it cannot but propose an analogical approach concerning any universal, including common forms, with all its consequences. In Z–H, by contrast, form is ontologically thick, it has ontological priority, and it can be said of many individuals. The last sentence may sound controversial in light of Z 13 and its ban on universals. To avoid getting trapped in this discussion, let me insist that,

Individuals, Form, Movement: From Lambda to Z–H

153

as a matter of fact, in Z–H form is characterized intensionally, with no extensional commitment (apart from being the form of at least one individual). Form is aition pro¯ton tou einai, to quote Z 17, 1041b28: it is the primary cause of a thing being what it is. From this nothing follows about the extensionality of a form. Perhaps a form F is a form of only one thing, and not by accident – as in the case of sun or moon. Perhaps a form G is predicated of several items, as the form of man is predicated of all of us. Characterizing form as ontologically thick envisions it only from the point of view of intensionality, irrespective of any consideration about extension. As such, it is not said more of many things than of one thing alone (whereas universals are naturally said of many items, and only accidentally of only one). What does matter instead is its being indivisible (Z 8, 1034a8: atomon gar to eidos), since it is the cause of an item being the unity it is. I do not want to say that the ontology of Z–H is exempt from aporias – to the contrary, it is not, as Z 13 dramatically reminds us, for this chapter stresses that no universal is a substance, when Z had argued that form is first substance and is supposed to be predicated of many items. I just want to stress that it seems to me better to take Lambda as half-way between the isolated ontology of individuals vindicated by Categories and the ontological priority attributed to forms by Z–H. Lambda tries to connect sensible and non-sensible substances by means of a unique first mover, a single particular principle of everything. Such an attempt makes a step towards the unified doctrine of substance Z–H argues for, but it still adheres to the ontology of individuals of Categories, and cannot but attribute analogical existence to anything other than individuals. Lambda is in this respect very close to Theophrast’s Metaphysics. It is important to stress that, in Lambda, non-sensible substances are not brought in gratuitously. It is not the case that one has a sort of theology in the second half of Lambda just because there happen to exist non-sensible substances, so that one has to deal with them anyway in theoretical science. Non-sensible substances have a clear connection to sensible substances, for in order to fully explain the movement we observe among sensible substances, we have to appeal to them as the ultimate source of movement. There is thus a close connection between these two realms. The discussion of sensible substances carried out in the first half of Lambda is not merely introductory to the second part, let alone a simple summary of physics; it is rather that, Lambda being a treatise on substance, no complete analysis of its subjectmatter can be attained without discussing those separate substances that function as the ultimate cause of the movement of all sensible substances. Hence, the second half of Lambda, Aristotle’s celebrated theology, has been rightly described as being “about the separate unchanging substances which

154

Marco Zingano

turn out to be unmoved movers”, for the cause of movement is the hinge between both realms.10 Z–H’s analysis of substance also argues for a close connection between both realms. In several passages, it is said that the whole discussion of sensible substances will shed light on the nature of non-sensible substances (e.g., Z 11, 1037a11–12 and Z 17, 1041a8–9). This points to the fact that the ultimate cause of movement is the unmoved mover(s). There is, though, an important difference between the two treatises on substance in this respect. Book Lambda takes it for granted that non-sensible substances exist; all its efforts are devoted to show that such substances cannot but be unmoved movers. It thus denies any plausibility to Plato’s Ideas, the main Academic candidates for non-sensible substances, for, no matter what Ideas are, they can play no role in the explanation of movement. We find a very different story in Z–H. We are now methodologically restricted to sensible substances, for they are the substances everyone agrees to, and any enquiry has to begin with what was previously agreed on (as argued for in Z 2). The discussion about sensible substances will show that something immanent to the individuals, to wit, form, has ontological priority and thickness. Form is thought of in contrast to matter, the two being the components of the individual, typically seen as the composite of matter and form. The upshot is that the form of sensible substances, due to its ontological thickness as a substance which is the cause of individuals being what they are, makes room for the existence of a substance of another nature; it is a sort of placeholder for non-sensible substances, which would not only be separable from matter, but would exist separate from any matter. Hence, if, based on independent reasons, one is compelled to posit the existence of such non-sensible substances, such a hypothesis will prove not to be problematic, for it has already been shown that an immaterial being (the form of sensible substances) has ontological thickness and priority. Now, according to Aristotle, the existence of an eternally ordered movement of sensible substances requires such a supposition. Hence one can go further and establish a domain of non-sensible substances. According to Z–H, the unmoved movers turn out to be separate unchanging substances, for, in order to play the role of ultimate source of an eternally ordered movement, they have to be unchanging, always in actuality, and so forth. Non-sensible substances are no longer taken for granted but are shown to play the role of unmoved movers in order to connect the whole universe. Z–H argues on behalf of them as possible substances, in so far as it attributes ontological priority to an immaterial substance, form. Given this demonstra-

10

Frede, M. 2000: Introduction, in: M. Frede/D. Charles (eds.), Aristotle’s Metaphysics Lambda. Symposium Aristotelicum, Oxford, 37.

Individuals, Form, Movement: From Lambda to Z–H

155

tion, non-sensible substances can be the case, if independent reasons posit their necessarily being the case. I think it is pretty fair to describe very cursorily the difference between the two treatises on substance as follows: in Lambda, separate unchanging substances – the non-sensible substances – turn out to be unmoved movers; in Z–H, unmoved movers turn out to be separate unchanging substances, the searched-for non-sensible substances. We have not, in Z–H, the part that corresponds to Lambda 6–10; but if we did, it would probably unfold a proof of unmoved movers turning out to be separate unchanging substances, the reverse order of what one actually finds in Lambda 6–10.

Bibliography Bonitz, H.: Commentarius in Aristotelis Metaphysicam, Hildesheim 21992. Crubellier, M.: Metaphysics Λ 4, in: M. Frede/D. Charles (eds.), Aristotle’s Metaphysics Lambda. Symposium Aristotelicum, Oxford 2000, 137– 160. Frede, M.: Introduction, in: M. Frede/D. Charles (eds.), Aristotle’s Metaphysics Lambda. Symposium Aristotelicum, Oxford 2000, 1–52. Frede, M./Patzig, G. (eds.): Aristoteles, Metaphysik Z’, München 1988. Frede, M./Charles, D. (eds.): Aristotle’s Metaphysics Lambda. Symposium Aristotelicum, Oxford 2000. Judson, L.: Formlessness and the Priority of Form: Metaphysics Zeta 7–9 and Lambda 3, in: M. Frede/D. Charles (eds.), Aristotle’s Metaphysics Lambda. Symposium Aristotelicum, Oxford 2000. Krämer, H.-J.: Aristoteles und die akademische Eidoslehre, in: Archiv für Geschichte der Philosophie, 55 1973, 119–187. Lesher, J.: Aristotle on Form, Substance, and Universals: a Dilemma, in: Phronesis, 16 1971, 169–178.

God as Pure Thinking. An Interpretation of Metaphysics Λ 7, 1072b14–26 1

STEPHAN HERZBERG

In Metaphysics Λ 6–7 Aristotle gives a more precise elaboration of several features which characterize the mode of being of the first principle (eternal, pure actuality, free of matter, metaphysically simple; moving without itself being moved, in the highest degree intelligible and desirable; pure thinking, best and eternal life). In this article I will be concerned exclusively with the feature of thinking in the sense that the first principle is not only the highest object of thinking, but it is itself pure thinking. When at Met. Λ 7, 1072b14– 26 Aristotle ascribes to the first principle a certain kind of life, and qualifies this more specifically as the activity of reason in the mode of contemplation, he argues for this feature in a significantly different way from the previous characteristics. In what follows I will develop and defend an interpretation of this passage according to which we are here dealing with an argument from analogy. On my reading, Aristotle wants to show in this passage how it is that we can determine the ‘pure actuality’ of the first principle based on the highest realization of human thought. He succeeds in doing so by showing that in human thinking there is a special ‘structure of actuality’, which, however, due to his ontological perfection has to be present in God in a higher form. I will show that, contrary to the usual reading of this passage, the interpretation developed in this essay requires fewer additional assumptions, is closer to the text and, in general, can better accommodate the unique topic discussed at Met. Λ 6–7.

1

The following article is a reworked and shortened version of chapter 4 of Herzberg, S. 2013: Menschliche und göttliche Kontemplation. Eine Untersuchung zum bios theoretikos bei Aristoteles, Heidelberg. I thank Manfred Weltecke for translating my text into English.

158

Stephan Herzberg

I. The Thesis (1072b14–16) “And its life is such as the best which we enjoy, and enjoy for but a short time. For it is ever in this state (which we cannot be)”.2 (1072b14–16) This is a thesis which consists of two parts: (1) the first principle leads a life (διαγωγή), i.e., the mode of its actuality or activity (ἐνέργεια)3 is identified as a form of life,4 and (2) this form of life is comparable with the best of our own, which we, however, can only enjoy for a short time. Both parts of the thesis deserve a detailed elucidation. By beginning the sentence with διαγωγή Aristotle wants to draw our attention to the fact that saying of the first principle, on which heaven and nature depend, that it leads a certain life, is an astonishing claim indeed.5 What do we know, at this point, about the mode of being of this first principle? In order to guarantee the eternal and uninterrupted motion of the first heaven, the first principle must be eternal and, in addition, in actuality. More precisely, it must be in actuality in the special way that it is not the actuality of an underlying potentiality. If it were the actuality of an underlying potentiality, then it would also be possible that its potentiality might, at times, not be actualized and would, therefore, not generate any motion. Yet since the motion has always existed, Aristotle reaches the conclusion that this principle has to be actuality according to its own essence, i.e. its essence is actuality. So it must not possess any kind of potentiality.6 Furthermore, being an actus purus of this kind the first principle is without matter, inasmuch as matter is the principle of potentiality. These features or predicates of perfection form a first series of predicates by which the essence of the first principle is characterized more precisely. Aristotle derives these characteristics in a purely conceptual way (λόγῳ):7 his starting point is the question of how the nature of

2 3 4 5

6 7

Barnes, J. 1984: The Complete Works of Aristotle, Princeton. Volume II. All translations are from Barnes unless otherwise indicated. See Horn, C. 2002: In welchem Sinn enthält Metaphysik Lambda eine Theologie? In: Jahrbuch für Religionsphilosophie, 1, 39 on this double meaning of ἐνέργεια. Cf. also E. N. X 4, 1175a12: ἡ δὲ ζωὴ ἐνέργειά τις. See Laks, A. 2000: Metaphysics Λ 7, in: M. Frede/D. Charles (eds.), Aristotle’s Metaphysics Lambda. Symposium Aristotelicum, Oxford, 232: “Whether one renders the term by ‘way of life’ or, perhaps more neutrally, by ‘occupation’, something more than mere existence is presupposed, namely a life.” See also Bordt, M. 2006: Aristoteles’ Metaphysik XII, Darmstadt, 116 f. Cf. Met. Λ 6, 1071b19 f.: δεῖ ἄρα εἶναι ἀρχὴν τοιαύτην ἧς ἡ οὐσία ἐνέργεια. See also Λ 7, 1072a32; Λ 8, 1074a35 f. Cf. Bordt (see note 5), 101.

God as Pure Thinking. An Interpretation of Metaphysics Λ 7, 1072b14–26

159

something that can guarantee an eternal and continuous motion must be conceived. This then leads him to the concept of ‘pure actuality’. This concept describes the essence of the first principle of motion and the other (two) characteristics are connected with this concept via a conceptual necessity.8 Within a second series of predicates the first principle is qualified further; more precisely, it is thus qualified with regard to the eternal and uninterrupted motion of the first heaven. The first feature in this series characterizes the first principle as moving other things, while it itself is unmoved.9 In support of this claim Aristotle gives the following argument: “There is therefore also something which moves them. And since that which is moved and moves is intermediate, there is a mover which moves without being moved, being eternal, substance, and actuality” (Λ 7, 1072a23–26).10 Aristotle develops this argument further at Phys. VIII 5. Here he proves the necessary existence of some unmoved moving being.11 The eternal and immaterial substance, which essence consists in actuality and which is identified as unmoved, moves in the way an object of desire and thought moves other things, and this is the second feature in this series of predicates. For objects of desire and thought move as mere final causes, without being moved themselves. In order to justify that the first principle does de facto move other things in this way,12 Aristotle tries to show that the primary instances of what is desirable and what is thinkable are identical: The primarily desirable is not that which just 8

9

10 11 12

To this first series other characteristics can be added, e.g. that owing to its pure actuality this principle “can in no way be otherwise than as it is” (Λ 7, 1072b8) and that it is “absolutely necessary” (b13); or the characteristics that this principle exists “separate from sensible things” (1073a4 f.), is “without parts and indivisible” (1073a6 f.; see also Λ 9, 1075a6 f.) and “one both in formula and in number” (Λ 8, 1074a36 f.; I owe this last point to Lloyd Gerson). Following Horn (see note 3) one may indeed call these “divine attributes” (32 f.) or also “predicates of perfection” (41). Bordt, M. 2011: Why Aristotle’s God is Not the Unmoved Mover, in: Oxford Studies in Ancient Philosophy, 40, 91–109 is right to point out that to be an unmoved mover is not an essential feature of the first substance, but merely a relational one, i.e., it has this feature only in relation to the realm of the perceptible and changeable substances (see previously Menn, S. 1992: Aristotle and Plato on God as Nous and as the Good, in: Review of Metaphysics, 45, 545). The first substance is later identified with God (1072b30), not insofar as it is an unmoved mover but insofar as it leads the best kind of life, which is the activity of reason. At this point it becomes clear that the second series of predicates is joined with the first series (ἀίδιον καὶ οὐσία καὶ ἐνέργεια οὖσα). There is no agreement as to which passage of Phys. VIII 5 Aristotle is referring here. See Laks (see note 5), 217–219, for more details on this question. See Bordt (see note 9), 107: “Aristotle is not interested only in some general claim to the effect that there are cases where something unmoved can move something else; rather, he is aiming to show that one very particular state of affairs actually obtains, namely that the unmoved mover in fact does move (what it moves) as an object of thought and desire”.

160

Stephan Herzberg

appears good to someone (τὸ φαινόμενον καλόν), but that which is real good (τὸ ὂν καλόν). The latter is not an object of appetite (ἐπιθυμία), but of rational desire or wish (βούλησις). This, however, implies a judgment of reason.13 The theoretical basis for such a definition of the πρῶτον ὀρεκτόν is Aristotle’s objective theory of the good:14 “For the apparent good is the object of appetite, and the real good is the primary object of wish. But desire is consequent on opinion rather than opinion on desire: for the thinking is the startingpoint” (1072a27–30). The paradigm among the objects of thought is determined by means of what is “in itself thinkable”, that is by means of a positive series of concepts (συστοιχία).15 In this positive series substance (οὐσία) comes first and within this genus of being the simple and fully actual substance (ἡ ἁπλῆ καὶ κατ’ ἐνέργειαν: 1072a30–32) comes first.16 The first intelligible, i.e. the highest object of thought, is thus the immaterial and essentially actual substance, which, as such, has the highest ontological determination (or ‘positivity’). In this positive systoichia we also find the καλόν, i.e. that which is desirable for its own sake and laudable.17 The first principle of motion is always the best or it is at least that which is analogous to the best (1072a34– b1).18 The unmoved mover, as the highest object of thought and desire, thus moves everything else by being its ultimate final cause. It does this solely by the attractiveness of its ontological perfection. As an absolute unchangeable and self-sufficient telos,19 which is separate from perceivable reality, it moves the first heaven by being loved (ὡς ἐρώμενον) and through this motion everything else is moved (1072b3 f.).20 Therefore Aristotle can say that heaven and 13 14 15

16

17 18

19

20

Cf. De an. III 9, 432b5–7; III 10, 433a24–30; Rhet. I 10, 1368b36–1369a4. Cf. Horn (see note 3), 35 f. On the origin of this ontological ordering instrument in the Old Academy see Krämer, H.-J. 1964: Der Ursprung der Geistmetaphysik. Untersuchungen zur Geschichte des Platonismus zwischen Platon und Plotin, Amsterdam, 153–159. With this, the series sketched in Met. Λ 1, 1069a20 f. (τῷ ἐφεξῆς) is differentiated further; in this passage Aristotle is merely dealing with οὐσία as the primary being as opposed to quality, quantity, motion, and negation. For the term καλόν see Met. M 3, 1078a31 f.; E. E. VIII 3, 1248b18–20. For this identification of the best with the highest object of thinking, see also Met. Α 2, 982b6 f.: τοῦτο δ’ ἐστὶ τἀγαθὸν ἑκάστου, ὅλως δὲ τὸ ἄριστον ἐν τῇ φύσει πάσῃ. The unmoved mover, as the primary object of wisdom, is also what is best in the entire cosmos. Cf. Met. Λ 7, 1072b2 f.; E. E. VIII 3, 1249b15 f. For Aristotle’s distinction between οὗ ἕνεκα τινός (the absolute aim) and οὗ ἕνεκα τινί (the beneficiary) see Gaiser, K. 1969: Das zweifache Telos bei Aristoteles, in: I. Düring (ed.), Naturphilosophie bei Aristoteles und Theophrast, Heidelberg, 97–113. The platonic background is emphasized by Richardson Lear, G. 2004: Happy Lives and the Highest Good. An Essay on Aristotle’s Nicomachean Ethics, Princeton, 79: “Aristotle’s Prime Mover is an object of desire of a special sort: It is an object of Platonic love. As such, it is a final cause as an object of aspiration, imitation, or approximation. Such an end may be, in principle, unchangeable as well as separable from the physical world.”

God as Pure Thinking. An Interpretation of Metaphysics Λ 7, 1072b14–26

161

the world of nature depend (ἤρτηται: 1072b13 f.) on such a principle. Yet the beings below the first heaven are also related to the first principle as their ultimate final cause. For every being tries to partake in the eternal and divine to the greatest extent possible.21 This striving towards the divine, which we find in the entire cosmos, is manifest in the fact that every being imitates the divine substance, i.e. its perfect (eternal and pure) actuality, according to the possibilities inherent in its own essence.22 Thus the entire cosmos is revealed as a universal interrelated whole that imitates the divine way of being.23 For the individual being to imitate the divine actuality in its own specific way is for it to realize its own nature in the best possible way.24 Through these two series of predicates we now know that the first principle of motion is a special kind of substance which, on the one hand, is essentially actuality (from which other predicates, e.g., that it is immaterial, invariant, necessary, indivisible, simple etc. may be derived) and, on the other hand, that it is something which is, in the highest degree, intelligible and desirable and which, thus, moves everything else as a principally unmoved and separate substance and as their final cause. At this point one might think that the characterization of the mode of being of the first principle is completed. The open questions which Aristotle had raised with regard to this third kind of substance, at least those pertaining to its existence and essence,25 now seem to have been clarified. It has been shown that an eternal and unmoved substance necessarily exists and that its essence is pure actuality.

21 22 23

24 25

Cf. De an. II 4, 415a29 f.: … ἵνα τοῦ ἀεὶ καὶ τοῦ θείου μετέχωσιν ᾗ δύνανται· πάντα γὰρ ἐκείνου ὀρέγεται … Cf. E. N. X 7, 1177b33 f.; De an. II 4, 415a26–b7; De gen. an. II 1, 731b24–732a11; De gen. et corr. II 10, 337a1–7, Met. IX 8, 1050b28 f. Cf. Sedley, D. 2007: Creationism and Its Critics in Antiquity, Berkeley–Los Angeles–London, 171: “It is, in short, scarcely an exaggeration to say that for Aristotle the entire functioning of the natural world, as also that of the heavens, is ultimately to be understood as a shared striving towards godlike actuality.” According to Sedley, Aristotle manages to harmonize two contrasting motives by having recourse to the platonic view of God as the highest object of imitation: that God is a transcendent substance separate from the world and that God is the highest explanatory principle of reality. See also Menn (see note 9), 573. Cf. Richardson Lear (see note 20), 86: “… when we approximate the divine to the extent possible for us, we realize our own nature.” Cf. the fourth aporia (Met. III 1, 995b13–18; III 2, 997a34–998a19), where Aristotle states that it is necessary to clarify whether there are other substances separate from those which we can perceive and whether these exist in one or several kinds. Concerning the question about the existence and mode of being of the separate substance see also Met. VII 2, 1028b27–32 (πότερον εἰσί τινες παρὰ τὰς αἰσθητὰς ἢ οὐκ εἰσί, καὶ αὗται πῶς εἰσί, καὶ πότερον ἔστι τις χωριστὴ οὐσία …) and Λ 1, 1069a33 f.

162

Stephan Herzberg

When at Met. Λ 7, 1072b14–30 Aristotle qualifies the ‘pure actuality’ of the first principle more precisely as life and ascribes a very specific kind of life to the first principle, i.e., the activity of thinking, which – as will emerge – this principle is according to its essence (1072b27),26 this is, in my view, not in line with the characteristics mentioned up to this point. Whereas the characteristics of the ‘first series’ were “conceptually” derived from the essential predicate of ‘pure actuality’ and whereas such a method of deduction could, in principle, also be given for the characteristics of the ‘second series’,27 this is impossible in the case of the characteristics of life and the activity of reason. First, life is only a certain kind of actuality among others (E. N. X 4, 1175a12: ἡ δὲ ζωὴ ἐνέργειά τις). Moreover, “life” is predicated in many ways (De an. II 2, 413a22; Top. VI 10, 148a26–31). The concept of actuality or activity (ἐνέργεια), which describes the essence of the unmoved mover, is an analogical concept which is applied in entirely different areas of being (Met. IX 6, 1048b6–9). Thus, it is, in principle, thinkable that the first principle – as an immaterial, unmoved and separate substance – could have an entirely different mode of actuality from the ones familiar to us, i.e., the mode of actuality which is life in the sense of the activity of reason. In E. N. X 8 Aristotle says that the activity of God “surpasses all others in blessedness” (1178b21 f.: ὥστε ἡ τοῦ θεοῦ ἐνέργεια, μακαριότητι διαφέρουσα …). In the famous fragment from the lost treatise “On prayer” Aristotle mentions the two options that the God “is either intellect or something more beyond intellect” (fr. 49 Rose).28 For Aristotle it is precisely the complete ontological determination and separation – which the first principle has on account of its ‘pure actuality’ – that make it into something which simpliciter or “by nature” is most known. Yet because of this the first principle is, from the perspective of our finite knowledge, which begins with perception, most remote.29 As is well known, Aristotle compares the relation of our reason to the things “which are by nature most evident of all” to the relation that the eyes of bats have to daylight (Met. II 1, 993b7–11). This epistemic distance of the first principle results from its ontological perfection or transcendence, even if, for Aristotle, this transcendence does not go as far as Plato’s ἐπέκεινα 26

27 28

29

God does not have an intellect, which he exercises, for then it would be possible, in principle, that this intellect might at one time not be actualised. Rather, God is the activity of the intellect itself and thus pure thinking or activity of reason. Cf. Menn (see note 9), 561 f.; Bordt (see note 9), 92. One could deduce the fact that God is unmoved also from the immateriality and see the highest intelligibility grounded in the complete actuality qua determination. ὁ θεὸς ἢ νοῦς ἐστὶν ἢ ἐπέκεινά τι τοῦ νοῦ. Like Oehler (1997, 60 f.) I take this alternative to be strictly disjunctive. The second alternative is a terminological reference to the Platonic ἐπέκεινα τῆς οὐσίας at Pol. 509b. Cf. Anal. Post. I 2, 71b33–72a5; Met. VII 3, 1029b4–12.

God as Pure Thinking. An Interpretation of Metaphysics Λ 7, 1072b14–26

163

τῆς οὐσίας: the first principle is οὐσία in the absolute primary sense. It manifests in a perfect and thus also paradigmatic way what it means to be, and, as such, it belongs – in contrast to Plato – to the logical space of being. On account of the transcendence of the first principle it is understandable that Aristotle expresses himself carefully in part (ii) of the thesis quoted above: “And its life is such as the best which we enjoy, and enjoy for but a short time” (1072b14–16). Here our best way of life serves as an element of a comparison (οἵα ἡ ἀρίστη).30 Therefore, Aristotle thinks that we can take the highest form of human life which we can enjoy as our departing point when we want to qualify the kind of life the first principle enjoys. In my view – and this is crucial for a proper understanding of this passage – we are dealing here with an assumption which stands in need of justification. This need of justification follows from the ontological perfection of the first principle we discussed earlier and from the resulting ‘epistemic distance’ from us. The justification for this thesis is developed in section 1072b18–24. According to my own reading of this passage, Aristotle claims to have shown why we are entitled to use the “to us better known” highest activity of human beings, contemplation (θεωρία),31 in order to characterize the perfect actuality of God more closely. Most commentators overlook that at 1072b14 f. Aristotle expresses a thesis which calls for a more detailed justification. According to the standard reading, Aristotle simply assumes that the mode of life of the first principle is of the same kind as our best form of life, i.e. θεωρία, without offering any further arguments for this assumption.32 When engaged in contemplation human beings exercise, for a short time, exactly the same kind of activity which God exercises at all times. On the basis of this assumption in 1072b18–24 Aristotle is supposed to be in a position to illustrate or explain divine thinking with reference to human thinking.33 However, against 30

31 32

33

Cf. Laks (see note 5), 233. In this context Laks rightly points out that this leaves it open whether this is an identity or a mere similarity. See also Kosman, A. 2000: Metaphysics Λ 9: Divine Thought, in: M. Frede/D. Charles (eds.), Aristotle’s Metaphysics Lambda. Symposium Aristotelicum, Oxford, 309: “Aristotle there sets out to give us an idea of the mode of existence of this being, known up to now only as a principle of activity, and he does so by means of a figurative account that relies upon simile to convey to the reader that mode of existence”. Cf. E. N. X 7, 1177a18; X 8, 1178b7–32. Cf. for example Ross, W. D. 1924 (ed.): Aristotle’s Metaphysics. A Revised Text with Introduction and Commentary, Oxford II, 378: “… Aristotle assumes that it must be such as the highest actuality or activity that we know, viz. νόησις, immediate or intuitive knowledge.” Cf. Bordt (see note 5), 116: According to Bordt, Aristotle can proceed in this way because there is no categorical difference between the rational activity of human beings and the rational activity of the first ousia. Ibid. 118: According to Bordt, what holds for human reason also holds with the exception of continuance and duration for the rational activity of the first principle.

164

Stephan Herzberg

the assumption of such a unity one can advance the following prima facie arguments: Nowhere does Aristotle state that divine and human thinking are of the same kind. Rather, in E. N. X 8 the question of what kind of similarity it is that holds between human and divine contemplation, is precisely left open (1178b27). A strong argument against the assumption of an identity in kind, with a gradual difference merely with regard to temporal duration, is the passage which concludes our section of the text: “If, then, God is always (ἀεί) in that good state in which we sometimes (ποτέ) are, this compels our wonder; and if in a higher degree (μᾶλλον), this compels it yet more. And God is in such a state (ἔχει δὲ ὧδε)” (1072b24–26; Barnes with modifications). That μᾶλλον should here carry a merely quantitative-temporal difference can be ruled out, as this difference is already expressed in the previous sentence.34 If it is the case that “even things different in species admit of degree” (E. N. VIII 2, 1155b14 f.35), then it is reasonable to suppose that this difference refers to the activity itself, i.e. to the essence of θεωρία. In this case the God would realize what it means to be active in the mode of contemplation in a way still higher than human beings, namely in a perfect, unlimited way. This kind of difference, or rather the eminence of divine contemplation, has to be elaborated further on the basis of the text. It suffices, at this point, to keep in mind that a merely temporal difference for μᾶλλον is ruled out, and that the indicated difference is explicitly maintained to exist (ἔχει δὲ ὧδε: 1072b26). In what follows I will develop an interpretation of the passage 1072b14–26 in which a unity of kind of divine and human contemplation is, contrary to the standard interpretation, precisely not assumed. Rather, on my view, in this passage Aristotle wants to show why it is legitimate at all to further characterize the form of life of the first principle on the basis of our most consummate form of life.36

34

35

36

Cf. Wedin, M. V. 1988: Mind and Imagination in Aristotle, New Haven/London, 228, 233: “But the passage indicates a difference in the state itself and, thus, certainly appears to countenance difference in kind.” See also Kosman (see note 30), 311: “The view that Aristotle offers, it seems, is not simply that God thinks as we do, only all the time rather than merely some of the time; it is rather that God engages in an activity that is like thinking, but something more.” See also E. E. VII 12, 1245b16–19. “… whether there is one species of friendship or more than one. Those who think there is only one because it admits of degrees have relied on an inadequate indication; for even things different in species admit of degree.” See also Gabriel, M. 2006: Gottes transzendenter Seinsvollzug. Zur Aristotelischen Ontotheologie im Λ der Metaphysik, in: Jahrbuch für Religionsphilosophie, 5, 99, 113n54, who reads the section 1072b20–24 as an analogical extrapolation. See previously Oehler, K. 2 1985: Die Lehre vom Noetischen und Dianoetischen Denken bei Platon und Aristoteles. Ein Beitrag zur Erforschung der Geschichte des Bewusstseinsproblems in der Antike, Hamburg, 203.

God as Pure Thinking. An Interpretation of Metaphysics Λ 7, 1072b14–26

165

II. Thinking as a suitable Candidate (1072b14–19) Now if, according to its very nature, the first principle is actuality, and if this actuality consists in leading (διαγωγή) a form of life, which is comparable to our best form of life, then this διαγωγή, which is not yet qualified any further, is true of the first principle always: “And its life is such as the best which we enjoy, and enjoy for but a short time. For it is ever in this state (which we cannot be)” (1072b14–16). Aristotle mentions an additional reason for the excellence of the first principle’s form of life: “since its actuality is also pleasure”.37 For if pleasure accompanies a perfect activity as a supervenient perfection,38 and if the first principle, according to its very nature, is actuality or activity, then the first principle is essentially also pleasure and, therefore, pure pleasure (cf. E. N. VII 15, 1154b25 f.).39 In view of this perfect activity with its pure pleasure, which is not yet qualified any further, human activities like waking, perceiving and thinking (to their various different degrees) are pleasant in a derivative sense. Hopes and memories in turn are pleasurable insofar as they relate to these activities: “And therefore waking, perception and thinking are very pleasant (ἥδιστον),40 and hopes and memories are so because of their reference to these (activities)” (1072b17 f.). Now, in what kind of activity does the διαγωγή of the first principle consist? How can this activity, to which pleasure belongs as such or simpliciter, be further qualified? In 1072b17–8, Aristotle lists three activities of human life which are each pleasant to a high degree (waking, perceiving, and thinking). The next sentence addresses thinking “in itself”: “And thinking in itself deals with that which is best in itself, and that which is thinking in the fullest sense with that which is best in the fullest sense” (ἡ δὲ νόησις ἡ καθ’ αὑτὴν τοῦ καθ’ αὑτὸ ἀρίστου, καὶ ἡ μάλιστα τοῦ μάλιστα: 1072b18–9; trans. Barnes with modifications). Now, it makes a great difference to how we understand this passage, whether we do or do not read Aristotle in such a way, that he assumes the unity in kind of divine and human thinking (or θεωρία) right from the beginning. Commentators making this assumption often think that the statement just quoted (1072b18–9) relates to the divine thinking itself. According to the ‘positive systoichia’ mentioned in 1072a31 f., the highest object of thinking would be the simple and essentially

37 38 39

40

ἐπεὶ καὶ ἡδονὴ ἡ ἐνέργεια (1072b16). On this see Laks (see note 5), 233 f. Cf. E. N. X 4, 1174b31–33; X 5, 1175a29 f. See also Bordt (see note 5), 117: According to Bordt, what pleasure is becomes paradigmatically clear from the real activity of the first principle, because in this case pleasure appears in its purest and fullest form. Because of the derivative character I read ἥδιστον in an elative sense (cf. also Laks (see note 5), 233).

166

Stephan Herzberg

actual substance. From this it would follow – though left unmentioned here by Aristotle – that the object of God’s thinking must be God himself. After all, the object of his thinking is “that which is best in the fullest sense”. Otherwise something else would be better, which would mean that God is not the highest being.41 This implicit divine thinking-himself would then be explained further by the following sentences (1072b20–24) with reference to human thinking. Because of the assumed identity of kind with divine thinking, human thinking likewise has a self-referential structure (αὑτὸν δὲ νοεῖ ὁ νοῦς), and has it in the same sense as divine thinking. One can, however, think of another way in which the expression ἡ δὲ νόησις ἡ καθ’ αὑτὴν in 1072b18 f. could be interpreted. I will develop this interpretation in what follows. Not only does it not need to assume a unity of kind, it can also avoid the assumption of a conclusion which is not drawn expressis verbis.42 This interpretation can give the passage 1072b14–26 an altogether more philosophically sophisticated sense insofar as it shows that this passage can be regarded as a consistent justification for why we are entitled to qualify the form of life of the first principle by starting from our own highest form of life. After Aristotle has mentioned waking, perceiving and thinking in 1072b17 from the human sphere, which are pleasant with respect to the ‘pure activity’ and ‘pure pleasure’ of the first principle, he selects from these

41

42

Cf. Ross (see note 32) II, 379: “In order to find the connexion between these two sentences, it seems necessary to suppose that when Aristotle says that the divine νόησις ἡ καθ’ αὑτήν is of τὸ καθ’ αὑτὸ ἄριστον he means the conclusion to be drawn ‘and therefore of the divine νοῦς itself’, which has been exhibited as the πρῶτον ὀρεκτόν (a27), in other words as the ἄριστον (a35). He then goes on to show how νοῦς knows itself”. Cf. Oehler, K. 1974: Aristotle on Self-Knowledge, in: Proceedings of the American Philosophical Society, 118, 499: “Since the highest type of activity is thought, Aristotle defines the nature of the Prime Mover to be thought, and thought in its purest form, that is as pure Noesis, without restriction to the sensory content of perception and imagination. The object of this highest, purest Noesis of the Prime Mover is that which is best in itself; that which is in the highest degree thought has as its object that which is in the highest degree good […] It is hardly surprising that Aristotle does not add expressis verbis the conclusion ‘therefore the Prime Mover thinks itself’. The notebook-like style of Book Lambda makes the enthymeme appropriate”. Norman, R. 1969: Aristotle’s philosopher-God, in: Phronesis, 14, 63–74 argues against such an ‘implicit syllogism’. According to him, such a syllogism would disrupt the continuity of the argument developed from 1072b15 to 1072b26 unnecessarily (68 f.). Regardless of this point, with which I agree, although it is not further developed based on the text, Norman also defends the problematic thesis that the unmoved mover has to be ascribed the same kind of thinking as humans, i.e. the αὑτὸν νοεῖν, which Norman (with reference to Aristotle’s analysis in De an. III 4, 429b5–9) contrasts as a “theoretical” kind of thinking with a “receptive” kind (67).

God as Pure Thinking. An Interpretation of Metaphysics Λ 7, 1072b14–26

167

three activities thinking or activity of reason (νόησις) in the next sentence: “And thinking in itself deals with that which is best in itself, and that which is thinking in the fullest sense with that which is best in the fullest sense” (1072b18–9). Thus the particle δέ in ἡ δὲ νόησις ἡ καθ’ αὑτὴν is not used in a connective-adversative sense – as in the standard interpretation – i.e., in such a way as to imply that human νόησις were here contrasted with divine νόησις. Rather, the particle δέ is understood in a connective-continuative sense; it refers back to the previously mentioned νόησις.43 This thinking is considered in a certain respect, namely “as such” (καθ’ αὑτήν). Especially if one takes into account the context of Aristotle’s psychology, the qualifier καθ’ αὑτό plays the role of highlighting a faculty with respect to its very own function (ἔργον), for the sake of which it came into being. It also highlights a faculty with respect to the objects (τὰ ἀντικείμενα) through which it is defined 44, or with respect to an effect it produces by itself.45 In the case of thinking this means to abstract from the specific human dependence on perception and imagination (φαντασία).46 This dependence is reflected on the one hand in the fact that we have to examine the perceptible objects with their immanent forms (τὸ εἶδος τὸ ἐνόν)47 if we want to acquire knowledge (ἐπιστήμη) about the world,48 and, on the other hand, in the fact that every act of human thinking is accompanied by some content of the imagination (φαντάσματα).49 This dependence of thinking on φαντασία and thus on perception implies that in human beings thinking cannot separate itself from their bodies (De an. I 1, 403a8–10; III 7, 431a16 f.). If, therefore, thinking is considered “in itself” this particular dependence of human thinking on sensibility is disregarded. Now, although it is the case that the thinking of God, as we learn in Met. Λ 9, is an eternal, non-procedural thinking not dependent on perception or imagination,50 it is not necessary to assume that at 1072b18 f. Aristotle

43 44 45 46 47 48

49

50

Cf. Denniston, J. D. 21950: The Greek Particles, Oxford, 162. In this way the different senses are determined by the “objects perceptible in themselves” (De an. II 6). Cf. the discussion on the movement of the soul (De an. I 3, 406a7, a11, b15). Cf. De an. III 3, 427b14–16; III 7, 431a14–17, 431b2–19. Cf. Met. VII 11, 1037a29; De an. III 8, 432a3–6. Cf. Anal. Post. I 31, 88a2–8; II 19. On this see the more detailed discussion in Herzberg, S. 2011: Wahrnehmung und Wissen bei Aristoteles. Zur epistemologischen Funktion der Wahrnehmung, Berlin–New York. Cf. De mem. 449b30–450a9. Both forms of dependence on the senses are expressed by Aristotle in the following sentence: “Hence no one can learn or understand anything in the absence of sense, and when the mind is actively aware of anything it is necessarily aware of it along with an image (ἀνάγκη ἅμα φάντασμά τι θεωρεῖν)” (De an. III 8, 432a7–9). This is especially emphasized by Wedin (see note 34), 229, 245.

168

Stephan Herzberg

already refers to divine thinking.51 It is more likely that this passage refers to thinking as it is treated in a general and structural manner in De an. III 4. In this chapter Aristotle wants to indicate clearly which differentia specifica defines the power of thinking and how thinking takes place. The faculty of reason is characterized, in an analogy to perception, as a receptive faculty (although not bound to an organ), which is “in itself” related to intelligible objects (νοητά): “The thinking part of the soul must therefore be, while impassible (ἀπαθές), capable of receiving (δεκτικόν) the form of an object […] Thinking must be related to what is thinkable, as sense is to what is sensible.” (429a15–18) If one disregards the specific dependence of human thinking on perception, then (theoretical) thinking is related “in itself” to what is of high ontological rank, namely to the simpliciter or “by nature” better-known essential structures which underlie reality (cf. Met. Λ 7, 1072a31). This ontological dignity is signalled by Aristotle with appropriate predicates like τίμιον, ἀγαθόν, θεῖον and their comparative forms. This is especially clear at E. N. X 7, 1177a14 f., where he says, of the capacity which underlies the activity in accordance with highest virtue, that due to its nature (κατὰ φύσιν) it has a cognitive access to the beautiful and divine things (ἔννοιαν ἔχειν περὶ καλῶν καὶ θείων; see also 1177a19–21). Yet Met. Λ 7, 1072b18 f. does not just declare the essential relation of thinking to the objects of high ontological rank, the passage also states that there is a dependence of the dignity of the act of thinking on the dignity of its object: “And thinking in itself deals with that which is best in itself and that which is thinking in the fullest sense with that which is best in the fullest sense” (1072b18 f.). This dependence on the dignity of the object, which Aristotle will explicitly come back to in Λ 9 (1074b29 f.), also shows itself in the pleasure which accompanies thinking: “the most complete is the most pleasant activity, and that of a well-conditioned organ in relation to the worthiest of its objects is the most complete” (E. N. X 4, 1174b21–23; trans. Barnes, slightly altered). Thus “as such” thinking (νόησις), if it is directed to the best objects, can grant us the highest pleasure. This is what distinguishes it from all other conscious activities. With this we have found the suitable candidate for the activity on the basis of which we can further qualify the ‘pure actuality’ which constitutes the essence of the first principle and to which belongs ‘absolute pleasure’.

51

Like Ross (see note 32 II, 379) and Oehler (see note 41, 499), for example.

God as Pure Thinking. An Interpretation of Metaphysics Λ 7, 1072b14–26

169

III. The Divine in Human Thinking (1072b19–24) After identifying human νόησις as a suitable candidate for an analogy to the divine mode of being, in a further step Aristotle subjects it to a detailed investigation by exploring the human intellect with regard to a divine element it contains (cf. 1072b23: ὃ δοκεῖ ὁ νοῦς θεῖον ἔχειν). Before I present my interpretation of this passage, I will have to mention two things: (1) in most cases Aristotle uses the predicate “divine” to describe the dignity of something of high ontological rank. This results from the realization of different characteristics to different degrees, e.g. of actuality, separateness, eternity or being unmoved, which stand in a certain order of rank to each other.52 Since the highest principle of reality has so far been identified as that which, by its very nature, is actuality (Λ 6, 1071b20; Λ 7, 1072a32), it is reasonable to think of the dignifying predicate θεῖον in the first instance in the sense of being actual. In my view, one cannot presuppose a more elaborate meaning of θεῖον at this point unless one makes the problematic assumption that in 1072b20 Aristotle has reached God’s thinking-himself as an unstated conclusion, and unless one regards this special self-referring cognitive act as what θεῖον means here. (2) If Aristotle wants to point out something divine in human thinking, and thus to identify an affinity with the mode of being of the highest principle, then, in my view, this passage has the function of stating a reason why we are entitled to characterize God’s ‘pure activity’ on the basis of the “for us better known” human thinking, i.e. “from the bottom up”. Since, as the highest οὐσία, God belongs to the logical space of being and is thus precisely not ἐπέκεινα τῆς οὐσίας, we can regard the uncovering of a special ‘structure of actuality’53 (and simultaneously of a high ontological dignity) in a certain activity of human life as proof of an affinity with the divine mode of being. Furthermore, we can claim – taking this as our starting point – to be able to gain knowledge of the kind and manner of his actuality, i.e., his essence. Thus, after having selected νόησις, which has the highest dignity in relation to the highest objects, from the preceding list of activities in 1072b18 f. Aristotle proceeds to examine this νόησις with regard to its ontological value. Now, the intellect or thinking is not a uniform phenomenon. Rather, we can distinguish different levels of this capacity to which belong different modes of activity. One may ask which mode of activity has the closest affinity to the mode of being of the highest principle. It is striking that Aristotle immediately begins with the intellect’s thinking itself: 52 53

Cf. Met. VI 1, 1026a20–29, Λ 8, 1073b3–5; De part. an. I 5, 644b25 f.; De cael. I 9, 279a32; De an. I 4, 408b29. For this term see also Buddensiek, F. 1999: Die Theorie des Glücks in Aristoteles’ Eudemischer Ethik, Göttingen, 236 f.

170

Stephan Herzberg

“And thinking thinks itself because it shares the nature of the object of thinking; for it becomes an object of thinking in coming into contact with and thinking its objects, so that thinking and the object of thinking are the same.”54 (1072b20 f.; trans. Barnes with modifications) The question of why Aristotle begins with a statement which, at first glance, is so hard to understand can only be answered by looking at De an. III 4. He seems to presuppose that the reader is familiar with this chapter. According to Aristotle, the human intellect, before it acquires any knowledge, exists in pure potentiality. It does not, therefore, belong to the things that exist in actuality: “it follows that it can have no nature of its own, other than that of having a certain capacity. Thus that in the soul which is called thought (by thought I mean that by which the soul thinks and judges) is, before it thinks, not actually any real thing” (De an. III 4, 429a21–24; 429b30–430a2). As part of the acquisition of knowledge the intellect receives the intelligible forms by which it becomes identical with the thinkable objects that are present in potentiality in the perceptible objects.55 Since, by its nature, the human intellect is a potentiality,56 it does not, as such, prior to the acquisition of knowledge, belong to the things which exist in actuality. It is for this reason that the possession of the immaterial intelligible objects, i.e., the fact that it knows these habitually, means that the intellect and what is thought of are the same (430a3 f.). Through its “participation in the intelligible” the intellect has become thinkable for itself. In this state, the intellect is on the level of the ‘first actuality’. While it is still something which is potentially a thinker (more precisely: a contemplator) it can, however, in contrast to an intellect which merely has the potential for knowledge acquisition, at any time, “when it wants” or “through itself”, move to the activity, i.e., the contemplation of the knowledge it has already acquired (De an. II 5, 417a26–28, b23 f.). This actualization consists then in nothing other than the intellect’s thinking itself: “When thought becomes each intelligible thing in the way in which a man who actually knows is said to do so57 (this happens when he is able 54 55

56 57

αὑτὸν δὲ νοεῖ ὁ νοῦς κατὰ μετάληψιν τοῦ νοητοῦ· νοητὸς γὰρ γίγνεται θιγγάνων καὶ νοῶν, ὥστε ταὐτὸν νοῦς καὶ νοητόν. This cognitive identity is one that is mediated through the intelligible forms, as Aristotle emphasizes especially at De an. III 8, 431b20–432a3. On form as a cognitive means of reference see Owens, J. 1980: Form and Cognition in Aristotle, in: Ancient Philosophy, 1, 17–27; Perler, D. 22004: Theorien der Intentionalität im Mittelalter, Frankfurt a. M.; Burnyeat, M. F. 2008: Aristotle’s Divine Intellect, Milwaukee, 20–24. Cf. also De an. III 5, 430a14 f. Cf. also De an. III 7, 431b16 f. (ὅλως δὲ ὁ νοῦς ἐστιν, ὁ κατ’ ἐνέργειαν, τὰ πράγματα). In this passage it becomes clear that the passage quoted above is about the intellect on the level of ‘first actuality’. See also Burnyeat (see note 55), 23.

God as Pure Thinking. An Interpretation of Metaphysics Λ 7, 1072b14–26

171

to exercise the power on his own initiative), its condition is still one of potentiality but in a different sense from the potentiality which preceded the acquisition of knowledge by learning or discovery; and thought is then able to think itself.”58 (De an. III 4, 429b5–10; trans. Barnes with modifications) So, in humans, the thinking-itself of the intellect refers to nothing other then the act by which one reflects on the epistemic content previously acquired. This thinking-itself, which has to be sharply distinguished from the reflexive consciousness of this self-reference,59 is thinking in the primary sense.60 While we can also speak of thinking when we consider a certain synthesis of thoughts in order to arrive at a judgment which can be true or false,61 the primary meaning of thinking is the contemplation of the contents of knowledge already acquired – with which one is already sufficiently familiar – in their different interconnections.62 According to De an. II 5, 417b6 f., we are here dealing with a “development into its true self or actuality” (εἰς αὑτὸ γὰρ ἡ ἐπίδοσις καὶ εἰς ἐντελέχειαν). Such an actualization confers greater joy on a cognitive act than the acquisition of knowledge (E. N. X 7, 1177a26 f.). Because Aristotle is here concerned with thinking in the primary and definitional sense, he addresses it right at the beginning of this passage. In this passage Aristotle also points out that the self-referential activity of the intellect is possible only because of its prior grasp of or sharing in intelligible objects (κατὰ μετάληψιν τοῦ νοητοῦ). Intellect becomes thinkable for itself only when it has received certain objects (1072b20 f.). The depen58

59

60

61

62

From what has been elaborated above, it becomes clear that it is possible – against the emendation of Bywater (δι’ αὑτοῦ), which is followed by Ross – to retain the variant of the surviving manuscripts (δὲ αὑτὸν). Cf. Kahn, C. H. 1992: Aristotle on Thinking, in: M. C. Nussbaum/A. Oksenberg Rorty (eds.), Essays on Aristotle’s De Anima, 372–375; Gerson, L. P. 2009: Ancient Epistemology, Cambridge, 78 f. Cf. Oehler (see note 41), 498. As in the case of the other cognitive acts of human beings (perception, opinion) the reflexive consciousness of this thinking is incidental or secondary (cf. Met. Λ 9, 1074b35 f.). Cf. Gerson (see note 58), 79: “It is this actualisation that enables the intellect to ‘think itself’. It is this latter actualisation that constitutes thinking in the primary, definitional sense.” Cf. the connection between διάνοια und ὑπόληψις in De an. III 3, 427b14–16, b25. Both are related to one another like the process to its result (cf. Hicks, R. D. 1907: Aristotle. De Anima. With translation, introduction and notes, Amsterdam, 457; Gerson (see note 58), 63): The discursive thinking on a certain state of affairs aims at a judgement which either takes the form of δόξα, ἐπιστήμη or φρόνησις. Cf. Dahl, N. O. 2011: Contemplation and eudaimonia in the Nicomachean Ethics, in: J. Miller (ed.), Aristotle’s Nicomachean Ethics. A Critical Guide, Cambridge, 70: “Contemplation, thus, turns out to be reflective appreciation of the nature of the world as revealed by Aristotelian science.“

172

Stephan Herzberg

dency on other objects for the actualization of νόησις reveals the nature of the human intellect, which consists in ‘pure potentiality’. This characterization of its essential nature does not only entail its basic receptivity63 and a merely limited self-sufficiency, insofar as self-referential thinking presupposes an arduous process of learning and is not simply available as such,64 but νόησις (or θεωρεῖν) is also determined in its content by the previously received intelligible objects and thus depends for its value on the value of the νοητά. This already shows the finitude of human thinking: the human mind is incapable of thinking itself in an unmediated way.65 Aristotle expresses this limitation of human thinking in the following sentences: “For that which is capable of receiving the object of thought, i.e. the substance,66 is thought. And it is active when it possesses this object. Therefore the latter rather than the former is the divine element which thought seems to contain, and the act of contemplation is what is most pleasant and best.”67 (1072b22–24) Thus the human intellect is receptive (δεκτικόν) according to its nature and it can be active only when it already has certain intelligible objects at its disposal (ἐνεργεῖ δὲ ἔχων), which then form the content of its contemplation (θεωρήματα: E. N. IX 4, 1166a26). With the help of his “triple scheme”68 Aristotle has distinguished different modal levels in which the intellect can exist and which correspond to different modes of thinking. On the basis of this, one can now ask what exactly it is about the human intellect that has the characteristic of being divine (θεῖον), i.e., possesses a ‘structure of actuality’ akin to the divine mode of being, to ‘pure actuality’.

63 64

65 66 67

68

Cf. Burnyeat, M. F. 2002: De Anima II 5, in: Phronesis, 47, 71. In E. N. X 7, 1177a27 f. Aristotle claims that the criterion of self-sufficiency applies most to the contemplating activity (ἥ τε λεγομένη αὐτάρκεια περὶ τὴν θεωρητικὴν μάλιστ’ ἂν εἴη). When the wise man has all the things which are necessary for life and leisure he can embark on the contemplation of the knowledge he has acquired previously. But only few people have the opportunity to acquire theoretical knowledge and even if they possess it there are hindrances to reflecting on it as such (see the precise observation in De an. II 5, 417a28). Thus human nature reveals itself as “in many ways in bondage”. One can indeed regard the acquisition of wisdom as beyond human power (Met. I 2, 982b28–30). Cf. Oehler, K. 1997: Subjektivität und Selbstbewußtsein in der Antike, Würzburg, 41. καὶ is here to be understood in an explicative sense (cf. Bordt (see note 5), 120). τὸ γὰρ δεκτικὸν τοῦ νοητοῦ καὶ τῆς οὐσίας νοῦς, ἐνεργεῖ δὲ ἔχων, ὥστ’ ἐκείνου μᾶλλον τοῦτο ὃ δοκεῖ ὁ νοῦς θεῖον ἔχειν, καὶ ἡ θεωρία τὸ ἥδιστον καὶ ἄριστον. Here I follow Ross and Jaeger who, contrary to the manuscripts and with the paraphrasis of the commentary of Ps.Alexander of Aphrodisias, prefer ἐκείνου μᾶλλον τοῦτο. Cf. Burnyeat (see note 63), 48.

God as Pure Thinking. An Interpretation of Metaphysics Λ 7, 1072b14–26

173

At this point, in the text there is an unclarity regarding the correct reading and translation of the Greek.69 The reason for this is that Aristotle mentions two elements, of which the one (= A) “more” (μᾶλλον) than the other (= B) realizes “the divine (θεῖον) which thought seems to contain”. However, he leaves the reference of both elements underdetermined by just using demonstrative pronouns. Moreover, the search for the respective object of reference is made more difficult by the fact that Aristotle does not say explicitly what the meaning of θεῖον is in this passage. If we understand Aristotle in such a way that he is looking for something divine in human thinking, namely in the sense of a structure of actuality which is similar to the ‘pure actuality’ of the first principle, then it may be natural to think – as in Ross’s and Jaeger’s reading of this passage – that τοῦτο (= A) refers to ἐνεργεῖ and ἐκείνου (= B) refers to δεκτικὸν. Since we are here dealing with human thinking, which can be explained more precisely with the “triple scheme”, the higher ontological dignity lies with ἐνεργεῖ, on the basis of possession (ἔχων70), i.e. the habitual knowledge, of the intelligible objects (‘first actuality’), and thus with the act of contemplating certain epistemic contents (‘second actuality’) with which it has previously become identical (1072b19–21). By contrast, the δεκτικὸν is the intellect on the level of the ‘first potentiality’ which has no other nature than to be potentially identical in character with its object (De an. III 4, 429a15 f., a21–25). Therefore the sentence is to be understood in such a way that “the latter”, i.e., the activity in the sense of the thinkingitself of human thinking, “rather than the former”, i.e., that “which is capable of receiving the object of thought”, is “the divine element which thought seems to contain”.71 From this it follows immediately that in human beings contemplation (θεωρία) is the most pleasant and best activity.72 69

70 71

72

The surviving manuscripts read ὥστ’ ἐκεῖνο (= A) μᾶλλον τούτου (= B) ὃ δοκεῖ ὁ νοῦς θεῖον ἔχειν, καὶ ἡ θεωρία τὸ ἥδιστον καὶ ἄριστον. However, in their respective editions Ross and Jaeger (whom many translators follow in this) read ὥστ’ ἐκείνου (= B) μᾶλλον τοῦτο (= A) ὃ δοκεῖ ὁ νοῦς θεῖον ἔχειν, καὶ ἡ θεωρία τὸ ἥδιστον καὶ ἄριστον with the paraphrasis of the commentary of Ps.-Alexander of Aphrodisias (698, 35 Hayduck). Cf. De an. II 5, 417a25, a32 f., b5. Cf. Kosman (see note 30), 310: “we learn that it is active thinking that is most divine in us”. Interestingly, the intellect “which is what it is by virtue of making all things” and which is essentially actuality and activity (De an. III 5, 430a18) is not included in this analysis. For a different interpretation see Reeve, C. D. C. 2012: Action, contemplation, and happiness: an essay on Aristotle, Cambridge–London, 215. If one keeps to the reading of the manuscripts, ἐκεῖνο (= A) can either be seen as referring to the δεκτικόν or the τοῦ νοητοῦ. However, the intellect on the level of the ‘first potentiality’ (δεκτικόν) cannot be of higher dignity than the intellect on the level of the ‘first actuality’ (ἔχων) or the ‘second actuality’ (ἐνεργεῖ). However, the intelligible object (τοῦ νοητοῦ) cannot be of higher dignity than the possession of knowledge of it or rather the actuality based on this possession. While it is indeed the case that the value of the act of thinking depends on the value of the object thought of and while the objects have priority insofar as it is through

174

Stephan Herzberg

If one looks back to the interpretation of Met. 1072b19–24 just developed, it becomes evident that it requires fewer additional assumptions (no presupposition of an identity of kind) than the standard interpretation, that it is closer to the text (no unstated conclusions) and that, in general, it is more faithful to Aristotle’s procedure in Met. Λ 6–7, which is guided by the nature of the subject matter (the ‘transcendence’ of the first principle). If this interpretation is correct, then Aristotle’s aim in this whole passage is to justify why we are entitled to take the “for us better known” human thinking as our point of departure when characterizing in more detail the mode of being of the eternal, unmoved and separate substance. Through an analysis of human thinking, which is guided by the “triple schema”, it becomes clear that among human activities contemplation, qua actualization of previously acquired knowledge, has the highest ontological dignity. Due to its special ‘structure of actuality’73 and its specific pleasure, θεωρία emerges as, among all human activities, most akin to the mode of being of the first principle, which is essentially actuality and purest pleasure. The διαγωγή of the first principle can, therefore, be characterized as θεωρία or νόησις; in 1072b15 it is θεωρία which can be called upon as a point of comparison.74 The God is always active in this magnificent way, which is possible for us to do only sometimes (1072b24 f.) One does not have to read the next sentence (εἰ δὲ μᾶλλον …) in order to realize already that besides the merely temporal duration of the act there has to be a deeper difference between human and divine contemplation which relates to the nature of this activity itself. Human contemplation is

73

74

them that the different capacities are defined (De an. II 4, 415a20–22; Met. IX 8) it is nevertheless true that prior to their being received by the intellect the intelligible objects lie in the individual perceptible objects, which have matter, only in potentiality (De an. III 4, 430a6 f.; III 8, 432a3–5). If it were otherwise, Aristotle would have no objective reason to introduce something like the ‘active intellect’ which – like a non-material medium – makes the intelligible objects, which reside in the perceptible individual things in potentiality, actually knowable, so that they can be received by the ‘passive intellect’ (De an. III 5, 430a15– 17). Contemplation is an activity (ἐνέργεια) in the narrow, technical sense of Met. IX 6, 1048b18–36. Learning, or the acquisition of knowledge, is a process which, as such, has an external goal and where the reaching of this goal also coincides with the temporal end of the process (it comes “to a standstill”: Phys. VIII 3, 247b11 f.). Thinking, however, in the sense of contemplation is a perfect activity: it has its goal in itself; in every moment of its actualization its goal is reached. An activity of this kind can be continued for any length of time. In this respect this kind of thinking is similar to the circular motion. Cf. Kosman (see note 30), 311: “The claim that God thinks turns out to mean that since the activity of thinking is divine, it may therefore be the clearest icon we have of the being of the divine principle whose essential nature is activity, and on which depend heaven and earth.”

God as Pure Thinking. An Interpretation of Metaphysics Λ 7, 1072b14–26

175

only possible because of a previous acquisition of knowledge which is based on an intellect which by its nature is ‘pure potentiality’. Because of this dependence on other objects, human contemplation is self-sufficient in a limited sense only, and in its content and dignity it depends in each case on the objects received (ἄλλο κύριον). It is a thinking that can think itself only via reference to ‘other beings’.75 This cognitive self-reference, mediated, as it is, by external objects, has only an “incidental” (ἐν παρέργῳ: Met. Λ 9, 1074b36) reflexive consciousness. As Wedin rightly points out against a presumed unity of kind, the fact that thinking-itself is, as Aristotle states, based on an ‘intellect in potentiality’, affects the mode of θεωρία itself. One is, therefore, not allowed to simply, and without further qualification, make θεωρία, as it occurs in humans, the standard, i.e., to carry it over in the very same sense to God.76 Aristotle himself draws our attention to the foundations upon which human thinking rests and that it is only a certain mode of this thinking which contains something divine, i.e., has an ontological affinity to God’s mode of being. Thus, in our passage, Aristotle gives us a further characterization of God’s mode of being: God is pure thinking without any potentiality. However, unlike the previous features, it cannot be “conceptually” deduced from the ‘pure actuality’, but only transferred to God – albeit with modifications – by taking as our starting point the “for us better known” highest human activity. We do not, at this point, know what exactly God’s θεωρία consists in (see Met. Λ 9); yet we do know this much: that God’s thinking, since he is ‘pure actuality’, cannot be based on an intellect which is potentiality by nature and, therefore, likewise cannot be based on an acquisition of knowledge in which ‘external’ intelligible objects are received (see E. E. VII 12, 1245b17–19). It is for this reason that divine thinking cannot be identical in kind with human thinking.77

75 76

77

Cf. Oehler (see note 65), 40–43. Wedin (see note 34, 233) argues against Norman’s thesis as follows: “First, if Norman is correct about sameness in degree, then man’s theoretic activity will be the measure of the god’s. For as activity the fact that our mind is also a potentiality does not enter the picture and that is what is relevant here. The point, rather, is that the two sorts of activity differ – unless, what Norman cannot allow, admixture of potentiality affects, downward, the quality of human theorizing.” (italics in the original) In the case of God there is no relation to external objects: he can think himself, i.e., pure thinking without it being necessary to have received external objects before (cf. Oehler (see note 41)). He is self-sufficient in an eminent sense. His thinking cannot be regarded as a mere instance of the thinking-itself referred to at 1072b19–21: it differs from this thinking in an essential way. In the case of God the reflexive consciousness, which appears only ἐν παρέργῳ in human cognition (1074b36), moves – due to the absence of a cognitive relation to the world – to the centre.

176

Stephan Herzberg

IV. A look at E. N. X 8 and a conclusion The method of argument we have just elaborated is significantly different from the one in E. N. X 8, 1178b7–32. In that passage, Aristotle provides a further reason for why the perfect happiness of humans is “a certain contemplative activity” (θεωρητική τις ἐνέργεια: 1178b7 f.). If we take the widespread opinion that the gods are blessed and happy to the highest degree78 as our starting point (ὑπειλήφαμεν: 1178b9), that they live79 – as all assume (πάντες ὑπειλήφασιν) – and are, therefore, active (ζῆν […] καὶ ἐνεργεῖν ἄρα: 1178b18 f.),80 and if we can rule out that the gods are either engaged in virtuous action in its various forms or in the production81 of things, then the activity of the God, which surpasses everything else in blessedness (μακαριότητι διαφέρουσα), has to be one of contemplation (1178b22). Now, if God is happy to the highest degree and exclusively engaged in the activity of contemplation, then there has to be an intrinsic connection between θεωρία and happiness. Among all human activities the one that is most akin (συγγενεστάτη) to the activity of God will be associated with the greatest happiness. For humans, life is happy “insofar as some kind of likeness (ὁμοίωμά τι) of such activity belongs to them” (1178b26 f.).82 Aristotle combines this similarity relation with a gradation model: “Happiness extends, then, just so far as contemplation does, and those to whom contemplation more fully belongs are more truly happy (καὶ οἷς μᾶλλον ὑπάρχει τὸ θεωρεῖν, καὶ εὐδαιμονεῖν), not accidentally, but in virtue of the contemplation […]”.83 This gives rise to the fundamental question of whether the gradual difference pertains only to the temporal extension, during which in God and human beings the same kind of contemplation takes place to a different degree (that what God is engaged in permanently human beings can do only for a limited time), or whether it pertains to the nature of contemplation itself, which would, in that case, exist “as such” in different degrees. The latter would mean that, in contrast to God, human beings would manifest what it

78 79 80 81 82

83

Cf. also E. N. I 12, 1101b23 f. On this endoxon see also Met. Λ 7, 1072b28 f. “Life” logically implies “activity” or “actuality”, but not vice versa (see above). “Life” is here qualified as rational life, namely that it consists either in action, production or contemplation (cf. Top. VI 6, 145a15 f.; Met. VI 1, 1025b25; E. N. VI 2, 1139a27 f.). In Aristotle’s writings the concept of similarity often occurs together with the concept of kinship, e.g., E. N. VI 2, 1139a10 f. (καθ’ ὁμοιότητά τινα καὶ οἰκειότητα); Top. I 7, 103a18 f. (ἅπαντα γὰρ τὰ τοιαῦτα συγγενῆ καὶ παραπλήσια ἀλλήλοις ἔοικεν εἶναι); Top. I 10, 104a19 f. (πάντα γὰρ ὅμοια καὶ συγγενῆ ταῦτ’ ἔοικεν εἶναι); Rhet. I 11, 1371b18 (πάντα τὰ συγγενῆ καὶ ὅμοια ἡδέα ὡς ἐπὶ τὸ πολύ). On this compare Kraut, R. 1989: Aristotle on the Human Good, Princeton, 39–77.

God as Pure Thinking. An Interpretation of Metaphysics Λ 7, 1072b14–26

177

means to be active in the form of contemplation only in a limited way and that it cannot be predicated of them in a univocal, but only in a derivative or analogous sense.84 Only God would manifest contemplation in an unlimited and perfect way. Looking towards him we could understand what it means, to lead a life of contemplation. The concept of similarity, which Aristotle uses at 1178b27 to characterize the relation between human and divine contemplation, can be interpreted in both ways. It can be applied to things of the same kind, which differ in the way in which they individually express the nature of their respective kind, as well as to things which do not stand to each other in the relation of belonging to the same genus, as in an analogy or in the pros hen–unity (see Top. I 17; E. N. I 4, 1096b26–29).85 A gradual difference (τὸ μᾶλλον καὶ τὸ ἧττον) is not only found in the category of quality (cf. Cat. 3b39–4a2), but for Aristotle things belonging to different kinds can also exist to different degrees: “Those who think there is only one (kind) because (friendship) admits of degrees (τὸ μᾶλλον καὶ τὸ ἧττον) have relied on an inadequate indication; for even things different in species (καὶ τὰ ἕτερα τῷ εἴδει) admit of degree” (E. N. VIII 2, 1155b13–15; trans. Barnes; slightly altered). Therefore, just because one can talk about the different degrees of something, this does not necessarily indicate that there is qualitative difference within one and the same kind. On the contrary, it can also be used for things differing in kind, as in the case of different kinds of friendship, for example,86 or of different genera of being.87 The fundamental question of whether divine and human contemplation coincide with regard to their kind cannot be decided on the basis of E. N. X 8. The main reason for this is the equivocal concept of similarity, which Aristotle does not specify any further in that text. In E. N. X 8 as well as in Met. Λ 7 he simply makes clear that contemplation can exist in different degrees (1178b29 f.; 1072b25). What follows from E. N. X 8 – and here especially from the emphasis on the superiority of divine happiness (1178b22) – leaves at least logical space for a reading which sees here an essential difference between divine and human contemplation.88 My reading of Met. Λ 7, 84 85

86 87 88

The indefinite pronoun in θεωρία τις (1178b32) would have a “weakening” function here (cf. Wedin (see note 34), 212n3). For Aristotle, “similarity” is not an irreducible concept. Rather, for him it is a concept based on a certain kind of unity or common feature ‘below’ numerical unity and which can, at the same time, indicate a gradual difference that eludes a definitive definition. (cf. Rapp 1992). Basically, it is true to say: “[F]or in so far as they have any identical attribute, in so far they are alike” (Top. I 17, 108a16 f.). Cf. E. N. VIII 5, 1157a12–14; VIII 8, 1158b4 f. Cf. Met. VII 1, 1028a25 f.; XIV 1, 1088a29 f. Cf. also Wedin (see note 34), 211, 229, 245.

178

Stephan Herzberg

1072b14–26 has suggested that we should positively assume such a difference in kind. I have tried to show that such an alternative interpretation is indeed possible and also better. Moreover, with regard to E. N. X 8 my interpretation has an interesting upshot concerning the way the argument unfolds. Whereas in E. N. X 8 Aristotle characterizes the perfect happiness of human beings by proceeding “top down”, i.e., by starting with the perfect happiness of God and making use of endoxa for the further characterization of God’s activity, the argument in Met. Λ 7, 1072b14–26 takes exactly the opposite approach. Here Aristotle proceeds “bottom up” starting from the highest form of human thinking and transferring this per analogiam to the divine mode of being.89 This can make better sense of the eternal, unmoved, separate substance whose essence is actuality. Because of the special ontological status, one has to assume that God manifests our best form of life, i.e. θεωρία, in an even higher way. Aristotle claims that this possibility is, in fact, realized by closing his argument with the following sentence: “If, then, God is always in that good state, this compels our wonder; and if in a better (μᾶλλον) this compels it even more. And God is in a better state (ἔχει δὲ ὧδε)” (1072b24–26; with E. E. 1245b16–19 and E. N. 1178b28–30).

Bibliography Barnes, J.: The Complete Works of Aristotle, Princeton 1984. Bordt, M.: Aristoteles’ Metaphysik XII, Darmstadt 2006. Bordt, M.: Why Aristotle’s God is Not the Unmoved Mover, in: Oxford Studies in Ancient Philosophy, 40 2011, 91–109. Buddensiek, F: Die Theorie des Glücks in Aristoteles’ Eudemischer Ethik, Göttingen. 1999. Burnyeat, M. F.: De Anima II 5, in: Phronesis, 47 2002, 28–90. Burnyeat, M. F.: Aristotle’s Divine Intellect, Milwaukee 2008. Dahl, N. O.: Contemplation and eudaimonia in the Nicomachean Ethics, in: J. Miller (ed.), Aristotle’s Nicomachean Ethics. A Critical Guide, Cambridge 2011, 66–91. Denniston, J. D.: The Greek Particles, Oxford 21950.

89

Cf. Oehler (see note 65), 58 f.: According to Oehler, although divine and human noetic thought are not identical, but different, they are not absolutely different. They stand to each other in a relation of analogy, namely of an analogia attributionis. Their characteristics correspond to each other, yet they do so in such a way that each of the characteristics of human noesis is lesser in degree when compared to the analogical characteristic of divine noesis.

God as Pure Thinking. An Interpretation of Metaphysics Λ 7, 1072b14–26

179

Gabriel, M.: Gottes transzendenter Seinsvollzug. Zur Aristotelischen Ontotheologie im Λ der Metaphysik, in: Jahrbuch für Religionsphilosophie, 5 2006, 97–119. Gaiser, K.: Das zweifache Telos bei Aristoteles, in: I. Düring (ed.), Naturphilosophie bei Aristoteles und Theophrast, Heidelberg 1969, 97–113. Gerson, L. P.: Ancient Epistemology, Cambridge 2009. Herzberg, S.: Wahrnehmung und Wissen bei Aristoteles. Zur epistemologischen Funktion der Wahrnehmung, Berlin–New York 2011. Herzberg, S.: Menschliche und göttliche Kontemplation. Eine Untersuchung zum bios theoretikos bei Aristoteles, Heidelberg 2013. Horn, C.: In welchem Sinn enthält Metaphysik Lambda eine Theologie?, in: Jahrbuch für Religionsphilosophie, 1 2002, 28–49. Kahn, C. H.: Aristotle on Thinking, in: M. C. Nussbaum/A. Oksenberg Rorty (eds.), Essays on Aristotle’s De Anima 1992, 359–379. Kosman, A.: Metaphysics Λ 9: Divine Thought, in: M. Frede/D. Charles (eds.), Aristotle’s Metaphysics Lambda. Symposium Aristotelicum, Oxford 2000, 307–326. Krämer, H.-J.: Der Ursprung der Geistmetaphysik. Untersuchungen zur Geschichte des Platonismus zwischen Platon und Plotin, Amsterdam 1964. Kraut, R.: Aristotle on the Human Good, Princeton 1989. Laks, A.: Metaphysics Λ 7, in: M. Frede/D. Charles (eds.), Aristotle’s Metaphysics Lambda. Symposium Aristotelicum, Oxford 2000, 207–243. Menn, S.: Aristotle and Plato on God as Nous and as the Good, in: Review of Metaphysics, 45 1992, 543–573. Norman, R.: Aristotle’s philosopher-God, in: Phronesis, 14 1969, 63–74. Oehler, K.: Aristotle on Self-Knowledge, in: Proceedings of the American Philosophical Society, 118 1974, 493–506. Oehler, K. : Die Lehre vom Noetischen und Dianoetischen Denken bei Platon und Aristoteles. Ein Beitrag zur Erforschung der Geschichte des Bewusstseinsproblems in der Antike, Hamburg 21985. Oehler, K.: Subjektivität und Selbstbewußtsein in der Antike, Würzburg 1997. Owens, J.: Form and Cognition in Aristotle, in: Ancient Philosophy, 1 1980, 17–27. Perler, D.: Theorien der Intentionalität im Mittelalter, Frankfurt a.M. 22004 Rapp, C.: Ähnlichkeit, Analogie und Homonymie bei Aristoteles, in: Zeitschrift für philosophische Forschung, 46 1992, 526–544. Richardson Lear, G.: Happy Lives and the Highest Good. An Essay on Aristotle’s Nicomachean Ethics, Princeton 2004. Ross, W. D. (ed.): Aristotle’s Metaphysics. A Revised Text with Introduction and Commentary, Oxford 1924.

180

Stephan Herzberg

Sedley, D.: Creationism and Its Critics in Antiquity, Berkeley–Los Angeles– London 2007. Wedin, M. V.: Mind and Imagination in Aristotle, New Haven/London 1988.

Unmoved Mover as Pure Act or Unmoved Mover in Act? The Mystery of a Subscript Iota SILVIA FAZZO

I. Prologue: raising a question in Aristotelian historiography In my present contribution, I will suggest the hypothesis that the theory of the Prime Mover as pure act, no doubt a main tenet of Aristotelianism in modern accounts, in fact originates from the plain statement that the Prime Mover must be in act; it slowly took its actual shape much later than it is supposed to have done. The process cautiously began with Alexander of Aphrodisias, then developed in a Neoplatonizing atmosphere, partly in Greek, partly in Arabic.1 In Greek, this happened during a time when scribes and scholars were less and less equipped to make full sense of the most ancient manuscripts, written in scriptio continua and uncial script. In such scripts, the subscript iota – which makes the difference between ‘in act’ (or ‘in activity’, ἐνεργείᾳ in the dative with subscript iota) and ‘act’ (ἐνέργεια in the nominative without subscript iota) – was not recorded in the majority of cases. Thus, ἐνέργεια written in capital letters can signify either ἐνέργεια or ἐνεργείᾳ. This happens in many more recent copies as well, namely in minuscule manuscripts, as extant manuscripts usually are.2 For this reason, it is commonly admitted that changes in the subscript iota hardly count as emendations, both in general and especially when dealing with manuscripts from the Aristotelian tradition.3 The point especially affects our case study, i.e. the opposition between ἐνεργείᾳ and ἐνέργεια as predicated of the first Unmoved Mover. For Aristotle, this is a non-sensible, separate substance. As such, the first Unmoved 1 2 3

See below, Appendix 1. “The first principle as ἐνέργεια in Michael of Ephesus, Plotinus and Averroes”. Other minuscule manuscripts have the iota but not as a subscript iota. Hence some further cases of ambiguity, see for example 1051b31 and ad loc. note 10 below. M. Burnyeat, ‘Kinesis vs. Energeia: A much-read passage in (but not of) Aristotle’s Metaphysics’, Oxford Studies in Ancient Philosophy, Volume XXXIV (2008) 219–92: 256. For an example, see e.g. Bruns’ critical apparatus of Alexander’s Quaestio 1.1 at p 4.11, with my note 15 below.

182

Silvia Fazzo

Mover is the main focus of Metaphysics book Lambda. As a Mover, it acts, by definition. The act of the first Unmoved Mover is eternal, unchangeable, free of potentiality. In this sense, it is ‘pure’. The point, however, is whether or not this separate substance, or any other separate substance, can be described as ἐνέργεια in itself. The question arises because that is currently given as if it were Aristotle’s own view and way of expression. However, this is not confirmed by my new critical edition of the book.4 The present contribution is designed to discuss the point in further detail. In this regard, we will see that the scribe of our earliest manuscript, J (Vindobonensis gr. phil. 100, mid-ninth century), sometimes hesitates between the dative and nominative (he refrains from adding accents, which was his standard way of suspending judgement). So eventually does E (Parisinus gr. 1853), at the beginning of the following century, in the middle of a Neoplatonizing atmosphere in Byzantium. As for the famous codex Ab of the Metaphysics, the Laurentianus 87.12 (twelfth to fourteenth century), there, as already Christ (1886) emphasized, the difference between a dative or a nominative singular form must be perceived by the reader, as it is largely omitted in writing.5 Hence a strict, albeit uncomfortable conclusion: the editorial choice between “act” and “in act”, ἐνέργεια and ἐνεργείᾳ, rests on a semantic basis.6 The written word is, so to speak, under-determined. Therefore, for a decision to be legitimate, we need to consider the theoretical role of the relevant sentence. The most challenging example of this problem is probably Metaphysics Λ 7, 1072a24–26, where ἐνέργεια is usually read and printed as a nominative form, as if the meaning were: the unmoved substance is ἐνέργεια. To these lines I devote the central part of my contribution, explaining why I cautiously opt for the dative ἐνεργείᾳ in this passage as well: when ἐνέργεια is interpreted

4

5

6

Fazzo, S.: Il libro Lambda della Metafisica di Aristotele, “Elenchos” LXI-1, Napoli 2012; Fazzo, S.: Commento al libro Lambda della Metafisica di Aristotele, “Elenchos” LXI-2, Napoli 2014. Christ gives the example of οὐσία: he notes that you should not ask the scribe of Ab whether he meant οὐσία or οὐσίᾳ: you have to find it out yourself (W. Christ, Aristotelis Metaphysica, recognovit, Teubner, Leipzig 1886, nova impressio correctior 1895). Christ’s description of Ab is not without impact, see see Fazzo 2012, 35–92: “Breve storia editoriale della Metafisica con particolare riferimento al codice Ab”. This especially applies if the word is transmitted without iota; for it can be shown that oldest manuscripts witness ΕΝΕΡΓΕΙΑ with a tacitly implied iota even in many places where the obvious meaning requires the dative, i.e., in our script, a subscript iota. Semantics have a slightly lesser bearing on the decision if a subscript iota is given in our manuscripts (see e.g. below ad 1071b20).

Unmoved Mover as Pure Act or Unmoved Mover in Act?

183

as a dative form and printed with a subscript iota, Lambda’s Unmoved Mover is not a pure act, but purely in act, ἐνεργείᾳ. Such a revision has its advantages. It is true that the pure act theory has always had an impressive impact factor; but it is not free of unwanted side effects, which can be improved by the suggested reading, ἐνεργείᾳ, in more than one respect. Firstly, the pure act theory, together with other aspects of Lambda’s alleged ‘theology’, poses a challenge to the book’s logical coherence (see below), especially in the frame of the syntax in use in the remaining books of Aristotle’s Metaphysics – with special reference to the actuality-potentiality theory of Book Theta. As a result, the syntax of ἐνέργεια is peculiar, so in the twentieth century editions of the Greek text of Metaphysics Lambda, as in late Neoplatonist readings of Aristotle: unlike Aristotle, they make of ἐνέργεια an absolute concept. In fact, Aristotle’s ἐνεργείᾳ, at least implicitly, is correlative with δυνάμει, constructed moreover with reference to an underlying subject, mainly a substance which switches from a state of δύναμις to a state of ἐνέργεια, i.e. from being δυνάμει to being ἐνεργείᾳ – hence the relevance of the dative case, when directly predicated of the Prime Mover’s way of being. This all is cleared away from book Lambda in most of the current readings. Aristotle is supposed to describe the first Unmoved Mover as a ‘pure ἐνέργεια’. As a result, the tight connection between Lambda and the central books of the Metaphysics remains unclear, and so does the unity of Aristotle’s project throughout these books. Indeed, the same language is spoken in Lambda as in Theta, or Zeta and Eta: based on that very system of thought, Lambda demonstrates a purely intelligible substance, which is entirely unmoved. Lambda does so via the general theory of substance as a focal meaning of being, which is built in the central books of the Metaphysics. Such coherence among Aristotle’s various books, which I consider now from the point of view of language, is a main issue at stake. The text of Lambda, at least as I edited it based on manuscript authority, talks about ‘being in act’, and not about ‘being act’, inter alia, at 1071b22, 1072a5–6, 1072a25, 1072b5. Otherwise, from the point of view of meaning, there is also a problem with the standard reading. Namely, one is likely to lose sight of the central value of Aristotle’s original assessment that the Unmoved Mover is ‘in act’. That is, one loses sight of the very way through which the Mover’s existence is being demonstrated in Lambda 6–7: the Mover must be eternal, and eternally in act, insofar as the effect if its activity – an eternally even movement – is proved to be eternal. Therefore, the Mover is permanently active, i.e., in act, i.e., acting always in the same way. In this sense, Aristotle’s first Mover is meant to be, not simply unmoved, but unchanged, ἀκίνητον: its moving

184

Silvia Fazzo

activity never changes. Changeless activity cannot but be eternal. Hence the eternity of the Mover which eternally acts: the eternity of the Mover is proved through the eternity of its activity and not the other way around (Λ 7, 1072a23–26).7 This general theory must have been regarded by Aristotle and his contemporaries as a real achievement of book Lambda, especially in the philosophical context of its original scope, i.e. the demonstration of an unmovable substance, an ἀκίνητος οὐσία, as announced at 1071b4 f.: “we must assert that it is necessary that there should be an eternal unmovable substance” (λεκτέον ὅτι ἀνάγκη εἶναι ἀίδιόν τινα οὐσίαν ἀκίνητον). This general theory should remain in the background of any particular reading of the book. In the context, it would make perfectly good sense to say that the Prime Mover’s act is free from potentiality; but the Mover itself is not an act: it acts. However, there is hardly any general account of Aristotle in modern times that fails to define the Prime Mover as ‘pure act’ (which is not the same as being in act). Henceforth, different scholars give different explanations of the formula. What would it mean that the unmoved substance, which the Prime Mover is, is ἐνέργεια? If this was really clear, it would not be difficult to conceive a substance which is pure ἐνέργεια, i.e. (probably) purely ἐνέργεια. The question is whether it really makes sense, in Aristotle’s terms and categories, i.e., ways of predication, to say that a substance is ἐνέργεια. Indeed, that the Mover is act is constantly repeated by Aristotle’s interpreters, but a proper explanation of what this means is often avoided. Instead, other Aristotelian texts are merely quoted (mostly in a footnote or parenthesis), as if their relevance to the actus purus theory were self-evident – which it is probably not (samples of alleged parallels are discussed in § VI.1, VII here below). However, not everyone quotes the same texts. The very fact that there is a wide discrepancy in the identification of textual foundation for the Prime Mover as pure act may be taken as a sign among others, that indeed the theory has scant textual support, and that no proof has been given.

7

Cf. Λ 7, 1072a23–26, a locus vexatus which I restituted without emendations and commented into detail in this very perspective, see ad loc. Fazzo 2012, 2014 and § VII below.

Unmoved Mover as Pure Act or Unmoved Mover in Act?

185

II. Act as activity (i.e. moving activity) or act as actuality, perfection and form? It also seems to be controversial whether this ‘act’, ἐνέργεια, is activity or actuality; the two diverge, since the latter is conflated with the meaning of ἐνέργεια as pure form in the standard vulgate. When commenting on Lambda 6, Enrico Berti suggested that this ‘act’, ἐνέργεια, should be understood in the more current sense of the word, as a kind of activity. This, with no doubt is more compatible with Berti’s proposal that the Unmoved Mover should be regarded as an efficient cause.8 The idea – as Berti recalls – was found difficult by the participants Symposium Aristotelicum on Lambda. This is not surprising, since the printed text they discussed had ἐνέργεια in the nominative case as a direct predicate of the Unmoved Mover. This requires a broader and less definite sense of ἐνέργεια: ἐνέργεια as actuality, susceptible to be converted into form, perfection, ἐντελέχεια; such terms do not directly mean or imply activity, but suggest a perfect state which is reached and accomplished already. Opting for the nominative would require this sense of ἐνέργεια. This makes sense, not in Aristotle himself, but within the tradition of exegesis. In this late Aristotelian frame, if we accept, as Aristotelian scholars mostly did, that the subject matter makes obstacle to the form to be accomplished and realized in its purest perfection,9 then one could conceive form, εἶδος, as a substance’s perfection, ἐντελέχεια, and ἐνέργεια as identical to ἐντελέχεια. This would allow the most perfect substance to be regarded as ἐντελέχεια, and as ἐνέργεια, in itself. So far the standard vulgate. Apparently, after all, the meaning would not be without support in the Corpus: the equivalence of ἐνέργεια with form is listed among the meanings of ἐνέργεια in Bonitz’s Index, s.v. In fact, closer analysis shows that such instances, most of which are in Metaphysics Eta 2–3, concern the form of sensible beings.10 8

9

10

E. Berti, “Unmoved Mover(s) as Efficient Cause(s) in Metaph. XII 6 th”. In: M. Frede, D. Charles, éd., Aristotle’s Metaphysics Lambda, Oxford, Oxford University Press, 2000, p. 181–206. See e.g. Alexander of Aphrodisias’ De providentia, 163.23–165.3 Zonta; the point is related to matter as a principium individuationis, on which see ibid. 157.14–18 (in Alessandro di Afrodisia. La provvidenza, a cura di S. Fazzo, traduzione dal greco di S. Fazzo, traduzione dall’arabo di M. Zonta, Milano: Rizzoli (BUR), 1999). Cf. Met. Eta 2, 1042b10 f.: λοιπὸν τὴν ὡς ἐνέργειαν οὐσίαν τῶν αἰσθητῶν εἰπεῖν τίς ἐστιν. This applies to the discussion in Eta 2–3 passim (which especially parallels De anima B 1.412a19–22), so that the Prime Unmoved Mover as a separate substance is precisely not at issue. Bonitz ibid. lists besides four Theta 8 passages, all of which refer to the activities (ἐνέργεια) of living beings (οἷον ἡ ὅρασις ἐν τῷ ὁρῶντι καὶ ἡ θεωρία ἐν τῷ θεωροῦντι καὶ ἡ ζωὴ ἐν τῇ ψυχῇ), once they are in act, ἐνεργείᾳ. At 1051b31, indeed, Ross prints ἐνέργειαι (at the plural nominative, hence ἐνέργειαι in the TLG text), but he recognizes that manuscripts

186

Silvia Fazzo

Therefore, they offer scanty support to the alleged occurrences in Metaphysics Lambda of a separate substance being itself act. Nonetheless, this kind of account thereafter became so usual that it is hardly surprising to encounter it in Aristotelian literature. Taken in this sense, the idea that the Unmoved Mover is pure act is often paired in general accounts with its being ‘pure form’. This is another claim which is typical of later Aristotelianism, but not entirely Aristotelian. As it, this latter idea cannot be found in Aristotle.

III. Alexander’s hylomorphism between exegesis and innovation By contrast, we are situated to trace the idea back to its likely origin, namely to its appearance in Alexander of Aphrodisias’ exegesis: this is due to Alexander’s hylomorphism, that is, Alexander’s characteristic tendency to develop Aristotle’s own hylomorphism far beyond its original boundaries, so that virtually every Aristotelian concept can be expressed by Alexander in terms of εἶδος and ὕλη.,11 As a matter of fact, there is a passage of Alexander, Quaestio I.25, 39.9–12, which includes the notion of ἐνέργεια as an Unmoved Mover’s predicate: the first substance is said to be ‘act’. Now, in this very passage, the assessment in question gives some firmer support to this other one, the Unmoved Mover as ‘form’ (εἶδος) 12 which is even more typical of Alexander’s exegesis.

11

12

(“codd.” in Ross’ critical apparatus) have ἐνεργείᾳ. So this does not count in favour of the nominative option concerning the word ἐνέργεια when predicated of a separate substance. Jaeger correctly prints the dative instead. Like every piece of Alexander’s system, this polarity has a basis in Aristotle; again, see e.g. De anima B 1. The reduction of most philosophical polarities to εἶδος and ὕλη on Alexander’s part, why ἐνέργεια and δύναμις are comparatively rare in Alexander’s own theories, in so far as one is substituted with form and the other with matter. In this framework, the first Mover, a separate substance no doubt, is for Alexander the special case of a εἶδος without ὕλη. On the whole process, see S. Fazzo, Aporia e sistema. La materia, la forma, il divino nelle Quaestiones di Alessandro di Afrodisia, “Pubblicazioni della Facoltà di Lettere e Filosofia dell’Università di Pavia”, ETS, Pisa 2002; on Alexander’s exegesis of Aristotle’s Unmoved Mover, Ead., L’exégèse de Métaphysique Lambda dans le De principiis et dans la Quaestio I.1 d’Alexandre d’Aphrodise in Laval théologique et philosophique, 2008, 607– 626. Alexander, Quaestio I.25, 39.9–12: “Among substances, according to Aristotle, there is one which is incorporeal and without body, a form without matter and separate, which is actuality separated from every potentiality, which [Aristotle] calls substance and intellect, and intellect in act, because it always thinks, and thinks the best one among beings, namely, itself.”

Unmoved Mover as Pure Act or Unmoved Mover in Act?

187

No doubt, given that Aristotle’s Unmoved Mover is a non-sensible substance, it is not entirely un-Aristotelian, in language and way of thinking, to take this a step further and to admit that a pure form and a pure ἐνέργεια exist as such. But this further step, although now a commonplace view, is not one that Aristotle himself can be shown to have taken, as far as we can see either from his extant writings, or from traces of his lost ones. In Aristotle, form, εἶδος, is always the form of something, of a substrate, so as act, ἐνέργεια is always act of a subject, as opposed to the subject’s δύναμις. By contrast, somehow, i.e., cautiously, Alexander took this further step, at least in some of his alleged writings. Why did he do so? As I will suggest, this belongs more to a conceptual reductionism, as generally practiced by Alexander while working on the whole of Aristotle’s theories, than to a crucial revision affecting the overall sense of them. In Alexander, after all, the basic alleged meaning of the Mover’s ‘being act’ is the same as ‘always in act’, ‘acting by its own essence, so that it cannot but act’ – a meaning, again, which, in itself, does not need to be based on the nominative option. Is the use of the nominative – Mover as ἐνέργεια – associated with the equivalence between ἐνέργεια and form, so to pave the way to Alexander’s main tenet, that Aristotle’s Prime Mover is a form without matter? It is probably so: according to Alexander, Aristotle’s first substance can be spoken of as form and ἐνέργεια.13 Through this, Alexander manages to produce a hylomorphic account of this first, non-sensible substance as well – as he does of every other being: the first Unmoved Mover is form separate from matter (Quaestio I.25, 39.14, 18, 30). Form, in turn, is seen by Alexander as implying intelligibility: “Form without matter is that which is intelligible at the highest degree” (ibid., 39.14). The first substance is object of thought not less than subject of thought. It is intelligible in act not less than intellect in act: the two go together. Hence the interest of a further detail: although Aristotle nowhere expressly recognizes the separate existence of a form (εἶδος) without matter, he often raises the question about the existence of purely intelligible substances, as an open issue. In Lambda 8, then, talking like someone who already got the answer, he refers once to the first Unmoved Mover as a quiddity (without matter) τὸ τί ἦν εἶναι (Lambda 8, 1074a35).14 This unique reference to a separate and unmaterial quiddity must have been taken into account by Alexander in support of his claim about the first Mover as εἶδος without matter.

13 14

Aristotle’s first substance is said to be εἶδός τι ἄυλον καὶ χωριστόν, ἐνέργειά τις οὖσα πάσης δυνάμεως κεχωρισμένη (Alexander, Quaestio I.25, 39.10 f., translated in the previous note). See on this passage note 29 below.

188

Silvia Fazzo

As a result, here as well as elsewhere, we see that Alexander develops any possible hidden implications of Aristotle’s texts in favour of his reading, while removing any major obstacle to his adaptation and paraphrases. As a result, thereafter, his kind of account has reached most Aristotelian literature. More precisely, we can distinguish between two levels in Alexander’s way of talking about the Mover’s act: when talking on his own, Alexander eventually introduces the idea of its being act itself, i.e. form without matter, act without potency. This is apparently found in a passage of Quaestio I.25 (see above), and possibly in the central part of Alexander’s De principiis as well (at least in its Arabic version of it, see Appendix below). More often still, when Aristotle’s text is more directly commented upon, Alexander is manifestly aware that Aristotle’s Unmoved Mover is not act but in act, not ἐνέργεια but ἐνεργείᾳ. For example, in the so called Quaestio I.1 (actually, an almost running paraphrase of parts of Lambda 6 and 7), Alexander deals with the point at issue very cautiously. See Quaestio I.1 passim (3.20, 21, 23; 4.10, 11, 14, 21), e.g. 4.11, where he says: “Form is intelligible, since it is something in act”. τὸ δὲ εἶδος νοητόν, ἐνεργείᾳ τι ὄν (ἐνεργεία τι ὄν V).15 More cautiously still, in a passage of the Quaestio I.25, Alexander refers to the active intellect 16 as to the one which Aristotle called ‘intellect in act’. The concept is expressed by νοῦν … τὸν κατ᾽ἐνέργειαν, 39.11. Saying κατ᾽ἐνέργειαν, instead of ἐνεργείᾳ, Alexander avoids any spelling ambiguities.17 Once more, we can wonder where Alexander might find that Aristotle does so: no such expression is to be found in Lambda, even not in those chapters where the text deals at length with intellect, i.e. in Lambda 7 and 9. As it appears, Alexander combines, in his exegesis, two different Aristotelian treatises, Lambda and De anima Γ 5, where the active νοῦς is by its own essence ‘in act’. Here as well as elsewhere, Alexander’s interpretation is highly influential. In modern exegesis as well as in critical editions, it seems that alleged passages about pure ἐνέργεια have been read as interconnected: the Prime Mover of Lambda is ἐνέργεια, being identical, as it is meant to be, with the active intellect of De anima Γ 5, which is also ἐνέργεια. Such a reading is the standard one since the late nineteenth century at least. Directly 15

16 17

Quaestio I.1, 4.11 Bruns. For or ἐνεργείᾳ the main manuscript, Venetus 258, has ἐνεργεία: the scribe signifies the dative through the accentuation. As it appears, he did not regard as necessary the iota to be spelled out. Cf. Alex. De anima 88.24 f.: νοῦς ποιητικός … τὸ χωρὶς ὕλης. Further on, regarding the same substance, Alexander says that it is “intelligible in act” (passim and 39.20: ἐνεργείᾳ ἐστὶ νοητή). So in the context it is clear that the meaning of κατ᾽ἐνέργειαν is the same as ἐνεργείᾳ.

Unmoved Mover as Pure Act or Unmoved Mover in Act?

189

or not, Alexander is one of its sources. Still, as we can see today, such alleged ‘pure act’ passages (most of them, if not even all of them) can also be read as if ἐνέργεια is in the dative case, which is more consistent with Aristotle’s standard way of talking about act and potency.

IV. A preliminary outline of an alternative assessment concerning the Unmoved Mover’s activity The idea that the first Unmoved Mover is acting can well be expressed in Aristotle by a simple dative ‘in act’. This can mean that it is always ‘in act’ in two ways. Firstly, the unrestricted chronological quantifier ‘always’ is to be gathered from the very absence of any restriction or precision; secondly, it is also analytically implied in the broadest notion of lack of change: Unmoved, in Greek: ἀκίνητον, is here something whose state does not change, and which will not therefore pass from activity to inactivity. If something ἀκίνητον acts, it always does so. This distinctive predicate apart, ‘in act by its own essence’, Aristotle’s non-sensible principle shares features with contemporary theories of principles in Plato’s school. In all of them, Unmoved, non-sensible – i.e., purely intelligible – substances are also given as principles of the sensible world. Aristotle’s strength lies not only in the fact that, unlike his fellow Platonists, he can produce a strict demonstration of the principles’ nature, number and necessary attributes. It is also the case that Aristotle’s non-sensible principle, Unmoved as it must be as such, is acting as a cause of movement and of becoming for our living world, unlike the principles in the theories held by Platonists, which were meant to be purely formal causes, according to Aristotle’s account of them in Met. Α 7, 988b2–5. This implies that the Aristotelian principle is not only an Unmoved, eternal substance, as those other principles were according to concurrent theories, but is always in actu, ‘in actuality’, ‘in activity’: in Greek, ἐνεργείᾳ. This is supported by my critical text of Lambda, unlike the standard assumption that the Aristotelian principle is act (ἐνέργεια). In fact, that the Unmoved Mover is ‘act’, or even ‘pure act’, is only a supposed feature of it, which does not arguably add any force to Aristotle’s argument. As a matter of fact, this odd expression is usually explained as if its meaning were exactly the same as what I suggest one should print instead: that the Unmoved Mover is simply in act (ἐνεργείᾳ). Namely, it is in act by its essence, it is always in act and cannot be but in act; this feature, which distinguishes it from any corruptible being, can be explained by saying that it has act as its proper nature and essence. This is an ontological feature

190

Silvia Fazzo

which Aristotle’s Unmoved Mover shares with any other eternal substance, namely, heavenly substance: eternal beings are never in potentiality: οὐθὲν δυνάμει ἀίδιον, as Θ 8 says (1050b7 f.: it is a parallel development about actuality and activity in eternal substances). No attempt is usually made to explain why, if Aristotle could so clearly express his thought in his ordinary language, he should have recourse to an exception and a singularity of this kind, which does not exactly fit his act-versus-potency theory. As a matter of fact, as far as I can see, the idea that Unmoved Mover(s) is (are) act(s) is not supported in Aristotle’s Corpus, even as it is currently printed, outside Metaphysics Λ and, on a peculiar understanding of it, in De anima Γ 5. But, if we look more closely at the manuscript tradition, it is not supported in these latter texts either, as we are about to see.

V. Crossing the boundaries: De anima Γ 5, 430a18 With all this, the theory of a pure act as first substance and divine principle became firmly established in Aristotelian studies. By 1862 it was not only central to a strict exegesis of Metaphysics Λ, but had also been applied to the text of De anima Γ 5, 430a18, where it is now a standard reading. It is worth examining the De anima section, so to avoid circularity between such alleged parallel passages. In fact, explaining De anima Γ 5 in light of Λ 7 and 9 and vice versa is, in itself, a current trend among commentators that goes back to at least 200 AD; discussing active intellect, they assume that Aristotle would have treated the subject similarly in those two very different works. This view can be detected already in Alexander of Aphrodisias, in one of the texts we quoted above, a partly independent treatise based on Alexander’s exegeses of Metaphysics Λ 6–7: namely in Quaestio I.25. In Alexander’s Quaestio I.25, at 39.11, we find the idea of a first substance as form and as act. Neither are these equations properly Aristotelian as such, nor does Alexander directly attribute these views and words to Aristotle. Nonetheless, what Alexander says about Aristotle’s intelligible, incorporeal and immaterial substance is interesting with respect to the history of the theory and of the transmitted text. Talking about the substance which is separate, without matter and without potentiality, Alexander says that Aristotle “calls it [i.e.] substance and intellect: an intellect in act” (ἣν οὐσίαν καὶ νοῦν καλεῖ, νοῦν δὲ τὸν κατ᾽ἐνέργειαν). Here, in order to express ‘in act’, Alexander does not use the dative form, but rather a more periphrastic prepositional phrase, κατ᾽ἐνέργειαν. Further on, regarding the same Aristotelian substance, Alexander says that such a substance is “intelligible in act” (ἐνεργείᾳ

Unmoved Mover as Pure Act or Unmoved Mover in Act?

191

ἐστὶ νοητή, ibid. 39.20 and passim); this means that such a substance is always thinking (a νοῦς in act) and always being thought (an intelligible, a νοητόν, in act) in so far as it always thinks the best of beings, namely, it thinks itself (39.11sf.). So much for Alexander on Aristotle’s first unmoved substance. Now, to which text of Aristotle does Alexander refer? No doubt, Metaphysics Lambda, chapter 7 (especially 1072b19–21), and chapter 9 (especially 1074b31–33), are the main reference texts about the Mover’s self thought. Indeed, Lambda 9 is an entire chapter of aporiai all of which are devoted to the notion of intellect.18 Still, one does not find, in book Lambda, the notion of an ‘intellect in act’ – nor, even less, of an intellect as act. This is perhaps unexpected, since Aristotle’s Unmoved Mover is so often described both as pure intellect and as pure act. Indeed, Lambda 7 discusses the act, i.e. the activity which is proper of the intellect, an activity which is life, see 7, 1072b26–28.19 An activity is at issue which is proper of the intellect. However, this does not amount to the theory of an intellect κατ᾽ἐνέργειαν yet. The closest parallel to Alexander’s νοῦς κατ᾽ἐνέργειαν in Aristotle’s Metaphysics is probably a reference to Anaxagoras’ νοῦς. This is recalled as being in act (ἐνεργείᾳ) in Lambda 6.20 Alexander may well use Anaxagoras’ intellect in act in order to revise Aristotle’s wording, when he talks about the νοῦς κατ᾽ἐνέργειαν. Among Aristotle’s texts, however, he seems to have in mind less the Metaphysics’ First Mover, than the active intellect of De anima Γ 5. The active νοῦς is by its essence in act (οὗτος … ὁ νοῦς τῇ οὐσίᾳ ὢν ἐνεργείᾳ) as opposed to potential intellect (ibid. Γ 5, 430a14–25: ὁ νοῦς … πάσχων … παθητικός … φθαρτός). But here there is a further problem. The reference to a νοῦς which is by its essence in act can be grasped only if we still read De anima, Γ 5, at 430a18 according to (for example) Bekker’s edition (1831) – an edition where the sentence could still be read in this way. Editing Aristotle’s De anima in 1862, Adolf Torstrik (in Ross’ view, an editor audacior quam fas est, this being said in general) was apparently the 18

19

20

Perhaps the very notion of intellect in act, which would be opposed to an ‘intellect in potency’, is beyond the scope of those Lambda chapters, which are entirely devoted to those substances which are note subject to change. However, the very concept of an intellect in act it seems to be incidentally referred to with τούτου in the sentence at 7, 1072b22 f., see Fazzo 2012 al loc., whose suggested reading of 1072b12 may explain an old textual crux, the printed text in the twentieth century. Aristotle focuses there on ἐκεῖνο, namely, the self-knowledge of the νοῦς, which is the divine component of it, see the next note. This applies to 7.1072b27 as well, if one follows manuscripts J and E (ἐκεῖνο, as referred to life) instead of Ab (ἐκεῖνος referred to νοῦς) as I intended to emphasize it in my edition (Fazzo 2012) ad loc. ὁ γὰρ νοῦς ἐνεργείᾳ, Lambda 6.1072a5s. (my edition cited).

192

Silvia Fazzo

first to inject a nominative reading into the definition of active intellect in the most seminal chapter of Aristotle’s psychology, at De anima Γ 5: the passage is the one we discussed above, 430a18. As a consequence, the halfhuman, half-divine Aristotelian νοῦς, which has some chance of being eternal even in its human part, is now said to be not only active (νοῦς … τῷ πάντα ποιεῖν, 430a14s.) but act itself. This reading is found in every printed edition and modern translation of Aristotle’s treatise I could check. In 1906, Hicks definitively accepted and justified the nominative reading, which he included in his editio maior.21 Thereafter, the nominative reading was integrated not only into the later editions, including Ross’ widely used OCT 1956 editio minor, but also – unsurprisingly – in the relevant Oxford Translation (i.e. Smith’s), now available at the Encyclopaedia Britannica website. Hicks decided the issue on the basis of the supposed parallels in Metaphysics Lambda, rejecting the main manuscript tradition of Aristotle’s De anima, which definitely supports the dative reading, and which therefore points to an in-act intellect. To tell the truth, the emendation, if it is one (admittedly, when subscript iota is concerned, the critical editor always has the last word), goes against the most obvious interpretation of the sentence at 430a18. The resulting syntax is also slightly less natural in the current text (oὗτος ὁ νοῦς … τῇ οὐσίᾳ ὢν ἐνέργεια, instead of the earlier construal, oὗτος ὁ νοῦς … τῇ οὐσίᾳ ὢν ἐνεργείᾳ). The meaning is, in my view at any rate, weaker, since the νοῦς … τῷ πάντα ποιεῖν is most obviously to be regarded as active, by its own essence (τῇ οὐσίᾳ), i.e. as in act, which means, in activity, rather than simply as ‘act’ itself. The intellectual and ideological frame is also relevant. The idea that the nominative ἐνέργεια refers to the active intellect can be found in Simplicius’ Greek exegesis (e.g. 244.35, 245.2). But this does not mean that Simplicius knew only the nominative option; like Alexander (and based on Alexander as well), he also has κατ᾽ ἐνέργειαν (243.20) and ἐνεργείᾳ (244.14) as can be seen in his commentary, where he says that the active intellect gets its perfection κατ᾽ ἐνέργειαν.22 Less directly, the Unmoved Mover is ἐνέργεια in Themistius as well. In his paraphrase of the Lambda, Themistius tries to conflate life and intellect through the first principle; this suggests that the theory of Mover as act was not established yet, but did not need to be regarded as impossible.23 21 22 23

Hicks, R. D., Aristotle. De anima, with Translation, Introduction and Notes, Cambridge 1906. αὐτὸς ὑφ᾽ἑαυτοῦ κατ᾽ ἐνέργειαν ἢ τελειούμενος ἢ ἤδη ὢν τέλειος, Simpl. CAG XI, 243.20 Hayduck. With this concern, Themistius tries to make sense of a difficult sentence at Met. Λ 7, 1072b26, on which see below. Themistius’ Lambda Paraphrase is lost in its original Greek version, but we have various versions of it which have been collated by Mauro Zonta. We

Unmoved Mover as Pure Act or Unmoved Mover in Act?

193

As an effect, the ontological role of actuality, ἐνέργεια, is emphasized; this reminds us less of Aristotle than of later developments of this concept from Plotinus onward.24 More generally, such a fortunate editorial intervention in the text of De anima presupposes a direct identification, both of Lambda’s Prime Mover with God and of the active intellect of De anima Γ 5 with God, and thus an indirect identification of Lambda’s Prime Mover with the active intellect of De anima Γ 5. This conflation is especially legitimized if we believe that theology was always a main concern of Aristotle, throughout the different parts of the Corpus; and that the final results of the various discussions, being concerned with theology, must agree with each other. Such results would pave the way to a Neoplatonizing theory of ”the purely free act” – an act which would be identical to the first unmoved substance by Aristotle, but will eventually overcome any kind of substance by Plotinus and his followers (see Appendix 1). After all, once a pure, non-sensible form (and a pure act, as identical to pure form) is allowed to count as a separate substance in Aristotle, this latter’s view and Plato’s appear more approximate than they do concerning, e.g., the theory of substance in Metaphysics Zeta. However, none of these points is uncontroversial enough to support the standard reading of the whole series of Lambda and De anima passages at issue. Still, Hicks believes that he can confirm Torstrik’s ‘restitution’ (as he calls it) at De anima Γ 5, 430a18, because he thinks that this does not depart from Aristotle’s usage – an assumption which probably needs to be reviewed. There are “plenty of examples”, Hicks says, of both nominative and dative forms. He obviously does not list the dative forms, which are so many; and for examples of the nominative he draws on three Metaphysics Lambda passages which he quotes ad loc.: 1072a25, 1072b27, 1071b20. But, as we shall see, no Lambda text about ἐνέργεια as a separate substance is so firmly established as to justify basing a different De anima text on it, with the active intellect of Γ 5 being ἐνέργεια at the nominative case. Hence, the other way around, it is impossible that De anima Γ 5 could help to support the nomina-

24

have the Arabic version from the Greek edited by Badawi, A. 1947: Aristu ‘inda l-‘Arab, Cairo, Aristu ‘inda l-‘Arab. Here the relevant passage (18,8–10) says: Wa-ka-ma¯ annahu fi‘lu azalı¯yu da¯’imu, ka-dß a¯lika huwa hø aya¯tu azalı¯yatu da¯’imatu. We also have a Hebrew version of the same by Mosheh Ibn Tibbon, from the Arabic, and a Latin version by M. Finzi from the Hebrew (these latter are edited by Landauer, S. 1903: Themistii in Aristotelis metaphysicorum librum 12 paraphrasis hebraice et latine (CAG V) 2, Berlin, see 21,22–24 and 24,14–16): U-kemo še-hu’ po‘el nisøhø iyyi matmid, kemo ken hu’ hø ayyim tamidim nisøhø iyyim: “Nec quandoque vivit, quandoque non vivit, nec diverso modo, perinde atque nos; sed ipse est ipsa vita, quoniam est actus, et actus est vita et eius perfectio”. I am especially grateful to Mauro Zonta for discussion on all of these texts. See e.g. Plotinus’ thoughts on Metaphysics Lambda at VI 1 15, VI.8 12, and Appendix 1 here below.

194

Silvia Fazzo

tive ἐνέργεια in Lambda.25 Let us therefore come back to the Lambda, starting from those passages which Hicks, the De anima’s editor, regarded as impressive enough to justify a textual adjustment at Γ 5, 430a18 as well.

VI. Two artefacts about the Unmoved Mover as a pure act? I shall then firstly explain why two well known Lambda passages – i.e. Λ 6, 1071b20, 22, and Λ 7, 1072b27 – do not testify that substance is a pure act, although both of them are called into question by Hicks as positive assessments of the first principle’s being act. In fact, their printed form is in part an artefact, in both cases. Moreover, it can be argued that neither of them refer to the Prime Unmoved Mover so that their meaning is not properly relevant to the central issue to be established, namely ‘is Aristotle’s Unmoved Mover an act?’ 1. Lambda 7, 1072b27 One of these alleged references to God as act can be found, judging from printed nineteenth and twentieth century editions, at 1072b27. In Ross’ edition (which is also the TLG text), the passage reads thus: “ἡ γὰρ νοῦ ἐνέργεια ζωή, ἐκεῖνος δὲ ἡ ἐνέργεια: ἐνέργεια δὲ ἡ καθ᾽ αὑτὴν ἐκείνου ζωὴ ἀρίστη καὶ ἀΐδιος.” Hence Ross’ Oxford Translation: “for the actuality of thought is life, and God is that actuality; and God’s self-dependent actuality is life most good and eternal.” But the reading is an artefact, because editors opt for the Ab reading ἐκεῖνος, whose antecedent is taken to be God in the previous line, whereas every other valuable manuscript of the Metaphysics apparently has ἐκεῖνο, a neuter pronoun (i.e. ἡ γὰρ νοῦ ἐνέργεια ζωή, ἐκεῖνο δὲ ἡ ἐνέργεια). It must say:

25

Hicks does not bring into the discussion Metaphysics Theta, although the most provoking nuances of the ἐνέργεια/δύναμις lexicon are apparently to be found in Theta 8. Still, it is possible that later Aristotelians took the path they took because they could find some lexical foundation in that book. At Theta 8, 1050b22–28, one can wonder whether it could be found a parallel supporting the current reading in De anima Γ 5.430a18. But on either case, οὐσία is taken with a depending genitive. οὐσία thus means the essence and quiddity of sensible beings. Therefore, the identification of a separate substance with act is not at issue. See moreover note 10 above.

Unmoved Mover as Pure Act or Unmoved Mover in Act?

195

“the Intellect’s ἐνέργεια is life: this (ἐκεῖνο) is (its) ἐνέργεια.”26 In this reading, which is original (in my view), ἐκεῖνο remains coupled with ἐνέργεια but not with intellect (νοῦς). The reading does not in fact allow to identify either ἐνέργεια or ἐκεῖνο (this latter, as referred to life, ζωή) with θεός or νοῦς. In particular, ἐνέργεια is disjoined with and differentiated from God: it is the ἐνέργεια of God. This is confirmed by the following sentence, where life is the very activity of his (ἐκείνου, i.e. God’s).27 Thus its eternal life (i.e., its proper ἐνέργεια) belongs (ὑπάρχει, 1072b26) to him. This overall meaning would yield a more natural sense as a whole, even if there were not a manuscript tradition pointing in that direction.

2. Lambda 6, 1071b22 As for 1071b19–22, we are in chapter 6, which is mainly devoted – always in my view and edition at least – to the heavens’ even and eternal movement; therefore, it is questionable whether any reference to an unmoved substance as such has direct support, either in the text, nor in the relevant part of the argument. This is Ross’ text: “δεῖ [20] ἄρα εἶναι ἀρχὴν τοιαύτην ἧς ἡ οὐσία ἐνέργεια. ἔτι τοίνυν ταύτας δεῖ τὰς οὐσίας εἶναι ἄνευ ὕλης: ἀϊδίους γὰρ δεῖ, εἴπερ γε καὶ ἄλλο τι ἀΐδιον. ἐνέργεια ἄρα.” The conclusion is here at issue point: ἐνέργεια ἄρα at 1071b22 in Ross’ edition. In fact, manuscripts E and J (the main manuscripts of the alpha family) bear a iota after ἐνέργεια,28 strongly supporting the dative option, ἐνέργείᾳ ἄρα, which means: “Therefore (they are) in act.”

26

27 28

Possibly, the reading ἐκεῖνο was abandoned because its reference is slightly opaque. Still, it has its place in the argument. It probably emphasizes ‘this’, i.e. life, ζωή, 1072b26, 27, as the noetic activity, in order to prepare for the following statement about God’s ζωή: this is found to be conceived in the way in which the activity of the νοῦς is life, ζωή: it is the best possible and eternal life (1072b28–30). On the other hand, the original neuter reading ἐκεῖνο does not directly refer either to God, θεός, in the previous line, 1072b25, or to νοῦς, in the same line 1072b27 – which makes ἐκεῖνο a lectio difficilior, and to some degree explains the success of ἐκεῖνος in printed editions. Therefore, ἐκείνου, at 1072b28, must have a different antecedent than the previous ἐκεῖνο, and this also helps to explain the success of ἐκεῖνος as a lectio facilior. Met. Λ 6, 1071b22: ἐνέργείᾳ EJ (ἐνεργειαι [sic] J): ἐνέργεια Ab MC Ross Jae Aru Arm ἐνέργειαι EΣγρ (quoted from Fazzo’s 2012 critical apparatus ad loc.).

196

Silvia Fazzo

Note that the previous occurrence of the word ἐνέργεια at 1071b20 is connected to οὐσία with a dependent genitive, i.e. essence’ rather than ‘substance’. I.e.: those beings which eternally act are such that acting is the very essence of them. The meaning of this latest expression must be gathered from Theta 8, a discussion, or a theory, to which Aristotle most probably refers here (see also 6, 1071b4, εἴρηται δὲ πῶς). οὐσία has a depending genitive and means ‘essence’. This makes a crucial difference from οὐσία as substance, let’s alone as a separate substance. Therefore 1071b20 counts as an argument in favour, not of the eternal substance at issue to be act, but rather about its being acting by essence and definition. The heavens are at issue, so this makes perfect sense. The point deserves attention because it is not uncontroversial. In chapter Lambda 6, Aristotle never talks about immobility but only about eternal heavenly motion as eternal actuality; as I argued elsewhere, the heavens are most likely to be at issue all the long of the chapter; this applies until 7, 1072a22, where the reference to ὁ πρῶτος οὐρανός is openly spelled out. Hence there is probably no obstacle to believing that the heavens are to be a subject of Chapter 6 as a whole. The chapter demonstrates an eternally moved substance (1071b12–22 especially), then recalls its (i.e. heaven’s) moving action on the sublunary world (end of Chap. 6, with open reference to De Gen. et Corr. II 9ss.). All of this (the heavens’ being moved and moving) is thus given as a premise. Therefore, no strict parallel can be drawn from Chapter 6 to legitimate a textual choice in favour of a pure act, a nominative ἐνέργεια, in the passage that shortly follows. This eventually concludes (at 7, 1072a24–26) that the First Mover of the universe is an unmoved, non-sensible, albeit always acting substance.

VII. A crux in twentieth century editions: Met. Λ 7, 1072a24–26 We come thus to our main point, Λ 7, 1072a26, without any previous occurrence of the First Mover, or of any separate substance, as ἐνέργεια.29 The internal connection in the argument must play a role for the text’s constitution. The current reading ἐνέργεια at 1072a26 is based on the previous alleged occurrence at 1071b22, which implicitly supports it. Once ἐνέργείᾳ is read at 1071b22, by contrast, it fully makes sense for ἐνέργείᾳ to be read at 1072a26 as well. This can be said as a preliminary remark. Still, admittedly, most words in the sentence at 7, 1072a24–26, which contain the central argument of Λ 6–7, have proved exceedingly difficult. (See Appen29

Another interesting passage further along at Met. Λ 8 1074a36 should probably be reconsidered on the basis of these words. Its exact meaning is not clear, especially since the passage

Unmoved Mover as Pure Act or Unmoved Mover in Act?

197

dix 2 below.) The main point is that the sentence at issue (introduced by τοίνυν at 1072a24 as a definitely conclusive particle) concludes the argument which is developed at length in the previous chapter. Concerning the word ἐνέργεια, one has to take into account the previous cases within the argument: 1071b22, 1072a5 f. In either case, the dative form (ἐνεργείᾳ) is attested in the manuscripts E and J (among others) and arguably preferable, as we have argued above concerning 1071b22 (see § VI.2). At 1072a25, the manuscripts give the word without subscript iota. This, once again, does not mean that one does not need to supply it (see above, § 1), but that one has to consider the argument as a whole.30 In the case at issue, the crucial reading in 1072a24–5 according to which “there is therefore an Unmoved Mover […] which is act” is meant to be justified by the former statement, if there really is one at 1071b22, that Unmoved Movers “are therefore act” (ἐνέργεια ἄρα). Without this previous statement, the τοίνυν conclusion, with ἐνέργεια in the nominative, is unexpected and unjustified. Above all, both in Λ 6 and 7, all of the substances at issue chiefly act. This is implied at 1071b12, 14, 17, 18. In this light, they must do so at 1071b22 as well. They act as any principle which always acts.

VIII. Epilogue With this contribution, while athetizing the (allegedly Aristotelian) pure-acttheory, I did not want to challenge the interest in it. This has been one of the most successful theories in the Aristotelian tradition. Its impact is impressive. It is even stronger than another well received idea, to which it was probably connected by Alexander of Aphrodisias, as we have seen above – namely, God’s, or the Prime principle’s, being pure form. As a matter of fact, it is hard to find a general account of Aristotle that lacks the definition of the Prime Mover as ‘pure act’. So far as an Aristotelian theology has attracted attention, the idea of God as an actus purus was no doubt regarded as a central feature of it.

30

was probably added from outside the argument’s main stream. It is not clear e.g. to what extent ἐντελέχεια there (whether at the nominative, as it appears, or at the dative, ἐντελεχείᾳ – a choice which is once more still open, in principle at least) does or does not have a meaning which is exactly identical to that of ἐνέργεια, and τὸ τί ἦν εἶναι to εἶδος. Here, the earliest manuscript (mid-ninth century), J, manifestly shows that the scribe hesitated between the dative and nominative (he refrained from adding accents, which was his standard way of suspending judgement). But already E, at the beginning of the following century, reads ἐνέργεια with the nominative (properispomenos in accent) and so does the subsequent tradition.

198

Silvia Fazzo

This was fortunate enough, and not without reason. Once the Unmoved Mover is identified as God, then we reach through this theory a basis for a rational theology: God is thus described, on the one hand, by the same standard lexicon in use in philosophy and physics, but, on the other hand, through a different syntax. This difference, whatever it may be, assigns to the very conception of the ‘divine’ a sensible axiological priority over all other kinds of substances. The latter can only have actuality, without ever being said to be ‘act’ or ‘actuality’ in themselves. As a result, the idea of a substance being ‘act’ or ‘actuality’, which does not have a proper explanation within Aristotle’s standard ontology, is felt to be prior to the idea of a substance being ‘in act’ or ‘in actuality’. Such a perception of axiological priority reinforces in its turn the actus purus theory. After all, there is no doubt that Aristotle’s Lambda would have had a much weaker impact if its value had been assessed according to its original intention. This especially applies if its intention was, as I argued it to be elsewhere, to construct a rival theory in concurrence with Academic theories of principles. There is definitely some contingency in Aristotle’s original reasons for writing the book. In Lambda, he has aimed to achieve what his contemporary fellow-Platonists failed to achieve (as he repeats over and over): a necessary proof concerning Unmoved, purely intelligible substances. These are reached by induction as substantial principles of both endless change and eternal permanency in the sublunary world: they permanently act, i.e., they are always in act. But if, and especially if, we find, at the end of the story, that the resulting theory, held to be Aristotelian throughout our history of ideas, is more appealing than this original theory which can be safely attributed to Aristotle, then there is an opportunity to consider some wider methodological problems which, it appears, are not always taken into account. On the other hand, it could very well be that the tradition of Metaphysics Λ, while adding some further mystical flavour to the theory of the Unmoved Mover, drops something down in its path. I refer at least to one interesting point, namely, to the role played by the substance-in-between, i.e., the eternal sensible unchangeable substance (de facto: the heavens, cf. 1072a22 f.) in the Prime Mover’s argument and in the structure of the book. The relevance of this intermediary substance, whose distinctive feature is to be eternally in act, was lost, or at least obscured, in the tradition.31 Its fate differed from that of the subscript iota of the dative. But the two losses are not entirely unrelated.

31

Fazzo 2014, “Introduzione”; § 3.3. Ead., Heavenly Matter in Aristotle: Metaphysics Lambda 2, in: Phronesis, 2013: 160–175.

Unmoved Mover as Pure Act or Unmoved Mover in Act?

199

Appendix 1 The first principle as ἐνέργεια in Michael of Ephesus, Plotinus and Averroes The idea of the first Mover as act (not yet: as ‘pure act’) is well established in Michael of Ephesus (alias pseudo-Alexander of Aphrodisias), as can be seen in his commentary to Met. Λ 6–7: “τὸ δὲ εἶδος νοητόν, ἐνέργειά τις ὄν” (CAG I, 687.6 Hayduck, “form is intelligible, since it is an act”, and passim). Correspondingly, however, it must be noted that Alexander’s Quaestio I.1 (a text which Michael appears to draw from, from 685.25 to 687.22) has ἐνεργείᾳ τι rather than ἐνέργεια τις, i.e.: “τὸ δὲ εἶδος νοητόν, ἐνεργείᾳ τι ὄν” (Suppl. Ar. II 2, 4.11 Bruns, “form is intelligible, since it is something in act”). As it appears, between Alexander and Michael (in the space of almost a millennium) substantial evolution in the perception of a rather Neoplatonizing idea of ἐνέργεια occurred: its introduction in Michael’s Lambda paraphrase did not especially require to be legitimated any longer. In the Lambda exegesis, among the Arabic Scholastics the same idea is always occasionally developed into the idea of a pure act, actus purus, fi‘lun mahø dø un. This is found in a passage of Averroes’ Long Commentary on the Metaphysics (Tafsı¯r ma¯ ba’d at-Tabı¯’at), 1568,12–13. There, Averroes explains Aristotle’s assessment: “and it [the principle] is a pure actuality, and everything which is pure actuality does not have in itself (any) potentiality” (fa-huwa fi‘lun mahø dø un wa-kullu ma¯ huwa fi‘lun mahø dø un fa-laysa fı¯hi quwwatun).32 In fact, the single parts of this paraphrases by Averroes are Aristotelian, but what need not be Aristotelian is the identity of the principle with its actuality. Probably, Averroes is led to introduce the notion at issue by the Matta¯ translation he relies on. Averroes is commenting upon the passage at 1071b19– 20 (“there must, then, be such a principle [ἀρχή], whose [ἧς] essence [οὐσία] is act [ἐνέργεια]”),33 which is altered or corrupt in the Matta¯ translation (1563,2 Bouyges). Matta¯ translates: “fa-yagˇibu idß a¯ anna yaku¯na mitßla ha¯dß a¯ l-mabda’i gˇawhar huwa fi‘l” 32 33

Bouyges, M. (ed.): Tafsı¯r ma¯ ba’d at-Tabı¯’at: Grand commentaire de la Métaphysique d’Aristote, Beyrouth 1952. Cf. Aristotle Lambda 6.1071b19 f.: δεῖ ἄρα εἶναι ἀρχὴν τοιαύτην ἧς ἡ οὐσία ἐνέργεια. This is correctly rendered in the more ancient translation by Ustath: fa-idß an yanbag˙ı¯ anna yaku¯na bad’ un mitßla hadß a¯ gˇawharu-h u fi‘l un, which closely reflects the original Greek phrase.

200

Silvia Fazzo

This can be translated word by word as “it is then necessary that like this principle a substance is act”, or better: “it is then necessary that a substance like this principle is act”. Thus, apparently, “whose” (ἧς) was neglected, or lost and the phrase could only be understood in this different way, as we have seen in Averroes’ paraphrase quoted above (1568.12–13 Bouyges). This gives a place in Averroes, too, to the idea of a substance that is act, or even, actus purus, here and further on.34 This idea of pure act, fi‘lun mahø dø un, does not seem to have a proper Aristotelian counterpart. It possibly echoes Avicenna instead.35 Among Greeks authors the idea of purity associated to ἐνέργεια is hardly found before Plotinus. Even then, the very idea of ἐνέργεια at issue is not Aristotelian. Plotinus does not claim that there is first substance which is act, but rather that the “purely free act” is not substance. In Enneads VI.8 [39] 16, Plotinus says that the νοῦς is ἐνέργημα – and more precisely: it is ἐνέργημα ἑαυτοῦ (Enn. VI.8 [39] 16, 16 f.). Further on, in Plato’s footsteps, he adds that νοῦς, in so far as it is ἐνέργημα, is beyond substance, ἐπέκεινα τῆς οὐσίας (VI.8 [39] 16.35, cf. Plat. Rep. 509b9, a parallel which shows that the Good is meant to be identical with νοῦς and ἐνέργημα). The point is strengthened at VI.8 [39] 20, where Plotinus contrasts ἐνέργεια with οὐσία, while stating the primacy of former on the latter. For he says: τελειότερον ἡ

34

35

See also ibid. 1599.7, which is the beginning of his Tafsı¯r in Arist. Met. Λ 7, 1072a30–b4, Averr. T 37, cf. R. Taylor, “Averroes on the Sharîah of the Philosophers”, in The JudeoChristian-Islamic Heritage: Philosophical & Theological Perspectives. Eds. Richard C. Taylor & Irfan A. Omar. Milwaukee, WI: Marquette University Press, 2012: 283–304, in part. 298 n. 11. Occurrences of fi‘lun mahø dø un have been pointed out by Rémi Brague in Avicennas’ commentary on the so called “Theology of Aristotle” – which is mainly made of Arabic adaptations from Plotinus – i.e. Avicenna, Al-Ta‘liqa¯t, passim and 32.25–33 in Badawi’s 1973 edition, see Brague’s review of C. Genequand, Alexander of Aphrodisias on the Cosmos (Leiden Brill 2001), in Studia Islamica 94, 2002: 17–18. As Peter Adamson notices (per litteras) the same Avicenna, when commenting on Aristotle’s Metaphysics, says that God is a pure good because He is always in act, but he apparently does not use the phrase “pure act”; Avicenna rather says “pure perfection” in paragraph 2 (kama¯l mahø dø ) (see Avicenna’s Shifa’: Metaphysics 8.6, § 3 in The Metaphysics of The Healing. Al-Shifâ: Al-Ilâhiyyât, ed. and trans. M. E. Marmura. Provo, Utah: Brigham Young University Press, 2005. I am grateful to Peter Adamson for his remarks on this points). This seems to conferm that Avicenna’s idea of pure act is not meant to be Aristotelian. Nonetheless, it apparently reached the Aristotelian commentary tradition later on. Among the reason for adopting it on Averroes’ part, one could consider Averroes’concern to show that the study of God as a cause is relevant to the science of metaphysics. So he associates God with a formal and final cause of being, in order to explain that relevance. As a matter of fact, in Arabic, “efficient cause” is rendered in Arabic as fa¯‘il, and the concept of God as act is very close to the concept of God as an efficient. See Peter Adamson’s paper “Averroes on Divine Causation” in P. Adamson/M. Di Giovanni, “Interpreting Averroes: Critical Essays” (forthcoming).

Unmoved Mover as Pure Act or Unmoved Mover in Act?

201

ἐνέργεια τῆς οὐσίας (VI.8 [39] 20.14, a passage with indirectly connects and associates ἐνέργεια with ἐνέργημα and νοῦς at VI.8 [39] 16.35). Hence Plotinus’ concept of a “purely free ἐνέργεια”, which is independent from substance: ἐνέργεια οὐ δουλεύσασα οὐσίᾳ καθαρῶς ἐλευθέρα (VI.8. 20.17 f.). Here, the idea of purity is contrasted with substance, and not directly with δύναμις. Therefore, Hadot is manifestly right when he notices that the Plotinian act at issue here «ne doit pas être entendu au sens aristotélicien»36 Now, if Plotinus has to be regarded as a source for the actus purus theory, an obvious question is how and when such a Neoplatonising theory was attributed to Aristotle, and when it was introduced in the exegesis of Metaphysics Lambda, as we have seen in Averroes’ exegesis of Aristotle’s Prime Mover at 1568, 1599 Bouyges. It is unclear whether or not any Arabic translations from Plotinus could have played a role. Enneads VI.8 [39] is not included among the sources of Lewis’ Plotiniana Arabica translated by in volume 2 of Henry-Schwyzer edition maior of the Enneads.37 Still, we know that what we have is incomplete so it is not at all unlikely that the original translation included VI.8 [39]. It has even been hypothesized that VI.8 [39] influenced the way that the Arabic Kindî-circle Adaptor of Plotinus understood the rest of the Enneads.38 In Arabic, the notion of a ‘pure act’, as referred to the Prime Mover, is also found in one passage of the Arabic translation of Alexander of Aphrodisias’ treatise On the principles of the Universe, § 64 Genequand. If authentic, this would be the first in time. On the one side, however, the passage is not entirely without suspicion: a Neoplatonizing lexicon could be introduced during the process of translation into Arabic, and moreover the passage is a trait d’union between two very different sections;39 as such, it could have been interpolated into the text. On the other side, if really Alexander used such an expression, it would be even clearer that Alexander distinguished the level of strict Aristotelian exegesis (as in Quaestiones I.1 and I.25, where Lambda’s intellect is called νοῦς κατ᾽ ἐνέργειαν, thus avoiding the nominative ἐνέργεια as directly referred to the first principle) from the level of his own

36

37 38 39

P. Hadot, Porphyre et Victorinus, 2 vols, “Études augustiennes”, Paris 1968, I 375. I am very indebted to Enrico Berti who firstly pointed me out the passages by Plotinus, Enn. VI 8, and the relevant remarks by Hadot. Plotini Opera – Tomus II: Enneades IV–V. Plotiniana Arabica ad codicum fidem Oxford 1964–82. R. C. Taylor, “Aquinas, the Plotiniana Arabica and the Metaphysics of Being and Actuality,” Journal of the History of Ideas 59 (1998a), 241–264. et dans la Quaestio I.1 d’Alexandre d’Aphrodise, in: Laval théologique et philosophique, 64, 681–690.

202

Silvia Fazzo

Aristotelianism which is highly responsible for later interpretations of Aristotle.40

Appendix 2 The crux philologorum at 1072a26 on the first Mover in act Concerning the text’s constitution of 1072a24–6 (see § VII above), various options have been considered for the intended wording and meaning of the text, since Brandis 182341 and Bonitz 1848. Ross and Jaeger decided that the text must be corrupt.42 Ross’ edition has an excision and two cruces (on both sides of τοίνυν). Jaeger’s edition has suspension dots within the text and a long-winded conjecture in the critical apparatus.43 In my edition, I follow manuscripts E (the first hand) and J, with no cruces nor brackets needed, so the passage reads as follows: 40

41

42

43

I am particularly grateful to Mauro Zonta, moreover and to Peter Adamson, Enrico Berti and Luca Bianchi for comments on the subject matter of this Appendix. Any errors or misunderstandings are all my own. Brandis has ἐπεὶ δὲ τὸ κινούμενον καὶ κινοῦν, καὶ μέσον τοίνυν ἔστι τι ὃ οὐ κινούμενον κινεῖ, ἀίδιον καὶ οὐσία καὶ ἐνέργεια οὖσα. This does not make entirely sense since the sought-for first and Unmoved principle (τι ὃ οὐ κινούμενον κινεῖ) is not supposed to be μέσον, but is the principle of the moved and moving principle, which is a μέσον. This problem prompted the excision of καὶ suggested by Bonitz and accepted by Ross. Moreover, Bonitz firstly points out a problem concerning the position of τοίνυν (see the following note), while suggesting the inversion of the τοίνυν ἔστι into ἔστι τοίνυν. Bonitz’ suggested reading would be then probably something like: ἐπεὶ δὲ τὸ κινούμενον καὶ κινοῦν μέσον ἔστι τοίνυν τι ὃ οὐ κινούμενον κινεῖ. But this does not produce a better Greek phrase, if τι (and not μέσον) is meant to be the subject of it, and if we consider that no punctuation was provided in ancient manuscripts (so to disjoin ἔστι from μέσον). The position of τοίνυν having been regarded as a major problem (cf. the previous note), see ad loc. Laks’ discussion in Id., Histoire, Doxographie, Vérité. Etudes sur Aristote, Théophraste et la philosophie présocratique, Louvain-la-Neuve, Peeters, 2007: 61–91 (a revised French version of Id. 2000: Metaphysics L 7, in: M. Frede/D. Charles (eds.), Aristotle’s Metaphysics Lambda. Symposium Aristotelicum, Oxford, 207–243). It seems to me that the importance of the question of whether or not τοίνυν can stand where it is in our passage has been overestimated. The transmitted position has some advantages in the sentence because it avoids ἔστι to be attached to the previous clause. As Alexandru points out, in later Greek at least, many examples can be found of τοίνυν at the beginning of a sentence (by the way, our τοίνυν is not at the beginning of a sentence, but of a main clause only), see S. Alexandru (Aristotle’s Metaphysics Lambda. Annotated Critical Edition based upon a Systematic Investigation of Greek, Latin, Arabic and Hebrew Sources, Brill, Leiden–Boston 2014: 139 (ad loc., although he still adopt Ross’ and Bonitz emendations). ἐπεὶ δὲ )τὸ κινούμενον καὶ οὐ κινοῦν ἔσχατον καὶ* τὸ κινούμενον καὶ κινοῦν [καὶ] μέσον, )ἕτερον (vel τρίτον)* τοίνυν ἔστι τι ὃ οὐ κινούμενον κινεῖ, ἀίδιον καὶ οὐσία καὶ ἐνέργεια οὖσα.

Unmoved Mover as Pure Act or Unmoved Mover in Act?

203

“ἐπεὶ δὲ κινούμενον καὶ κινοῦν καὶ μέσον, τοίνυν ἔστι τι ὃ οὐ κινούμενον κινεῖ, ἀίδιον καὶ οὐσία καὶ ἐνεργείᾳ οὖσα.” “And since it [i.e., the heavenly substance] moves and is moved and is intermediate, there is something which moves it without being moved, which is eternal, it is substance and it is in activity.” The meaning of the two clauses – the causal protasis and the main clause – is closely interconnected, so that the subject of the former is the object of the transitive predicate (κινεῖ) of the latter – a predicate whose eternal activation justifies the ἐνεργείᾳ-mode to be attributed to the subject of the main clause without restrictions. A comparison to Ross’ text and translation reveals more than one difference. Ross translates: “And since that which moves and is moved is intermediate, there is something which moves without being moved, being eternal, substance, and actuality”. His text has: “ἐπεὶ δὲ τὸ κινούμενον καὶ κινοῦν [καὶ] μέσον, †τοίνυν† ἔστι τι ὃ οὐ κινούμενον κινεῖ ἀίδιον καὶ οὐσία καὶ ἐνέργεια οὖσα.” Jaeger accepts from Ross that the text is corrupt and does not make any further attempt to defend its transmitted readings. He rather proposes in his critical apparatus the possible implementation of a supposed lacuna at Met. Λ 7, 1072a24–5. Both editors print τὸ κινούμενον, rather than κινούμενον, following part of the manuscripts and the second hand (E2) in E.44 Both editors print ἐνέργεια in the nominative in without discussion.45

Bibliography Alexandru, S.: Aristotle’s Metaphysics Lambda. Annotated Critical Edition based upon a Systematic Investigation of Greek, Latin, Arabic and Hebrew Sources, Brill, Leiden–Boston 2014. Badawi, A.: Aristu ‘inda l-‘Arab, Cairo 1947. Berti, E.: Unmoved Mover(s) as Efficient Cause(s) in Metaph. XII 6 th, in: M. Frede, D. Charles, eds., Aristotle’s Metaphysics Lambda, Oxford 2000, Oxford University Press, 2000, p. 181–206. 44 45

As Alexandru 2014: 99 ad loc. and Mauro Zonta (per litteras) notice, the without τό. In either reading, there is also a further peculiarity concerning the syntactic agreement of the participle οὖσα, which are pointed out with a few similar cases in Fazzo 2014 Commento al libro Lambda 2014, 215 f. while suggesting a philosophical interpretation of it.

204

Silvia Fazzo

Bouyges, M. (ed.): Tafsı¯r ma¯ ba’d at-Tabı¯’at: Grand commentaire de la Métaphysique d’Aristote, Beyrouth 1952. Brague, R.: review of Alexander of Aphrodisias on the Cosmos by Charles Genequand (Leiden Brill 2001), in: Studia Islamica 94 2002, 17–18. Burnyeat, M.: Kinesis vs. Energeia: A much-read passage in (but not of) Aristotle’s Metaphysics, in: Oxford Studies in Ancient Philosophy, Volume XXXIV(2008) 219–92: 256. Fazzo, S.: Aporia e sistema. La materia, la forma, il divino nelle Quaestiones di Alessandro di Afrodisia, “Pubblicazioni della Facoltà di Lettere e Filosofia dell’Università di Pavia”, Pisa 2002. Fazzo, S.: L’exégèse de Métaphysique Lambda dans le De principiis et dans la Quaestio I.1 d’Alexandre d’Aphrodise in Laval théologique et philosophique 2008, 607–626. Fazzo, S.: Il libro Lambda della Metafisica di Aristotele, Napoli 2012. Fazzo, S.: Heavenly Matter in Aristotle: Metaphysics Lambda 2, in: Phronesis 2013, 160–175. Fazzo, S.: Commento al libro Lambda della Metafisica di Aristotele, Napoli 2014. Fazzo, S./Zonta, M. Alessandro di Afrodisia. La provvidenza, a cura di S. Fazzo, traduzione dal greco di S. Fazzo, traduzione dall’arabo di M. Zonta, Milano: Rizzoli (BUR) 1998. Hadot, P.: Porphyre et Victorinus, 2 vols, “Études augustiennes”, Paris 1968, I 375. Harlfinger, D.: Zur Überlieferungsgeschichte der “Metaphysik”, in: P. Aubenque (ed.), Études sur la Métaphysique d’Aristote, Paris 1979, 7–36. Hayduck, M. (ed.): Simplicii in libros Aristotelis de anima commentaria edidit Michael Hayduck (= CAG XI), Berolini 1882. Hayduck, M. (ed.): Alexandri Aphrodisiensis in Aristotelis metaphysica commentaria, consilio et auctoritate Academiae Litterarum Regiae Borussicae edidit Michael Hayduck. (= CAG I), Berolini 1891. Henry, P./Schwyzer, H.-R. Plotini Opera, t. I–III, ed. Oxford 1964–82. Tomus II: Enneades IV–V. Plotiniana Arabica ad codicum fidem anglice vertit Geoffrey Lewis. Hicks, R. D.: Aristotle. De anima, with Translation, Introduction and Notes, Cambridge 1906. Laks, A.: Metaphysics L 7, in: M. Frede/D. Charles (eds.), Aristotle’s Metaphysics Lambda. Symposium Aristotelicum, Oxford 2000, 207–243. I am particularly grateful to Christoph Horn and to the colleagues of his seminar in Bonn with whom this paper was discussed in February 2013; to Enrico Berti, Carlo Maria Mazzucchi, Mauro Zonta, for discussion and comments, to Natasha Dubokovich, Francesco Bravi and Chiara Borsatti for their kind help and support.

Unmoved Mover as Pure Act or Unmoved Mover in Act?

205

Laks, A: Aristote, Métaphysique Lambda 7: une presentation. In: Id., Histoire, Doxographie, Vérité. Etudes sur Aristote, Théophraste et la philosophie présocratique, Louvain-la-Neuve, Peeters. 2007, 61–91. Marmura, M. E. (ed.): The Metaphysics of The Healing. Al-Shifâ: Al-Ilâhiyyât, ed. and trans. M. E. Marmura. Provo, Utah: Brigham Young University Press, 2005. Landauer, S.: Themistii in Aristotelis metaphysicorum librum 12 paraphrasis hebraice et latine (Commentaria in Aristotelem Graecae, V.2, Berlin 1903. Ross, W. D.: Aristotelis De Anima, Oxford 1956. Taylor, R. C.: “Aquinas, the Plotiniana Arabica and the Metaphysics of Being and Actuality,” Journal of the History of Ideas 59 1998, 241–264. Taylor, R. C.: “Averroes on the Sharîah of the Philosophers”, in The JudeoChristian-Islamic Heritage: Philosophical & Theological Perspectives. Eds. Richard C. Taylor & Irfan A. Omar. Milwaukee, WI: Marquette University Press 2012, 283–304

The Causality of the Prime Mover in Metaphysics Λ ALBERTO ROSS

I. Introduction The aim of this paper is to propose a critical revision of the traditional interpretation of Met. Λ, according to which the Aristotelian Prime Mover is the final cause that explains the eternal movement of the heaven, rather than a formal or an efficient cause postulated for the same purpose. The importance of this discussion is widely acknowledged, and it is a subject upon which there is no unanimous agreement among interpreters. The different accounts of the causal relationship between the Unmoved Mover and the physical world have given rise to one of the most recurrent controversies in the history of the transmission of the Corpus Aristotelicum, and it is a question that does not admit simple or simplistic answers. A good example of the long tradition of attempts to redress this exegetical problem is the discussion recorded by Simplicius in his commentary on Physics VIII. At the end of his paraphrase, the Greek commentator introduces a digression in the following terms: “Some think that Aristotle says the primary mover – which he hymns as mind, eternity and god – is only a final cause and not also an efficient cause of the world and in particular of the heaven, since it is eternal and consequently ungenerated. They think this because they hear him often saying that it causes motion as the object of love, and often celebrating it as a final cause. It is a good idea, then, to prove that here too he is consistent with his teacher in calling god not only a final cause but also an efficient cause both of the entire world and the heaven.”1 This is an ancient formulation of the question regarding the causality of the Prime Mover, and we find that it distinguishes between two aspects of the discussion that it is always important to differentiate. On the one hand, Simplicius refers to the dilemma of the kind of causality of the Prime Mover – 1

Simplicius: In Aristotelis Physicorum Libros Commentaria (2 vols.), Berlin: Reimer, 1882, 1360, 24–31.

208

Alberto Ross

in this case, whether he is a final or an efficient cause. On the other hand, he explains that the very defenders of the final cause hold that such causality is, in particular or especially, exercised upon the heavens. This distinction is highly relevant since it allows us to see that what was under discussion from Antiquity was not only the kind of cause exercised by the Prime Mover, but also the domain over which it extends. If we compare this formulation of the problem with the contemporary discussion, we will find that the terms are not exactly the same and the intentions are also different. Simplicius is talking about an efficient principle that is not only a “moving” cause, but also a “productive” one, and his main purpose is to illustrate the harmony between Aristotle and Plato. This is not the case for contemporary scholars, but it is interesting to note that the difficulties for a correct comprehension of the Aristotelian position on that point and the different alternative interpretations offered by the Corpus are present from Antiquity. I will return later to this passage in more detail, but now I just want to point out the long and ancient tradition of the controversy. The discussion about the kind of relation between the Prime Mover and the world has been subject to continuous revision. In the contemporary discussion about the causality of the Prime Mover, the main references are found between chapters 6 to 10 of Met. Λ, where the most relevant theses of this debate are contained. They are as follows: 1. Motion is eternal, and it implies that there must be an eternal substance.2 2. Motion is not explained by something that is capable of moving things or capable of acting on them if it is not actually doing so.3 3. There must be a principle whose substance is actuality.4 4. The eternal substance has no matter.5 5. There must be a mover that moves without being moved.6 6. The object of desire and the object of thought move in this way (without being moved).7 7. The eternal substance is necessary, good, a first principle, separate, impassive, unalterable and God.8 8. The heavens and the world of nature depend on this principle.9

2 3 4 5 6 7 8 9

See See See See See See See See

Met. Met. Met. Met. Met. Met. Met. Met.

1071b3–11. 1071b12–17. 1071b17–20. 1071b20–22. 1072a24–27. 1072a27–28. 1072b4 ff. 1072b13–14.

The Causality of the Prime Mover in Metaphysics Λ

209

9. The life of this principle is such as the best which we enjoy, and it is “thinking on thinking”.10 10. The universe is analogous to an army and a household.11 The logical reconstruction of these theses has generated all sorts of discussions. One of the most polemical issues associated with this topic is precisely the clarification of how this Prime Mover moves, being a contemplative intellect and having these characteristics. Unfortunately, this is a question that Aristotle does not answer explicitly in the text, so an exegetical exercise is necessary to illustrate the point.12 The order of the exposition in this work will run as follows: Firstly, I will present the traditional reconstruction of Met. Λ. Secondly, I will introduce the main objections to this interpretation and some of the new accounts of the causality of the Prime Mover, which are precisely elaborated from criticisms of the traditional reading of the text. Thirdly, I will offer a reply to these objections and to the new accounts that appeal to different parts of the Corpus. As a conclusion, I will try to prove that it is still possible to defend the traditional interpretation of Met. Λ, in spite of the strong objections against it.13

II. Traditional Interpretation of Met. Λ The so-called traditional interpretation of Met. Λ tries to explain the correct reconstruction of the aforementioned theses. In order to do that, it proposes that the Prime Mover causes eternal motion by being an object of love or desire, and that it implies the conclusion of its teleological causality. Its ousia is energeia, and its activity is pure contemplation. According to this interpretation, movement of the first sphere takes place because it imitates this Prime Mover’s activity. The soul of the first sphere, as an expression of love, gives rise to a physical image of eternal contemplation, where the physical image is the eternal movement of the sphere. The life of the Prime Mover is a continuous activity of pure thought, but the spheres cannot reproduce this, so

10 11 12 13

See Met. 1074b29–35. See Met. 1075a11 ff. See Laks, A. 2000: Metaphysics Λ 7, in: M. Frede/D. Charles (eds.), Aristotle’s Metaphysics Lambda. Symposium Aristotelicum, Oxford, 207–243, 221. An earlier version of this work appears in Ross, A. 2007a: Dios, eternidad y movimiento en Aristóteles, Pamplona and Ross, A. 2007b: La causalidad del Primer Motor en Met. XII, in: Diánoia, 59, 3–26.

210

Alberto Ross

they do the next best thing and perform the only perfect, continuous physical movement, i.e. movement in a circle. In this reading of the text, the Prime Mover appears as the final cause that explains the eternal motion of heaven. According to this interpretation, the Prime Mover moves the first heaven directly as its telos. Since it moves by inspiring love and desire, this implies that the first heaven is capable of feeling love and desire, that is to say, it has a soul. This seems to be supported by what Aristotle claims in De cael: the first heaven, the planets, the sun and the moon are living beings.14 Alexander of Aphrodisias was one of the first defenders of this interpretation,15 and after him, many ancient,16 medieval 17 and contemporary18 commentators have done the same with different refinements.19 Before Alexander, Theophrastus had offered an interesting testimony for corroborating this interpretation, as I will mention later.20 However, despite this long tradition of inter-

14 15

16 17 18

19

20

See De cael. 285a29 ff, 292a20 ff and 292b1 ff. See Alejandro de Afrodisia: Quaest. XVIII, p. 62, 16–34; y XIX, p. 63, 18–26; and also Berti, E. 2000a: Il movimento del cielo in Alessandro di Afrodisia, in: A. Brancacci (ed.), La filosofia in età imperiale. Le scuole e le tradizioni filosofiche, Napoles, 225–243. See Themistius: In Aristotelis Metaphyicorum librum XII paraphrasis hebraice et latine. Berlin, 1903, 19–20 and 31–55; and Simplicius: In Ph. (see note 1) 1360, 24–31 See Aquinas, Thomae Aquinatis: In duodecim libros Metaphysicorum Aristotelis expositio. Ed. R. M. Spiazzi, Turin–Rome, 1964, l. n. 1594–1595. See Ross, W. D. 1924 (ed.): Aristotle’s Metaphysics. A Revised Text with Introduction and Commentary, Oxford, cxxx; Reale, G. (ed.) 21968: Aristotele. La metafisica, Napoles, 588; Elders, L. (ed.) 1972: Aristotle’s Theology, A Commentary on Book L of Metaphysics, Assen, 35–43; Menn, S. 1992: Aristotle and Plato on God as Nous and as the Good, in: Review of Metaphysics, 45, 570–573; Natali, C. 1997: Causa motrice e causa finale nel libro Lambda Della Metafisica de Aristotele, in: Méthexis, 10, 105–123; Gómez-Lobo, A. 1994: Aristóteles y el aristotelismo antiguo, in: A. García Marqués/J. García Huidobro (eds.), Razón y praxis, Valparaíso, 65; Boeri, M. 1999: Una aproximación a la noción aristotélica de Dios, in: Tópicos: Revista de Filosofía de Santa Fé, 71–77; Sedley, D. 2000: Metaphysics L 10, in: M. Frede/D. Charles (eds.), Aristotle’s Metaphysics Lambda. Symposium Aristotelicum, Oxford, 327–336; Laks (see note 12), 242 and Laks, A. 2013: Cátedra Tópicos: Los motores inmóviles de Aristóteles: una introducción sencilla a un problema complejo, México, 14–22; Sharples, R. 2002: Alexander of Aphrodisias and the End of Aristotelian Theology, in: T. Kobusch/M. Erler (eds.), Metaphysik und Religion, Müchen– Leipzig, 1–40; Botter, B. 2005: Dio e Divino in Aristotele, Bonn, 191–195; Rashed, M. 2011: Alexandre d’Aphrodise. Commentaire perdu à la Physique d’Aristote (Livres IV– VIII), Berlin–Boston, 134–140; Gourinat, J. B. 2012: Le premier moteur selon Physique VIII et Métaphysique Λ: physique et philosophie premiere, in: M. Bonelli (ed.), Physique et métaphysique chez Aristote, 201–214 and Baghdassarian, F. 2015: La question du divin chez Aristote. Discours sur les dieux et science du principe, Louvain-la-Neuve, 161–196. A good summary of the differences among defenders of the traditional interpretation can be found in Berti, E. 1997: Da chi è amato il motore immobile? Su Aristotele, Metaph. XII 6–7, in: Méthexis, 10, 66–75. Theophrastus, Met. 5a23–b 10.

The Causality of the Prime Mover in Metaphysics Λ

211

preters, in recent years some scholars have disagreed with this reading and have offered some interesting objections to reject this account.21 In order to reconstruct this discussion, these arguments against the traditional interpretation of Met. Λ will be presented in the following section.

III. Objections to the traditional interpretation Most of the objections to the traditional interpretation reject the model of final causality as an explanation of the relation between the Aristotelian God and the world, and defend the identification of the Prime Mover as an efficient or formal cause. Some of the most important arguments adduced in support of these positions are the following: 1. The Prime Mover is designated by Aristotle as kine¯tikon and poie¯tikon in Met. Λ 6.22 Such a characterization is crucial in identifying what kind of cause the Prime Mover is. The suffix “ikon” indicates the capacity to do something, which is normally used in the Corpus to refer to an efficient cause.23 Hence, the Prime Mover must not be a final cause, but an efficient one.24 In reply to this objection, some defenders of the traditional interpretation have claimed that chapter 7 – as opposed to chapter 6 – introduces teleological language in order to clarify the question.25 However, this is not a real obstacle for interpreting the Prime Mover as an efficient cause, because there 21

22 23 24 25

See Broadie, S. 1993: Que fait le premier moteur d’Aristote? Sur la théologie du livre Lambda de la « Métaphysique », in: Philosophique de la France et de l’Étranger, 183, 375–411; Judson, L. 1994: Heavenly Motion and the Unmoved Mover, in: M. L. Gill/J. G. Lennox (eds.), Self Motion: From Aristotle to Newton, Princeton, 155–171; Kosman, A. 1994: Aristotle’s Prime Mover, in: M. L. Gill/J. G. Lennox (eds.), Self Motion: From Aristotle to Newton, Princeton, 135–153; Berti, E. 2000b: Unmoved mover(s) as efficient cause(s) in Metaphysics L 6, in: M. Frede/D. Charles (eds.), Aristotle’s Metaphysics Lambda. Symposium Aristotelicum, Oxford, 181–206 and Berti, E. 2012: The Finality of Arisotle’s Unmoved Mover in the Metaphysics Book 12, Chapters 7 and 10, in: Nova et Vetera, 10, 863–876; Bradshaw, D. 2001: A new look at the First Mover, in: Journal of the History of Philosophy, 39, 1–22; Salis, R. 2009: La causalidad del motor inmóvil según Pseudo Alejandro, in: Estudios de Filosofia, 40, 201–209; Stevens, A. 2011: La causalité de l’intellect dans la Métaphysique et le traité De l’âme, in: La causalité chez Aristote, ed. L. Couloubaritsis/ S. Delcomminette, Paris–Bruxelles–Vrin–Ousia, 125–137. See Met. 1071b12–17. See GC 324b13–14. See Berti 2000b (see note 21), 186; Bradshaw (see note 21), 7; and Salis (see note 21), 202. See Met. 1072a26–b4.

212

Alberto Ross

are other arguments that have the purpose of showing that this is not the correct reading of those texts. One of these arguments is the following: 2. The reference to objects of desire or love in chapter 7 does not necessarily imply that the Prime Mover is a final cause. On one hand, the adverb ho¯s in the expression ho¯s ero¯menon could just mean “as if”. In other words, the reference to the object of love or desire in the text could simply be an analogy to indicate the way it moves – without contact – but not the kind of causality of this principle of motion.26 On the other hand, there are other passages from the same chapter, as well as from Λ 10, that also seem to refer to an efficient cause and not to a final one (e.g. the arguments to prove that the Prime Mover has no magnitude).27 Furthermore, Aristotle says: “We must consider also in which of two ways the nature of the universe contains the good or the highest good, whether as something separate and by itself, or as the order of the parts. Probably in both ways, as an army does, for the good is found both in order and in the leader, and more in the latter”.28 This reference is taken as a sign of the efficient causality of the Prime Mover, an interpretation suggested by the metaphor of the leader. A third objection to the traditional interpretation goes as follows: 3. In order to explain the causal relation between the Prime Mover and the first moved, it is embarrassing for the traditional interpretation that Aristotle says nothing about “imitation” (mimesis) in the text: contemporary criticism of such a view would therefore assert that the traditional interpretation is nothing but a Neoplatonic reconstruction.29 In addition, we find the following objection: 4. Another problem with the traditional interpretation is that Aristotle never claims that more than one object is loved or desired in the production of the first motion.30 However, this interpretation assumes that the sphere-soul engages in a pair of activities of love or desire, one of which – the love for the Prime Mover – somehow gives rise to the other’s desire 26 27 28 29 30

See Berti 2000b (see note 21), 200–206 and 2012 (see note 21), 863–868; Salis (see note 21), 203; and Stevens (see note 21), 125–138. See Broadie (see note 21), 378–379; Bradshaw (see note 21), 8. The texts are: Met. 1072b30 ff, 1073a5–11, 1075a11 ff. Met. Λ 10, 1075a11–15. See Broadie (see note 21), 379; and Salis (see note 21), 204–206. See Met. 1072a26–27.

The Causality of the Prime Mover in Metaphysics Λ

213

for motion in the sphere.31 Therefore, there are actually two desires, not one. This reading of the text would not be the most economical one. Along the same lines, we have the following criticism: 5. Met. Λ 7 distinguishes two senses of ‘that for the sake of which’, i.e. as ‘objective’ (tinos) and as ‘beneficiary’ (tini) of action.32 It is clear that the Prime Mover is not the beneficiary of the motion of the first sphere. However, the imitation story is not easy to link with our understanding of ‘objective’.33 If the sphere-soul’s love of divine contemplation takes the form of an impulse to imitate it in a physical medium, then in relation to this imitative activity, the divine contemplation is not really an objective one. Strictly speaking, divine contemplation would not be the final cause of motion, but rather an exemplary cause.34 Furthermore, there would be another problem: 6. If this were the case (i.e. if the Prime Mover were an exemplary cause), although the motion of the sphere depends on the highest activity as final cause (in the odd sense of ‘exemplary’), then the cosmos would be alive forever with order and beauty unfounded on the good. Thus, it is not certain that the processes within the cosmos take place because they or their ends are fair and good, since the goodness of the principle on which everything else depends is ultimately irrelevant to its existence.35 A final objection to the traditional interpretation of Met. Λ would be the following: 7. If the Prime Mover is the primary object of thought as well as desire, it would imply that the first heaven is capable of some kind of contemplation. But why, then, does it not imitate the Mover in the most direct way possible, i.e. by contemplating that which the Mover contemplates? An imperfect contemplation would be a closer approximation to the perfection of the divine life than spatial movement.36

31 32 33 34 35 36

See Broadie (see note 21), 380; and Bradshaw (see note 21), 7. See Met. 1072b1–3. See Broadie (see note 21), 382; Berti 2000b (see note 21), 187 and 201. See Broadie (see note 21), 382. See Broadie (see note 21), 384; Berti 2000b (see note 21), 201; and Bradshaw (see note 21), 8. See Bradshaw (see note 21), 8.

214

Alberto Ross

These are some of the main criticisms of the traditional account of Met. Λ. In addition to these, objectors and other commentators have introduced new accounts of the text. S. Broadie, A. Kosman and A. Stevens, for example, have suggested that the Prime Mover is a form or a certain kind of soul.37 They hold this, despite Aristotle’s contention that it is impossible to explain the eternity of change from a cause that is capable of accidental motion, which seems to be the case with a soul.38 Nevertheless, in favor of this position, it is possible to say that a divine self-moving principle would not have this problem in particular; first, because it would not be subject to the causal agency of a periekhon39 (because the periekhon would be itself) 40 and, second, because this lack of periekhon also implies that there is no surrounding environment in relation to which it might be said to move accidentally from one place to another. It is important to remember that the motion of the sphere is self-contained because of its circular, eternal and continuous nature.41 Thus, it seems possible to say without contradiction that the Prime Mover is a form or a certain kind of soul. I will return to this argument later. L. Judson has introduced another interpretation. He disagrees with the idea of a Prime Mover as a self-mover. Judson asserts that the Prime Mover is a transcendent entity quite distinct from the heavenly sphere which causes its motion,42 although like Broadie, Kosman and Stevens he rejects the final cause as an explanation of the relation between heaven and the Prime Mover.43 He supports the idea that Aristotle was entertaining a notion of an efficient cause sufficiently wide to subsume a final cause. On the basis of the distinction between proximate and remote causes, he claims that desire is the efficient proximate cause of motion; the object of desire (the Unmoved Mover) would be its remote efficient cause. E. Berti has a similar opinion. He admits a total independence of the Prime Mover from the world.44 Berti defends the claim that the reference to the objects of love and desire is only a comparison that notes how both the heavenly mover and the object of human desire move while remaining unmoved and move by means of something that is moved, but the metaphor would in no way be a reference to a teleological account. The Unmoved

37 38 39 40 41 42 43 44

See Broadie (see note 21), 387; Kosman (see note 21), 139; and Stevens (see note 21), 125– 138. See Phys. 258b13–16, 259b7–37. See Kosman (see note 21), 142. See De cael. 279a24, 278b23, 284a7. See Kosman (see note 21), 146. See Judson (see note 21), 155–157 See Judson (see note 21), 164–167. See Berti 2000b (see note 21), 202 and 2012 (see note 21), 865–868.

The Causality of the Prime Mover in Metaphysics Λ

215

Mover would be an efficient cause, and a necessary principle as a condition of the good of the heaven and other things. The Prime Mover permits them to attain their good (i.e. the circular movement or eternal reproduction), and it exists in the highest possible state, so it is an end, but just for itself. Other interpreters, just as D. Bradshaw, maintain that the efficient causality of the formal and final cause and its connection with the noetic activity of the Prime Mover is the key point that no one has made adequately clear.45 He asserts that the Prime Mover “thinks itself” only in the sense that all active intellect thinks itself, and that the direct objects of its contemplation are the forms.46 Yet, because of the identity of intellect and its object, the Prime Mover simply represents the forms. The Prime Mover would be the forms as self-subsistent, so it would move not only the first heaven, but also all things, as an object of love insofar as all things aspire to realize their proper form.47 These are some of the main arguments against the traditional interpretation of Met. Λ and some of the recent accounts of the text. Having presented these objections, in the following section I will attempt a reply to the criticisms listed above.

IV. Defense of the traditional interpretation of Met. Λ The order of exposition in this section will be as follows: First, I will attempt to prove that there is no problem in accepting that the Prime Mover, being an object of love, desire or intellection is a final cause (attending to objections (3), (4), (5), (6) and (7)). Second, I will offer an explanation in order to show the compatibility of the traditional account of Met. Λ, and those passages that seem to refer to an efficient cause (attending to objections (1) and (2)). To begin with objection (3), it is true that Aristotle never states explicitly that the first heaven “imitates” the activity of the Prime Mover. In addition, regarding objection (5), it is true that the Prime Mover according to the classical interpretation would seem to be an “exemplary” cause. However, it is possible to mention some passages from other parts of the Corpus where Aristotle introduces something like an “exemplary” cause as a final cause, and some passages that suggest that the first heaven imitates the activity of the Prime Mover, despite the fact that there is no explicit reference to this type of account in Met. Λ. It is accepted, as Philoponus says, that Aristotle 45 46 47

See Bradshaw (see note 21), 18. See Bradshaw (see note 21), 12–13. See Bradshaw (see note 21), 15.

216

Alberto Ross

does not distinguish the “exemplary” cause as an independent kind of aitı¯a,48 but it is possible to discern Aristotle’s opinion of this kind of explanation from some passages that I will refer to below. The first text that helps us to understand the nature of this kind of account is a well-known passage of De an. II 4: “(…) the most natural act is the production of another like itself, an animal producing an animal, a plant a plant, in order that, as far as its nature allows, it may partake in the eternal and divine. That is the goal towards which all things strive, that for the sake of which they do whatsoever their nature renders possible. The phrase ‘for the sake of which’ is ambiguous; it may mean either the end to achieve which, or the being in whose interest, the act is done. Since then no living thing is able to partake in what is eternal and divine by uninterrupted continuance (for nothing perishable can for ever remain one and the same), it tries to achieve that end in the only way possible to it, and success is possible in varying degrees; so it remains not indeed as the self-same individual but continues its existence in something like itself – not numerically but specifically one.”49 It is true that the passage does not refer to the imitation of God or the Prime Mover by the heavens, as Berti has remarked.50 However, the text is useful to illustrate the Aristotelian use of explanations from something similar to an “exemplary” cause. It talks about the reproduction as a way to partake in the eternal and divine. This is not a minor point, because it implies two consequences: (i) Aristotle recognizes the relationship between two different kinds of substances or realities (the sublunary and the heavenly ones) in terms of imitation, and (ii) this occurs in the context of a teleological account. In fact, De an. II 4 introduces the same senses of ‘for the sake of which’ as Met. Λ 7. In addition to this reference, we find a similar idea in De gen. et corr.: “God therefore adopted the remaining alternative and fulfilled the perfection of the universe by making coming-to-be uninterrupted; for the greatest possible coherence would thus be secured to existence, because that coming-to-be should itself come-to-be perpetually is the closest approximation to eternal being. The cause of this as we have often said is circular motion; for that is the only motion which is continuous. That, too, is why all the other things – the things, I mean, which are reciprocal48 49 50

See Philoponus, In Phys. 241, 15–19. De an. 415a26–b7. See Berti 2000b (see note 21), 201.

The Causality of the Prime Mover in Metaphysics Λ

217

ly transformed in virtue of their qualities and their powers, e.g. the simple bodies – imitate circular motion. […]. Hence it is by imitating circular motion that rectilinear motion too is continuous.”51 In this text, the relationship between the corruptible world and the incorruptible one in terms of imitation is also confirmed. In this case, the “example” or the “model” consists in simple bodies imitating the circular motion of the heavens. Thus, it seems to be true from both passages that Aristotle uses explanations that appeal to an “exemplary” cause, and that this kind of explanation is a teleological account for him; it is not necessarily a Neoplatonic reconstruction. In order to link these passages with Met. Λ, it is important to remember what kind of relationship is assumed by Aristotle when he speaks of different types of substances. He claims in Λ 7 that there must be a mover which moves without being moved, and there must be an eternal and active substance in order to explain the unceasing motion of the heavens.52 We find a similar idea in Phys. VIII 6, where he says that the Prime Mover imparts motion always in the same way because it does not itself change in relation to that which is moved by it.53 In contrast, a moved mover would impart another kind of motion because it stands in varying relations or positions regarding the things that it moves, and so produces contrary motions.54 According to these passages, there is a causal connection between the Prime Mover, the heavens and the sublunary world, thanks to their similar but nonidentical properties. Eternity subsists in all of them but in a different way, and that is what makes their causal relationship possible. The main difference between Met. Λ and the theory of Phys. VIII would be the kind of causality attributed to the Prime Mover. If there is an analogical relationship between the connection of the sublunary world and the heavens and the connection of the heavens and the Prime Mover, it is possible to say that the heavens imitate the latter, despite the fact that Aristotle does not use the term mime¯sis in Lambda. The passages cited from Met. Λ 7 and Phys. VIII 6 suggest that these relationships are analogical in the Aristotelian account of motion. In regard to the kind of imitation that takes place (i.e. the circular motion of the first heaven), it is necessary to add something. It is possible to find two indirect but important references that could support the traditional interpretation on this point. First, Plato’s mention of circular movement as the

51 52 53 54

De gen. et corr. 336b27–337a7. See Met. 1072a19–26. See Phys. 260a5. See Phys. 260a7–10.

218

Alberto Ross

model of human reflection.55 This reference is relevant because it shows that the analogy between thinking and circular motion was not unusual in Greek thought. Second, if it is correct to take Theophrastus’ Metaphysics as a critical reaction against part of Aristotelian philosophy, then this provides an indirect support for the traditional interpretation.56 In this book, Theophrastus asserts that “it is hard to see how it can be that, though the heavenly bodies have a natural desire, they pursue not rest but motion”57 and “if the Prime Mover is the cause of the circular motion, it will not be the cause of the best motion; for the movement of the soul is better, and first and above all that of thought, from which also springs desire”.58 From both references, it is possible to infer that circular motion was a reasonable way to imitate the Prime Mover for Aristotle. Indeed, according to Theophrastus, it was a critical point of his theory, so it is possible to say that it is not just a Neoplatonic interpretation. Bradshaw’s objection (7) would be, in fact, an objection to the Aristotelian position itself, not to its classical interpretation. If what we have said is true, objection (6) – about the irrelevant existence of the Prime Mover – is no longer a problem, despite the fact that it provides a good reason for suspicion regarding the truth of the Aristotelian position, as Theophrastus’ objection likewise does. As for objections (4) and (5) – the number of objects of love and the two kinds of ‘for the sake of which’ – it is necessary to refer to some of our previous remarks. It seems that Aristotle had two different kinds of teleology in mind. One of them is intrinsic, and the other is extrinsic.59 In virtue of the first one, every natural substance looks for its own perfection, and in virtue of the second one the cosmos is ordered. Again, we can refer to the passages De an. II 4 and De gen. et corr. II 11. Thus, if it is necessary to admit that the traditional account of the relationship between the Prime Mover and the first heaven implies two different desires, it is necessary to say that this is also true of all teleological processes.60 The soul of the first heaven would have two desires, in the same way that living things wish to reproduce them-

55 56 57 58 59 60

See Plato, Tim. 34 a; and Leg. 898 a. See Ross, W. D./Forbes, F. H. (eds.) 1929: Theophrastus. Metaphysics, Oxford, 43. Theophrastus, Met. 5a23–25. Theophrastus, Met. 5b7–10. See Vigo, A. 1994: Naturaleza y finalidad, in: A. García Marqués/J. García Huidobro (eds.) 1994: Razón y praxis, Valparaíso, 49–52. On the parallelisms between Met. Λ and other parts of the Corpus concerning teleology see Fazzo, S. 2004: Sur la composition du traité dit motu animalium: contribution à l’analyse de la théorie aritotélicienne du premier moteur, in: A. Laks/M. Rashed (eds.), Aristote et le mouvement des animaux. Dix études sur le De motu animalium, Paris, 203–229, especially 227–229.

The Causality of the Prime Mover in Metaphysics Λ

219

selves and to partake in what is eternal and divine, or in the same way that simple elements tend to engage in reciprocal transformations and at the same time imitate the eternal motion of the heavens. Strictly speaking, Aristotle does not talk about two desires, but about two aspects of one and the same desire.61 As for the compatibility of this account with the passages of this book that seem to refer to an efficient cause, some observations would be appropriate. First, we have the passage Λ 6, where Aristotle states that if there is something that is capable of moving (kine¯tikon) things or capable of acting (poie¯tikon) upon them, but is not actually doing so, there will be no movement.62 Some commentators understand these lines as a reference to the efficient causality of the Prime Mover. However, there are several reasons to differ. First, it is crucial to notice the negative framing of this passage. In other words, those lines refer only to what the Prime Mover is not. The argument asserts: 1. If there is something capable of moving things or acting upon them and it is not actually doing so, there will not be movement. 2. It is not the case that there is no movement. 3. Then, there cannot be something capable of moving things or acting upon them and not actually doing so. According to this reconstruction, it is possible to support an economical reading of the argument. What is important here is to emphasize what kind of principle is not capable of explaining the existence of motion, and this purpose is not a positive description. From the negation of the consequent follows the negation of the antecedent – and that is all. Hence, the description of a principle that is capable of moving and acting and is not doing so, could be just a description aimed at denying its existence, and it is unnecessary to link it with the nature of the Prime Mover. Actually, the Platonic forms provide according to Aristotle one such instance.63 A similar explanation can be given for the arguments aimed at demonstrating that the Prime Mover has no magnitude.64 It is perfectly possible to conceive an argument that proves, first, that a body or a magnitude cannot move eternally and,

61

62 63 64

Barbara Botter introduces another answer to this objection from a different version of the text. This version is taken from Alexander’s Commentary on Metaphysics referred to by Averroes (see Botter (see note 18), 193). See Met. 1071b12–17. See Met. 1071b12–17. See Phys. 266a12–266b20.

220

Alberto Ross

second, that a final cause explains the eternal motion without itself being moved. The first demonstration is negative, so it does not have to determine the description of the posterior cause mentioned. In addition to these considerations, it is true that the terms kine¯tikon and poie¯tikon usually refer to an efficient cause in the Corpus, but this is not always the case. There are some parts of the Corpus where Aristotle talks about two different senses of kinou¯n. The clearest is the passage of Phys. VII 2, where Aristotle says: “that which is the first mover of a thing – in the sense that it supplies not that for the sake of which, but the source of the motion – is always together with that which is moved by it”.65 Here, kinou¯n can bear two meanings, i.e. as an efficient causa and as final causa. If we accept that kine¯tikon and poie¯tikon are synonyms in Met. Λ 6 and that kine¯tikon and kinou¯n are the same (the first one in potency and the second one in action), this reference does not necessarily give a definitive indication of the efficient causality of the Prime Mover. André Laks has pointed out in this connection that the suffix “-ikon” is less constraining than criticisms of the traditional interpretation suggest.66 It indicates, in general terms, that there is a certain relation between the radical and what is said to be “-ikon”. Laks also observes that there is a disjunction (e¯) between the two terms that could introduce the second term as either an alternative or an equivalent one. The opponents of the traditional interpretation assume immediately the second option, but it is possible to read the text in the first way. In any case, the term kine¯tikon would refer to the first mover as a principle of motion in general and not specifically to the efficient cause. In regards to the use of the adverb ho¯s in Λ 7, it is true that this expression could just mean “as if”. In this sense, the formula ho¯s ero¯menon would have no more than a metaphorical value. However, it is also correct to read it with a modal force, i.e. meaning “in so far as”.67 The first option makes sense, as long as we accept that the Prime Mover introduced earlier is an efficient cause. But if this is not the case, as I have attempted to prove, it is better to adopt the most natural reading of the text, i.e., that the Prime Mover is an object of love and desire, so it moves in so far as it is loved and desired. In order to support that point and in addition to the arguments already mentioned, we could return to the passage of Simplicius referred to at the beginning of this work.68 According to the text, some think that Aristotle says that the Prime Mover is only a final cause because they hear him 65 66 67 68

See Phys. 243a32–33. See Laks (see note 12), 242 and Fazzo, S. 2014: Commento al libro della Metafisica di Aristotele, Napoles, 290–295. See Laks (see note 12), 221. See Simplicius: In Ph. (see note 1) 1360, 24–31.

The Causality of the Prime Mover in Metaphysics Λ

221

often saying that it causes motion as the object of love, and often celebrating it as a final cause. It is interesting to notice the affirmation of Simplicius about the origin of the controversy. According to the Greek commentator, the reason this problem arises within Aristotelian doctrine is the thesis of Metaphysics Λ that describes the Prime Mover as a principle that moves ho¯s ero¯menon. Simplicius’ text makes no direct reference to this book of Aristotle’s, but rather to a thesis that has been heard from him many times (pollakis). Nevertheless, it is well known to us where such a description appears and it was also thus for the ancients.69 So what can be established with certainty is that, from Antiquity, the expression ho¯s ero¯menon was generally read it with a modal force, i.e. meaning “in so far as”. Thus, Simplicius noticed a tension between what is asserted in the Metaphysics and the possible attribution of an efficient causality to the Prime Mover in order to defend the harmony between Aristotle and Plato. Lastly I would like to mention the analogies of Met. Λ 10 between the universe and an army and a household. This part of the book seems to consolidate the account of chapters 6 to 9 by a switch of focus: it no longer deals with the specific intellectual activity of the Prime Mover, but rather with its characterization as a unifying final cause. It starts with a reference to “the nature of the universe”, and continues by describing a stratified “jointarrangement” of the entire world. This issue admits also a minimalist reading. The nucleus of the analogy could be just the fact that, in an army, the general does not depend on order, but that order depends on him; furthermore, everything in the universe is ordered together, as in a household. Both analogies can reflect the causal power of the Prime Mover as a final cause. It is responsible for the order of the cosmos because thanks to it the heavens are moved eternally, and thanks to this movement the eternal succession of coming-to-be and passing-away is possible. The similarity of the different parts of the universe reveals its unity and this order depends on its main element, i.e. the Prime Mover. Aristotle states in Met. Λ 8 that the motions of the heavens are jointly arranged, so he has to focus his attention on the sublunary world. Sublunary things live in varying degrees “towards the whole”, which is compared to the way in which various members of the household act in different degrees “towards what is communal”. The level of participation is higher among free men than among slaves, but the order in the house implies that the high

69

It is commonly accepted that Simplicius knew this text and it has even been suggested that he may have written a commentary on it. See Hadot, I. 1987: Simplicius, sa vie, son œuvre, sa survie, Berlin–New York; and Baltussen, H. 2008: Philosophy and exegesis in Simplicus, London.

222

Alberto Ross

level of perfection is related to the low level of randomness. The universe works in a similar way. Randomness exists only in the sublunary world and hence has no place in heaven. Nonetheless, if the Prime Mover does not move, the remaining relationships do not work. Therefore, it is possible to say that the order depends on it, and that the analogies of Λ 10 are perfectly compatible with the traditional interpretation of Lambda; it is not necessary to look for an efficient cause.

V. Commentaries on the new accounts of Met. Λ I have attempted to answer the most important objections against the traditional interpretation of Met. Λ. Now, I will introduce some remarks related to new accounts of the Prime Mover, particularly, the following: (i) The Prime Mover is the remote efficient cause of motion. (ii) The Prime Mover is the soul of the first heaven. (iii) The Prime Mover is the same as the forms of the natural substances. In regards to (i), Aristotle distinguishes between proximate and remote causes; this distinction is, however, valid in a series composed of the same kind or species of cause. The causal modalities (tropoi) mentioned in Phys. II 3 and Met. V 2 are distinctions within the same kind (eidos) of cause.70 Thus, it seems to me that it is not easy to apply this distinction in Met. Λ with respect to “energetic” or “nonenergetic” causes. If they are, strictly speaking, two different kinds of aitiai, the distinction between proximate and remote in them will not apply. As for (ii), it is important to recognize that it implies a more economical interpretation than the traditional one. In its defense, the argument of the lack of a periekhon is offered, in order to emphasize that this soul would not be affected by it – because it does not exist – and it would not have accidental motion because there is no surrounding environment in relation to which it might be said that the Prime Mover is moved accidentally from one place to another. Despite this fact, I would like to introduce three considerations: First, in explaining the eternity of motion, Aristotle finds a problem not only in the external influence of a periekhon, but also in the things that happen within the hylomorphic substance, because animals move themselves only with one kind of motion (i.e. the local one).71 The natural motion “within” 70 71

See Phys. 195a26 ff. and Met. 1013b28 ff. See Phys. 259b6–20.

The Causality of the Prime Mover in Metaphysics Λ

223

animals affects the causal power of the soul and it could be the case of the soul of the universe, in spite of the lack of a periekhon. This account is taken from Phys. VIII 6, but there are no reasons to suggest it does not work in Met. Λ. Second, there is another problem about (ii). This account introduces a new complication in that the immobility of the Prime Mover would not depend on itself, but on the lack of a periekhon. In this case, the Prime Mover could itself be affected by something external, but that is not the case since the external factor is not present. The Prime Mover would be absolutely unmoved by accident, and not by itself. Finally, it is important to take into consideration that Aristotle mentions, in Phys. VIII and Met. Λ, that the Prime Mover must be separated and unmixed (as the “nous” referred to by Anaxagoras).72 In contrast, a soul must be mixed with a body. Thus, if we say that the Prime Mover is a soul, in the sense that it is an immaterial principle connected to a body but not mixed with it, then there is no real difference between this account and the traditional one. Concerning (iii) (i.e. the identification of the Prime Mover with the forms of the natural substances), it is worthwhile to take into consideration the following: If this were the case, the Prime Mover would be the proximate cause of all types of motion in view of the different types of form. However, this contradicts the observation of Phys. VIII 6 and Met. Λ 7, according to which the Prime Mover imparts motion always in the same way because it does not itself change in relation to that which is moved by it, and in contrast, the moved mover imparts another kind of motion because it stands in varying relations to the things that it moves, so it produces contrary motions.73 Actually, Aristotle himself explains in another passage why the set of souls cannot be the main explanation of the eternity of motion, considered as something continuous and eternal.74 Bradshaw justifies his account with a reference to Phys. II 7, where Aristotle says that form, mover and ‘for the sake of which’ often coincide.75 If it is used to assimilate the final causality of the Prime Mover with its efficient power, the reference is problematic. In fact, the text refers to an identification of the causes in specie, not in number (for example, a man generates a man),76 and it only implies that an efficient cause shares the same species with its effect. It is not enough to enable us to identify the Prime Mover as an efficient cause and not a final one. However, there is another text that could be used in order to defend thesis (iii). In De an. II 4, Aristotle presents the soul in these terms: It is a 72 73 74 75 76

See See See See See

Phys. 256b20–27 and Met. 1072b4 ff. Phys. 260a5–10. Phys. 258b16 ff. Bradshaw (see note 21), 18–19; the reference is to Phys. 198a25–26. Phys. 198a26–27.

224

Alberto Ross

formal cause because “the essence is identical with the cause of its being, and here, in the case of the living things, their being is to live”.77 In another sense, the soul is a final cause because “all natural bodies are organs of the soul”.78 Finally, the soul is conceived as an efficient cause because it is “the original source of local movement”79 and the same goes for all other qualitative or quantitative changes.80 Thus, Aristotle can say without contradiction that the soul is form, mover and ‘for the sake of which’, because the issue explained (the explanandum) is different in each of the three cases. However, the Prime Mover in Phys. VIII, as also in Met. Λ, explains the same issue, i.e. eternity of motion. Hence, there is just one explanandum, and so in this case it is not possible to use the same kind of description that Aristotle introduces in De an. II 4 in order to explain the causal power of soul. Then, if the Prime Mover is the cause of the eternity of motion and nature depends on it just in this way, then the Prime Mover must be one, and one only, of the following causes: efficient, formal or final. In view of the replies to objections (1) to (7) presented in the previous section and the arguments against positions (i) to (iii) presented here, I think that it is possible to assert that the traditional interpretation of Met. Λ could still be acceptable, i.e. that the Prime Mover is a final cause. Perhaps it implies an exegetical reconstruction, but the nature of the text seems to demand an answer of this kind. Obviously, this conclusion maintains the tension between the accounts of the Prime Mover in Phys. VIII and the version of Met. Λ. The resolution to this problem is an undoubted advantage of the interpretations that I set out to reject in this work. However, it is also true that the traditional reconstruction of the Lambda seems to be an answer to the philosophical problems inherited from the Physics, and this is not a minor point. The Unmoved Mover of Phys. VIII moves without affection and this is possible because the First Mover is without magnitude.81 How does it happen? It seems that Phys. VIII does not have a satisfying answer to this question, but Met. Λ offers an original alternative in order to fill this gap in our comprehension of Aristotle’s metaphysics.

77 78 79 80 81

De an. 415b12–13. De an. 415b18–19. De an. 415b21–22. See De an. 415b23–27. Phys. 267b2–4.

The Causality of the Prime Mover in Metaphysics Λ

225

Bibliography Alexander of Aphrodisias: Scripta Minora. Questiones, De Fato, De Mixtione (Suplementum Aristotelicum 2.2). Ed. I. Bruns, Berlin 1892. Baghdassarian, F.: La question du divin chez Aristote. Discours sur les dieux et science du principe, Louvain-la-Neuve 2015. Baltussen, H.: Philosophy and exegesis in Simplicus, London 2008. Barnes, J.: The Complete Works of Aristotle, Princeton 1995. Berti, E.: Da chi è amato il motore immobile?: Su Aristotele, Metaph. XII 6– 7, in: Méthexis 1997, 10, 59–82. Berti, E.: Il movimento del cielo in Alessandro di Afrodisia, in: A. Brancacci (ed.), La filosofia in età imperiale. Le scuole e le tradizioni filosofiche, Napoles 2000a, 225–243. Berti, E.: Unmoved mover(s) as efficient cause(s) in Metaphysics L 6, in: M. Frede/D. Charles (eds.), Aristotle’s Metaphysics Lambda. Symposium Aristotelicum, Oxford 2000b, 181–206. Berti, E.: The Finality of Arisotle’s Unmoved Mover in the Metaphysics Book 12, Chapters 7 and 10, in: Nova et Vetera 2012, 10, 863–876. Boeri, M.: Una aproximación a la noción aristotélica de Dios, in: Tópicos: Revista de Filosofía de Santa Fé 1999, 63–89. Bonelli, M. (ed.): Physique et métaphysique chez Aristote, Paris 2012. Botter, B.: Dio e Divino in Aristotele, Bonn 2005. Bradshaw, D.: A new look at the First Mover, in: Journal of the History of Philosophy, 39 2001, 1–22. Brancacci, A. (ed.): La filosofia in età imperiale. Le scuole e le tradizioni filosofiche, Napoles 2000. Broadie, S. 1993: Que Fait le Premier Moteur d’Aristote? Sur la théologie du livre Lambda de la « Métaphysique », in: Philosophique de la France et de l’Étranger, 183, 375–411. Elders, L. (ed.) 1972: Aristotle’s Theology, A Commentary on Book L of Metaphysics, Assen. Fazzo, S.: Sur la composition du traité dit motu animalium: contribution à l’analyse de la théorie aritotélicienne du premier moteur, in: A. Laks/M. Rashed (eds.), Aristote et le mouvement des animaux. Dix études sur le De motu animalium, Paris 2004, 203–229. Fazzo, S.: Il libro Lambda della Metafisica di Aristotele, Napoles 2012. Fazzo, S.: Commento al libro Lambda della Metafisica di Aristotele, Napoles 2014. Frede, M./Charles, D. (eds.): Aristotle’s Metaphysics Lambda. Symposium Aristotelicum, Oxford 2000. García Marqués, A./García Huidobro, J. (eds.): Razón y praxis, Valparaíso 1994.

226

Alberto Ross

Gill, M. L./Lennox, J. G. (eds.): Self Motion: From Aristotle to Newton, Princeton 1994. Gómez-Lobo, A.: Aristóteles y el aristotelismo antiguo, in: A. García Marqués/J. García Huidobro (eds.), Razón y praxis, Valparaíso 1994, 51– 68. Gourinat, J. B.: Le premier moteur selon Physique VIII et Métaphysique Λ: physique et philosophie premie`re, in: M. Bonelli (ed.), Physique et métaphysique chez Aristote, Paris 2012, 175–206. Gracia, J. (ed.): Concepciones de la metafísica, Madrid 1998. Hadot, I.: Simplicius, sa vie, son œuvre, sa survie, Berlin–New York 1987. Joachim, H. H.: Aristotelis De Generatione et Corruptione, Oxford 1922. Judson, L.: Heavenly Motion and the Unmoved Mover, in: M. L. Gill/J. G. Lennox (eds.), Self Motion: From Aristotle to Newton, Princeton 1994, 155–171. Kobusch, T./Erler, M. (eds.): Metaphysik und Religion, München–Leipzig 2002. Kosman, A.: Aristotle’s Prime Mover, in: M. L. Gill/J. G. Lennox (eds.), Self Motion: From Aristotle to Newton, Princeton 1994, 135–153. Laks, A.: Metaphysics L 7, in: M. Frede/D. Charles (eds.), Aristotle’s Metaphysics Lambda. Symposium Aristotelicum, Oxford 2000, 207–243. Laks, A.: Cátedra Tópicos: Los motores inmóviles de Aristóteles: una introducción sencilla a un problema complejo, México 2013. Laks, A./Rashed, M. (eds.): Aristote et le mouvement des animaux, Dix etudes sur le De motu animalium, Paris 2004. Menn, S.: Aristotle and Plato on God as Nous and as the Good, in: Review of Metaphysics, 45 1992, 543–573. Natali, C.: Causa motrice e causa finale nel libro Lambda Della Metafisica de Aristotele, in: Méthexis 1997, 10, 105–123. Philoponus: In Aristotelis Physicorum Libros Commentaria (2 vols.). Berlin, 1887–1888. Plato: Opera. Ed. J. Burnet. Oxford, 1900–1906. Rashed, M.: Alexandre d’Aphrodise. Commentaire perdu à la Physique d’Aristote (Livres IV–VIII), Berlin–Boston 2011 Reale, G. (ed.): Aristotele. La metafisica, Napoles 1968. Ross, A.: Dios, eternidad y movimiento en Aristóteles, Pamplona 2007a. Ross, A.: La causalidad del Primer Motor en Met. XII, in: Diánoia, 59, 3– 262007b. Ross, W. D. (ed.): Aristotle’s Metaphysics. A Revised Text with Introduction and Commentary, Oxford 1924. Ross, W. D. (ed.): Aristotle’s Physics, Oxford 1936. Ross, W. D. (ed.): Aristotle’s De Anima, Oxford 1961. Ross, W. D./Forbes, F. H. (eds.): Theophrastus. Metaphysics, Oxford 1929.

The Causality of the Prime Mover in Metaphysics Λ

227

Salis, R.: La causalidad del motor inmóvil según Pseudo Alejandro, in: Estudios de Filosofia, 40, 199–221 2009. Sedley, D.: Metaphysics L 10, in: M. Frede/D. Charles (eds.), Aristotle’s Metaphysics Lambda. Symposium Aristotelicum, Oxford 2000, 327– 350. Sharples, R.: Alexander of Aphrodisias and the End of Aristotelian Theology, in: T. Kobusch/M. Erler (eds.), Metaphysik und Religion, München– Leipzig 2002. Simplicius: In Aristotelis Physicorum Libros Commentaria (2 vols.), Berlin: Reimer, 1882. Simplicius: On Aristotle’s Physics 8.1–5, tr. I. Bodnàr/M. Chase/M. Share. Ithaca, 2012. Simplicius: On Aristotle’s Physics 8.6–10, tr. R. McKirahan. Ithaca, 2001. Stevens, A.: La causalité de l’intellect dans la Métaphysique et le traité De l’âme, in: La causalité chez Aristote, ed. L. Couloubaritsis/S. Delcomminette, Paris–Bruxelles–Vrin–Ousia 2011, 125–137. Themistius: In Aristotelis Metaphyicorum librum XII paraphrasis hebraice et latine. Berlin, 1903. Thomae Aquinatis: In duodecim libros Metaphysicorum Aristotelis expositio. Ed. R. M. Spiazzi, Turin–Rome, 1964. Vigo, A.: Naturaleza y finalidad, in:, A. García Marqués/J. García Huidobro (eds.) 1994: Razón y praxis, Valparaíso 1994.

Aristotle’s Silence about the Prime Mover’s Noe¯sis

1

MARIA LIATSI

In chapter nine of Metaphysics Lambda Aristotle asks what the object of the Prime Mover’s thought might be. In chapter seven he makes the general statement that the nous thinks itself by participation in the thinkability of its object (1072b19 f.: αὐτὸν δὲ νοεῖ ὁ νοῦς κατὰ μετάληψιν τοῦ νοητοῦ); this statement is general insofar as it includes both the human nous and the nous of the Prime Mover. So the next, unavoidable question is what the Prime Mover is thinking about. The answer to this question has remained controversial since the time Theophrastus2, and a consensus has never been reached. Among the approaches to this problem it is possible to distinguish between a formal and a material type. The formal type interprets the self-thinking of the Prime Mover as the self-reference of his thought, but does not really answer the question about the nature of the self-reference of the Prime Mover’s thought. On the other hand, the material way of answering the question interprets the self-knowledge of the Prime Mover as one which, in thinking itself, also includes the ideas or forms of being, i.e. the main structures of being, so that his self-thought also thinks the world. These are the two ways of interpreting the Prime Mover’s thought, and both of them refer to the text of book Lambda, which resembles the Sphinx. 1

2

This article results, last but not least, from the discussion during our meeting at the University of Bonn in December 2010. An earlier version of the paper was delivered before this assembly. I am indebted to all members of the conference in Bonn for helpful remarks about the problems which are connected with the subject treated here. I am conscious of having benefited particularly from discussions with Jonathan Beere, Enrico Berti, Klaus Corcilius, Christoph Horn, Oliver Primavesi, Carlos Steel, Marco Zingano. Needless to say, any imperfections are my own. Laks, A./Most, G. W. (eds.) 1993: Théophraste, Métaphysique, Paris, 5a14–6a15. Cf. Ross, W. D./Forbes, F. H. (eds.) 1929: Theophrastus. Metaphysics, Oxford. See Frede, D. 1971: Theophrasts Kritik am unbewegten Beweger des Aristoteles, in: Phronesis, 16, 65–69. Cf. Alexander and Pseudo-Alexander 1891, in: M. Hayduck (ed.), Commentaria in Aristotelem Graeca, vol. I, Berlin, Praefatio V–X, Freudenthal, J./Fränkel, S. 1885: Die durch Averroes erhaltenen Fragmente Alexanders zur Metaphysik des Aristoteles: Berlin, 69 ff. See Rosemann, P. W. 1986: Das Aristotelische Problem der Selbstbezüglichkeit des Unbewegten Bewegers in der Kommentierung Ibn Rushds, in: Zeitschrift für philosophische Forschung, 40, 543–560.

230

Maria Liatsi

Aristotle’s own method of answering the question seems to be plausible: he simply follows a yes-or-no-scheme. His axiom is that the nous of the Prime Mover is the most divine thing of all (1074b15 f.: δοκεῖ μὲν εἶναι τῶν φαινομένων θειότατον). Then he formulates a series of pairs of alternatives, each time eliminating one alternative and using the other as an introduction to the next pair, eventually arriving at his final answer. The result of his procedure is the famous conclusion that the object of the nous of the Prime Mover is not just anything but is something equal to its own ontological status (1074b26: τὸ θειότατον καὶ τιμιώτατον νοεῖ, καὶ οὐ μεταβάλλει). Hence, it can only be itself (1074b33 f.: αὐτὸν ἄρα νοεῖ, εἴπερ ἐστι τὸ κράτιστον). Therefore, the nous of the Prime Mover is one with its object, and its thought is a thinking of thinking (1074b34: καὶ ἐστιν ἡ νόησις νοήσεως νόησις). In contrast to the human thought about thinking, the thought of the Prime Mover is not just reflexive “on the side” (ἐν παρέργῳ, 1074b36), but has no object other than itself. This result raises the next question as to how it is possible that knowledge could have itself as its object. Aristotle’s answer (1075a3 ff.) is that in the case of things that have no matter, thought 3 and the object of thought are not different (οὐχ ἑτέρου οὖν ὄντος τοῦ νοουμένου καὶ τοῦ νοῦ, ὅσα μὴ ὕλην ἔχει), and therefore the divine thought and its object will be the same: the thinking will be one with the object of its thought (ἡ νόησις τῷ νοουμένῳ μία). There is not a single word about the content of the Prime Mover’s selfknowledge. In the following passage Aristotle discusses the formal structure of a possible object of the divine thought, specifically whether it is composite or not (1075a5 f.: εἰ σύνθετον τὸ νοούμενον), and the answer is no, it is not; otherwise the divine thought would change from one part of the whole to another (1075a6: μεταβάλλοι γὰρ ἂν ἐν τοῖς μἑρεσι τοῦ ὅλου), which is not compatible with the Prime Mover’s autonomy. This passing from one part to the other would presuppose immaterial forms and the series of numbers, which are not objects of the Prime Mover’s thought, because forms and numbers are ordered successively, passing from part to part, and this process of thinking is only possible in a certain period of time, which is characteristic only of the human mind which is concerned with compound things (1075a7 f.), but it is not characteristic of the divine mind which has only itself as its object, and has it in toto, as a whole (1075a10), and is not dependent on the mediation of linguistic symbols like words and sentences: to the contrary, it consists of a single act of immediate insight.

3

Burnyeat suggests to translate νοῦς as ‘intellect’ and the verb νοεῖν as ‘understand’. See Burnyeat, M. F. 2008: Aristotle’s Divine Intellect, Milwaukee, 9–63, esp. 10–19.

Aristotle’s Silence about the Prime Mover’s Noe¯sis

231

As far as we can see, the text,4 of the ninth chapter of Lambda is a logical movement from alternative to alternative that finally brings Aristotle into a position in which he can make no other statement than this: the Prime Mover is autonomous and thinks only himself. On the basis of the text of Lambda, there seems to be no other conclusion: it excludes from the Prime Mover any knowledge of the world. This interpretation of the text was adopted by many scholars in modern times, among them Eduard Zeller,5 Albert Schwegler,6 David Ross,7 Francis Cornford,8 William Guthrie,9 and others. This was, in any case, the traditional interpretation of Lambda in the nineteenth and twentieth century. Nevertheless, there was from time to time a feeling of uneasiness among scholars about whether Aristotle really had no more to say concerning the Prime Mover and his relation to the world. Many attempts have been made to deny that the text of Lambda is the complete doctrine of Aristotle. Conversely, attempts have been made to interpret the text of Lambda in a manner which does not share the assumptions of Aristotle. Such is the case with, for instance, the attempt to modify the Prime Mover from a causa finalis,10 into a causa efficiens11 which is incompatible with Aristotle’s well-known doctrine of the eternity of the world.12 Apart from this, when someone says that the Prime Mover is a causa efficiens, what does

4 5 6 7 8 9 10

11

12

Cf. Sandbach, F. H. 1954: A Transposition in Aristotle’s Metaphysics L 9, 1074 b, in: Mnemosyne, 8, 39–43. Zeller, E. 41921: Die Philosophie der Griechen in ihrer geschichtlichen Entwicklung, Leipzig, 382. Schwegler, A. (ed.) 1847: Die Metaphysik des Aristoteles. Grundtext, Übersetzung und Kommentar nebst erläuternden Abhandlungen, Tübingen. Ross, W. D. 1924: Aristotle’s Metaphysics. A Revised Text with Introduction and Commentary, Oxford. Cornford, F. M. 1932: Before and After Socrates, Cambridge. Guthrie, W. K. C. 1950: The Greeks and their Gods, London. On the view that the Prime Mover is a causa finalis see for example Genequand, C. (ed.) 1986: Ibn Rushd’s Metaphysics. A translation with Introduction of Ibn Rushd’s commentary on Aristotle’s Metaphysics, Book Lambda, Leiden, 41–42. Schüssler, W. 2001: Als protestantischer Theologe in philosophischem Material, in: I. Nord/Y. Spiegel (eds.): Spurensuche. Lebens- und Denkwege Paul Tillichs, 139–151, esp. 151. Cf. Laks, A. 2000: Metaphysics L 7, in: M. Frede/D. Charles (eds.), Aristotle’s Metaphysics Lambda. Symposium Aristotelicum, Oxford, 207–243, esp. 243. For example Berti, E. 2000b: Unmoved mover(s) as efficient cause(s) in Metaphysics L 6, in: M. Frede/D. Charles (eds.), Aristotle’s Metaphysics Lambda. Symposium Aristotelicum, Oxford, 181–206. Cf. Politis, V. 2000: Aristotle and the Metaphysics, London–New York, 278. Wieland, W. 1960: Die Ewigkeit der Welt (Der Streit zwischen Joannes Philoponus und Simplicius), in: D. Henrich (ed.), Die Gegenwart der Griechen im neueren Denken. Festschrift H.-G. Gadamer, Tübingen, 291–316.

232

Maria Liatsi

this mean for the whole ontological system of Aristotle’s universe? What are the implications of such a position and where does it lead us? What is the next step? Another false inference is that the Prime Mover knows all things, because it is the first principle of all, so that its self-knowledge must be knowledge of the world and the essences of all beings in consequence of its being the first principle. But the fact is that Aristotle gives many reasons for the Prime Mover’s immutability but says not a single word about the Prime Mover’s omniscience.13 All these and other speculations are without any basis in the text of Lambda, although they became widespread doctrines in later ages. Perhaps the reason for Aristotle’s silence was his lack of interest in the many theological implications connected with all this. On the other hand, there is no doubt that the feeling of uneasiness with regard to the text of Lambda is comprehensible. It is, in spite of all that has been said, difficult to understand why Aristotle is so reticent about the Prime Mover’s knowledge. Therefore, annotators have suspected that the Lambda is only an abridged, fragmentary form of Aristotle’s complete theology, which we do not have. But we do not know. Nevertheless, scholars go far beyond what is delivered by the text of Lambda. In this situation one has to make a decision, namely, one has to decide what one is going to interpret. Do we interpret the text of Lambda as it lies before us on the table, the text that exists? Or do we interpret it based on our subjectivity which may be full of prejudices, yielding a text which does not really exist, but is only a conclusion drawn from our own inferences? As we all know, the borderline between these two attitudes is very often not clear, more vague than distinct, and so is the result of our reasoning about the text. I will mention two examples. The first example concerns the age-old attack on what is called “a Narcissus-like theology”, meaning by this the doctrine of Lambda chapter 9, in combination with the belief that this cannot be the real opinion of Aristotle, in any case not the opinion of the later Aristotle after he had become a man of wisdom, and so chapter 9 of Lambda is said to be “a provisional draft, later on supplanted by Lambda 7”, including “God’s omniscience”, and so intended to resolve the big quarrel about noesis noeseos”. This opinion was uttered not long ago, in the year 2000, by Jacques Brunschwig.14

13 14

Cf. Kretzmann, N. 1966: Omniscience and Immutability, in: Journal of Philosophy, 63, 409–421. Brunschwig, J. 2000: Metaphysics Λ 9: A Short-Lived Thought-Experiment? In: M. Frede/ D. Charles (eds.), Aristotle’s Metaphysics Lambda. Symposium Aristotelicum, Oxford, 275–306.

Aristotle’s Silence about the Prime Mover’s Noe¯sis

233

There is no truth in it. The mythological figure of Narcissus says mighty little, practically nothing, about the meaning of the figure of self-reference in logic and philosophy. The figure of self-reference is concerned with the highest point within the system of Aristotle, and the figure of Narcissus has no relevance whatsoever to the question of whether God’s knowledge is pure self-reference or not. The comparison with Narcissus is blind to the dialectical aspect of the figure of self-reference. Brunschwig’s comparison is a category mistake; it is not a contribution toward solving the problem but a problem in itself. The second example I want to cite is a single point in the commentary on Lambda written by Michael Bordt,15 whose main thesis regarding chapter 9 is to totally deny the self-referential character of the Prime Mover’s thought. He denies this in spite of the generally accepted fact that there is absolutely no thought, human or divine, which is completely without any moment of self-reference, because this is impossible for logical reasons. Bordt starts his interpretation of Lambda 9 with the general statement that νόησις νοήσεως is not self-referential for two reasons.16 (1) Bordt holds that already in chapter 7 Aristotle had mentioned the problem of νοῦς without touching on the topic of self-reference (ohne “die Vorstellung einer Selbstreflexion”). This is a false statement, however, because in chapter 7 Aristotle argues that the First Mover is the highest being and that his activity is thinking, but not thinking of just anything, but only of the best, 1072b18–9: ἡ δὲ νόησις ἡ καθ᾽ αὑτὴν τοῦ καθ᾽ αὑτὸ ἀρίστου, καὶ ἡ μάλιστα τοῦ μάλιστα. But what the best is, Aristotle has already said: the being of the First Mover is the best, therefore the First Mover’s thought thinks itself. This conclusion has the character of an enthymeme, because this conclusion, after all, is the only possible one. In the next passage, 1072b19–24, Aristotle is not speaking of the Prime Mover’s thought, as Bordt maintains, but about human thought and its kind of reflexivity. This is evident from the distinction between potentiality and actuality of thought. In 1072b22–3, Aristotle differentiates between the νοῦς as τὸ δεκτικὸν τοῦ νοητοῦ καὶ τῆς οὐσίας and the νοῦς ἔχων, i.e. between the νοῦς which is only able to receive its object, which means the νοῦς as potentiality, and the νοῦς in its active form as νόησις, which really thinks and is nothing other than thinking and has its object in its thinking. Only this last one is concerned with the divine thought itself, as Aristotle puts it: ὥστ᾽ ἐκείνου μᾶλλον τοῦτο ὃ δοκεῖ ὁ νοῦς θεῖον ἔχειν (1072b23). And only in this

15 16

Bordt, M. 2006: Aristoteles’ Metaphysik XII, Darmstadt. Bordt (see note 15), 147.

234

Maria Liatsi

active form is it the case that the θεωρία is τὸ ἥδιστον καὶ ἄριστον (1072b24). And this is the status in which God is forever: ὁ θεὸς ἀεί, but for us human beings it is only experienced sometimes (ὡς ἡμεῖς ποτέ, 1072b25). Not only Bordt in his commentary, but many other interpreters did not realize here the distinction between the divine and the human νοῦς. Moreover, the human mind is always dependent on other things, because only other things impel the human νοῦς to think and to transform its potentiality into actuality: νοῦς δὲ ὑπὸ τοῦ νοητοῦ κινεῖται (1072a30). This last quoted sentence is the most important axiom of Aristotle’s theory of knowledge. (2) Bordt’s next argument is as follows. He believes that Aristotle refers to the objects of the theoretical science (e.g. first philosophy), i.e. the eternal theoretical truths concerning the world around us, and that these truths would be the objects of the Prime Mover’s thought. Thus the Prime Mover’s thought would not be self-referential. But this interpretation does not fit Aristotle’s assumption that objects of theoretical science are connected with ὕλη, and that they become objects of theoretical science only “by abstraction” (ἐξ ἀφαιρέσεως, Metaphysics Κ 3, 1061a29. Cf. 4, 1061b17 ff. Α 2, 982a223 ff. Ζ 3, 1029a11–33). In Aristotle’s view only two objects are pure form, pure εἶδος without ὕλη, and not only “by abstraction”. These εἴδη are the Prime Mover and the fifty-five Movers of the celestial spheres. Therefore, Michael Bordt’s denial of the self-referential character of νόησις νοήσεως is inconsistent with Aristotle’s assumptions and with the logic of his system. These and other attempts to deny the Prime Mover’s self-reference have their origin in the already mentioned belief, which is not confirmed by the text, that the Prime Mover’s thought has more objects than just itself. It is a historical fact that the material approach to interpreting Metaphysics Lambda was favoured by Platonism in the period after Aristotle and in Late Antiquity, and moreover by Christian theology in form of the so-called Interpretatio Christiana. The doctrine of the Ideas as God’s thoughts deeply influenced the interpretation of νόησις νοήσεως throughout the Middle Ages and even in modern times. In spite of this development, there is reason to doubt whether Lambda is a theology in the traditional sense at all. Michael Frede, for instance, doubts that Lambda is a theology and states that Lambda’s chapters 2–5 are not “a preliminary discussion and that the real subject of Lambda is not God or the divine substances”, and that, as Frede puts it, on the contrary, the sensible substances in Lambda “form an integral part of theology itself as a universal science”.17 Many interpreters in the past already

17

Frede, M. 2000a: Introduction, in: M. Frede/D. Charles (eds.), Aristotle’s Metaphysics Lambda. Symposium Aristotelicum, Oxford, 9.

Aristotle’s Silence about the Prime Mover’s Noe¯sis

235

had the same impression and concluded that Aristotle was gradually forced, by the logical pressure of his Physics and by the movement from alternative to alternative, to the final conclusion that the First Unmoved Mover, God, is pure energy (ἐνέργεια) and thinks only himself – a result which is forced upon him by his own premises. The question is, what did Aristotle do in this situation, when he was driven to this conclusion by the presuppositions of his own system? I think there is a consensus that Aristotle was not a wholehearted theologian,18 and I, at least, can imagine that he had certain difficulties in saying something about God’s behaviour. Fortunately, in this case we are in a position to know what he thought about it. In De partibus animalium I 5, 644b22–645a6, he compares the inquiry into the ungenerated and eternal beings with the inquiry into generated and perishable ones. He says: “The former are excellent beyond compare and divine, but less accessible to knowledge (τιμίας οὔσας καὶ θείας ἐλάττους ἡμῖν ὑπάρχειν θεωρίας). Both departments, however, have their special charm. The scanty conceptions to which we can attain celestial things give us, from their excellence, more pleasure than all our knowledge of the world in which we live, just as a half glimpse of persons that we love is more delightful than a leisurely view of other things, whatever their number and dimensions. On the other hand, in certitude and in completeness our knowledge of terrestrial things has the advantage. Moreover, their greater nearness and affinity to us balances somewhat the loftier interest of the heavenly things that are the objects of the highest philosophy (ἀντικαταλλάττεταί τι πρὸς την` περὶ τὰ θεῖα φιλοσοφίαν). Having already treated of the celestial world, as far as our conjectures could reach (διήλθομεν τὸ φαινόμενον ἡμῖν), we proceed to treat of animals”.19 This text gives an indication of why Aristotle is so reticent to say anything about the heavenly things, which are the objects of our loftier interest. In this situation, the best he could do was to define the divine mind as a sort of self-reference, which is inscrutable. Because no human being is able to understand the divine mind from a human perspective, i.e. from without, Aristotle did feel the inadequacy of this relationship. The rest was silence. This, of course, is only one possible explanation of Aristotle’s silence about any further investigation concerning the phrase νόησις νοήσεως. Nevertheless, there are doubts. Perhaps the text of Lambda really is no more than 18

19

Cf. Horn, C. 2002: In welchem Sinn enthält Metaphysik Lambda eine Theologie? In: Jahrbuch für Religionsphilosophie, 1, 28–49. See also De Koninck, T. 1999: Aristotle on God as Thought Thinking Itself, in: L. P. Gerson (ed.), Aristotle. Critical Assessments I: Logic and Metaphysics, London, 365–402. Translation by A. L. Peck, Aristotle. Parts of Animals. The Loeb Classical Library 323, Cambridge, Mass.–London 1937, ad loc.

236

Maria Liatsi

a short draft, used by Aristotle in his lectures, open to further detailed oral supplements, as Werner Jaeger presumed. Perhaps, then, while teaching, Aristotle made statements about God’s knowledge of the world, similar to the doctrine of his fellow-Platonists in the Academy, in a certain analogy with Speusippus and Xenocrates.20 It is not impossible, but improbable. Otherwise, Theophrastus would have said something about it. We do not know. What we know is that Aristotle was much more interested in the experience of the immanent world, of nature, than in speculations on the beyond, the hereafter, and the life to come or not to come. Within the historical framework of Platonism, his discretion with regard to systems of ontological derivation or emanation is highly remarkable, anyway. As far as I can see, there are four possible explanations for Aristotle’s silence about the Prime Mover’s νόησις νοήσεως: 1. The text of Lambda is a short draft for lectures. 2. The text of Lambda is incomplete, only a fragment. 3. The text of Lambda is complete, but does not contain Aristotle’s complete thoughts about the topic. Reservatio mentalis. 4. The text of Lambda is complete and contains what Aristotle had in mind about the topic, but we have not understood it. Let us analyse the fourth possibility. For Aristotle the movement, κίνησις, is eternal.21 The eternal movement must have a τέλος. From this τέλος the eter20

21

There is only one passage which could belong to such a framework. This is Fragment 49, a sentence of his dialogue Περὶ εὐχῆς which runs: “God is νοῦς or something beyond νοῦς (ὅτι ὁ θεὸς ἢ νοῦς ἔστιν ἢ καὶ ἐπέκεινα τὶ τοῦ νοῦ), a fragment from Simplikios’ Commentary on Aristotle’s De caelo II 12. 292b10 (CAG VII p. 485, ed. Heiberg, Berlin 1894). It seems that Aristotle here aims to distinguish two conceptions of God: (1) God is νοῦς, which would be in accordance with Lambda, and (2) God is even beyond νοῦς, which could be an echo of Politeia VI 509 b, where the phrase ἐπέκεινα τῆς οὐσίας is alluding to a last principle beyond νοῦς. H. Flashar in his recently published commentary to Aristotle’s Fragmente I (2006, p. 167) denies the possibility of taking this fragment as evidence for a Platonic position espoused by Aristotle in his dialogue Περὶ εὐχῆς. On the contrary, he thinks that Aristotle, at the end of his dialogue, was discussing the question whether it makes sense to address prayers to God who is so far away from human beings, and how we should estimate the moral quality of prayers in general. Another possibility is that Aristotle refers here to questions which had been discussed in the academy, particularly between Plato, Speusippus and Xenocrates. In any case, there can be no doubt about Aristotle’s own position, for he criticizes the doctrine of Speusippus emphatically at the end of Lambda. Through the metaphor of the army and its leader and the quotation from the Iliad Aristotle confirms his own position: God is νοῦς, he is the Unmoved Mover, and, hence, he is the final cause of the world. This seems to be the message of his first philosophy in Lambda. Cf. Kosman, A. 1994: Aristotle’s Prime Mover, in: M. L. Gill/J. G. Lennox (eds.), Self Motion: From Aristotle to Newton, Princeton, 135–154, esp. 149–150.

Aristotle’s Silence about the Prime Mover’s Noe¯sis

237

nal movement is at a permanent distance, which is always the same. This τέλος is the First Mover, the πρῶτον κινοῦν, without any movement; it is unmoved. It is the τέλος of everything that is in a state of movement. Therefore, it is the highest kind of being. The question is, whether such a being does exist. Aristotle says “yes”, it is the pure energy, the pure ἐνέργεια. For Aristotle the manifestation of this pure ἐνέργεια is ἡ θεωρία (1072b24 ff.), the knowledge. The knowledge is something τέλειον, that is, complete, selfsufficient and for the sake of itself; ἥδιστον, the most delightful; and ἄριστον, the very best. The θεωρεῖν, the νοεῖν is always active, because the essence of νόησις as νόησις is activity which never rests. For this reason the kind of being of the First or Prime Mover can only be the being of νοῦς in its active form, and that is νόησις. And the object of it can only be itself, as the ἄριστον, which it is. Therefore, the highest being is νόησις νοήσεως (1074b34 f.), the pure knowledge of itself.22 After all, we can say that it is a mistake to believe that Aristotle thinks of a personal mind, of a person, of the personality of God. He is only interest-

22

This highest being seems to us (human beings) even in its form and not only in its possible content so hopelessly circular (νόησις νοήσεως νόησις) and so puzzling to apprehend. Perhaps we could speak, hypothetically, about the first principle that in its perpetual reflexive activity includes the basic eternal, modern speaking, natural laws of the universe, i.e. the rotation of the celestial bodies, the ceaseless interchange of the four material elements (earth, air, fire and water; GC II 10. 336b25 – 337a15; Met. Θ 8. 1050b28–30), or the constant reproductive cycles of living things, plants and animals (De an. II 4. 415a26–b7; GA II 1. 731b24 – 732a11). This does not mean that this principle is knowledge of the world, but only that in this way of understanding it, it is an ἀρχή τῆς κινήσεως, because it provokes the ceaseless movement of the heaven and of the natural world, which means that it provokes life, moreover it is life (1072b27: ἡ γὰρ νοῦ ἐνέργεια ζωή), as it provokes the continuous attempt of the universe to reach the final aim (τέλος) of the κίνησις, which is the first principle itself. That simply means, it moves everything as the final cause. The universe depends necessarily on such a principle (1072b13 f.: ἐκ τοιαύτης ἄρα ἀρχῆς ἤρτηται ὁ οὐρανὸς καὶ ἡ φύσις), which could explain why „heaven“ and „nature“ actually have to move in a circular way: in order to achieve constancy, one might answer. And why must constancy be achieved? Because the phenomena, or in other words empirical experience, show for Aristotle that there is this constancy, as well as a τάξις in the universe, and this must be explained. The ἀρχή that Aristotle calls πρῶτον κινοῦν ἀκίνητον or νόησις νοήσεως consists, metaphorically speaking, in a condensed knowledge and in an intensive certainty of the „why“ (ἕνεκα οὗ) of the constancy, of the regularity and of the certain order of the κίνησις, not in knowledge of the world. In an analogical way, in human life the activity of νοῦς, i.e. the θεωρία, is the highest form of knowledge, the best and most delightful ἐνέργεια, leading to the most complete happiness (εὐδαιμονία). Cf. Kahn, C. H. 1985: The Place of the Prime Mover in Aristotle’s teleology, in: A. Gotthelf (ed.), Aristotle on Nature and Living Things, Bristol–Pittsburgh, 183–205, esp. 184: „the eternal guarantee and ultimate driving (drawing) force for all patterns of teleology in nature as well as in human life and action“. Cf. also Guthrie, W. K. C. 1933: The Development of Aristotle’s Theology I, in: CQ, 27, 162–171, esp. 167. Politis (see note 11), 288–290.

238

Maria Liatsi

ed in finding a being of the highest rank, of the highest position, of the highest order. What he is absolutely not seeking is a self-thinking mind in a personal sense, a kind of René Descartes as God. And this exactly is the reason, why we have not found and shall not find in the context of Lambda a statement about the togetherness of the highest being and the world, and not a single word about the cosmic creation, the origin of the universe by that highest being, or about the maintenance of the world. A further relationship, other than the one that all things have to this removed τέλος, towards which they always tend and which they try to imitate, i.e. between this highest being and the world, is undefined and undetermined. No further explanation is necessary, because for Aristotle the world is eternal, without a beginning and without an end. This means that the whole connection of the moved world with the moving principle is purely ontological and has in Aristotle’s view nothing to do with a personal God or Divine Creator. What he really does, is to explain the phenomenon and the concept of movement (κίνησις) by means of ontology alone – and we cannot forget that Aristotle begins always from the phenomena, not only in his Metaphysics. Everything else is a picture, a metaphor, like the phrases ὡς ὀρεκτόν (1072a26) and ὡς ἐρώμενον (1072b3).23 He knows that there is no more to say for him, i.e. from a human perspective – hence the use of pictures and metaphors. His only interest and intention is to answer the question, is there something like a highest being, which could be the keystone of a doctrine of being: a first mover, a first heaven, a first substance or something like that; the purpose of this inquiry is only in order to come to an end, because the worst thing for Aristotle, as we know from many of his writings, is an open system, not closed, left incomplete by a regressus in infinitum. This means that the Unmoved Mover as the highest point of first philosophy is outside of any direct connection with the world and with human beings. It is a pure “meta-physical” construction with regard to his doctrine of movement, that is to his Physics, without any idea of creation or direction or providence through a divine principle. He calls it θεός (1072b25–30), but this is a name, the name for the highest being within Aristotle’s ontological and natural order 24 – far away from everything that 23

24

Cf. Elders, L. (ed.) 1972: Aristotle’s Theology, A Commentary on Book L of Metaphysics, Assen, 173–174. Genequand (ed.), Ibn Rushd’s Metaphysics, (s. note 10), 36–39. De Filippo 1994, 393–409, esp. 400–401. Laks (see note 10) 207–243, esp. 223–224. See H.-G. Gadamer: “Dieses Seiende ist nicht durch unsere Sinne, sondern allein denkend erfahrbar – also ganz wie in der platonischen Wendung zur Idee, aber es ist dennoch nicht eine allgemeine Idee, sondern hat ebenso konkretes Fürsichsein wie die Wesen der Natur, es ist das, was die religiöse Überlieferung „Gott“ nennt”. In: Gadamer, H.-G. 1948: Aristoteles’ Metaphysik XII, Frankfurt a.M., 5. Cf. Broadie, S. 1993: Que Fait le Premier Moteur d’Aristote? Sur la théologie du livre Lambda de la « Métaphysique », in: Philosophique de

Aristotle’s Silence about the Prime Mover’s Noe¯sis

239

was later on said by others to be the theological meaning of νόησις νοήσεως. From this point of view the history of interpretation of the expression νόησις νοήσεως is a history of misinterpretation, a Historia calamitatum. So far, Aristotle’s so-called silence is not really a silence, but, strictly speaking, our own failure to understand what Aristotle is saying in Lambda about the highest point of his doctrine of κίνησις.25 If we repeat, once more, the comparison between the formal and material types of explanation with regard to the Prime Mover’s knowledge, we are inclined to say that after the entry of christian theology in late antiquity, there is no doubt that the material interpretation played the dominant role and did so up to modern times, even up to the present. But, on the other hand, after this material type of interpretation was settled, book Lambda in its material understanding as an instrument of theology had a stabilizing influence on christian dogmatism, and christian philosophers and theologians made use of it. Lambda, therefore, had an influence on christian theology not only in the East, but also in the West. However, in the nineteenth century scholars began to reformulate the problem in a way that excluded knowledge of the world from the Prime Mover, among them Eduard Zeller, who was followed in the twentieth century by others. For these scholars the activity of the Prime Mover was to think itself, in accordance with the wording of the Aristotelian text. But they did not explain what it means that a thought is thinking itself. In other words, they did not realise that they are confronted with the logical problem of self-reference, because Aristotle’s assertion here is the assertion of the reflexivity of the thought of the Prime Mover. Therefore, it is necessary to ask how Aristotle understands reflexivity. Reflexivity is a relation, but a special one. In the Categories, ch. 7, Aristotle is concerned with reflexive relations; he knew of this problem from Plato’s treatment of this subject in Charmides and Sophistes. In the Charmides Plato doubts the possibility of a knowledge, which is its own object. But in the Sophistes, Plato makes some statements which could lead to the solution of the problem in the Charmides, particularly Plato’s analysis of the notion of identity (ταὐτόν) in connection with the system of the μέγιστα γένη. It was

25

la France et de l’Étranger, 183, 375–411. Laks (see note 10). Sedley, D. 2000: Metaphysics L 10, in: M. Frede/D. Charles (eds.), Aristotle’s Metaphysics Lambda. Symposium Aristotelicum, Oxford, 327–350. Horn, C. 2008: Aristoteles in Metaphysik Lambda über das Verhältnis von Mythologie und Rationalität, in: J. Halfwassen/M. Gabriel (eds.), Kunst, Metaphysik und Mythologie, Heidelberg, 19–31. Cf. Manuwald, B. 1989: Studien zum Unbewegten Beweger in der Naturphilosophie des Aristoteles, Stuttgart.

240

Maria Liatsi

the mathematician Erhard Scheibe,26 who in 1967 drew some conclusions from Plato’s text, which are confirmed by systematic considerations and lead to the result that reflexive relations presuppose identity between the relata of these relations, what means that reflexive relations have their basis in the reflexivity of the identity-relation, which has its basis in identity. Participation in the genos of identity is the only ground for the possibility of reflexive relations. Since E. Scheibe’s observations, we have good reasons for the assumption that, concerning the problem of identity and reflexivity, Aristotle followed Plato’s lead. It is far more than an assumption, because it is confirmed by the texts of Aristotle, in which we find the beginnings of a logic of relations, of which he was the founder in the history of logic. In Metaphysics Δ, ch. 9, he defines the same as a relationship: “Clearly, therefore, sameness is a unity of the being either of more than one thing or of one thing when it is treated as more than one, i.e., when we say a thing is the same as itself, for we treat it as two” (1018a7 f.). Most important in this statement is not only that identity functions as the basis of reflexive relations, but that identity itself, which, by its nature, is nothing else than the unity of a plurality, is defined as a relation. Consequently, Aristotle in Metaphysics Δ, ch. 15, where he is enlisting the kinds of relations, does not forget identity when defining sameness as a relation. It seems more than plausible to apply Aristotle’s own definition of sameness to the interpretation of his statements about the self-referential character of knowledge.27 The main mistake of many interpreters was their belief that

26 27

Scheibe, E. 1967: Über Relativbegriffe in der Philosophie Platons, in: Phronesis, 12, 28–49. This is the method of K. Oehler (Oehler, K. 1976: Aristotle on Self-knowledge, in: Proceedings of the American Philosophical Society, 118, 463–506, Oehler, K. 1984: Der Unbewegte Beweger des Aristoteles, Frankfurt a.M. [German Translation]) in his article “Aristotle on Self-knowledge”. Among the many proposals which commentators have advanced so far in order to solve the problem of self-reference in Aristotle’s doctrine of the Unmoved Mover, I see no better way than Oehler’s method of applying Aristotle’s own Logic of Relations to Aristotle’s assertions about the self-referential character of knowledge. With regard to this subject, my analysis in this paper follows therefore Oehler’s method, which he first applied in 1974 in his article on Aristotle’s Self-Knowledge. Cf. also Oehler, K. 1981: Logic of Relations and Inference from Signs in Aristotle, in: Ars Semeiotica, 4, 237–246; German translation: Oehler, K. 1982: Die Anfänge der Relationenlogik und der Zeichenschluß bei Aristoteles, in: Zeitschrift für Semiotik, 4, 259–266. The linguistic background of Oehler’s interpretation of νόησις νοήσεως is his investigation of the semantic field of the notions νοῦς, νόησις, νόημα, νοεῖν, νοεῖσθαι, νοητόν, διάνοια, διανόησις, διανοητικόν, διανοητόν, διανόημα, διανοεῖσθαι, etc., which is published in his work: Die Lehre vom Noetischen und Dianoetischen Denken bei Platon und Aristoteles (Oehler, K. 21985: Die Lehre vom Noetischen und Dianoetischen Denken bei Platon und Aristoteles. Ein Beitrag zur Erforschung der Geschichte des Bewusstseinsproblems in der Antike, Hamburg).

Aristotle’s Silence about the Prime Mover’s Noe¯sis

241

what Aristotle, in describing the Prime Mover’s activity, called νόησις νοήσεως, the thinking of thinking, would be an identity of thinking and the thought in the sense of a misunderstood unio mystica, i.e. without a distinction between the νόησις and the νοητόν. The fact is that the Prime Mover thinks himself, because his thinking is identical with his thinking his thought, i.e. himself. His thinking is thinking of his thinking. Nevertheless, although the νόησις is also its own object, there is, λόγῳ, a distinction between the νόησις and its object. Aristotle was not a mystic. In De anima III 4, 430a3–4 Aristotle says: “In the case of objects which involve no matter, what thinks (τὸ νοοῦν) and what is thought (τὸ νοούμενον) are identical”. Here, too, the basic presupposition is the Aristotelian notion of identity, which means that the asserted identity of the thinking and the thought is a relation, not a mystic unity. The structured unity of actual thinking with its object consists of two aspects, which Aristotle always has in mind when he speaks about identity in this framework. These two aspects are thinking and being thought. And it is not by accident that he formulates this distinction also in connection with the question, whether the Prime Mover has its well-being through its thinking or through its being thought: ἔτι εἰ ἄλλο τὸ νοεῖν καὶ τὸ νοεῖσθαι, κατὰ πότερον αὐτῷ τὸ εὖ ὑπάρχει; οὐδὲ γὰρ ταὐτὸ τὸ εἶναι νοήσει καὶ νοουμένῳ (Metaphysics Λ 9, 1074b36–38). Commentators are disappointed that Aristotle does not answer this question, although he uses very often the figure of enthymeme, if the matter under discussion is apparent, as in this case. Whoever knows the Aristotelian doctrine of being knows the answer, without need of a statement by Aristotle. The answer depends on what Aristotle is saying in Met. Λ 9, 1074b35 f.: “knowledge and perception and opinion and understanding have always something else as their object, and themselves only by the way (ἐν παρέργῳ)”. The reflexivity of these intentional acts is self-referential, is nothing else than self-reference, but it is only possible in a roundabout way with regard to an object (ἐν παρέργῳ), and follows the general premise of the Aristotelian theory of knowledge that there is a primacy of things over knowledge, which implies the receptive character of perception and thought. The objects impel the mere faculty of perception and of thought to their active form, thereby changing the potential state of these faculties into actual performance. In this process the perceiving of something leads to the awareness that we perceive, the thinking of something leads to the consciousness that we think.28 So, the

28

Cf. Eudemian Ethics VII 12. 1245a7–10: αἰσθανόμενος μέν γὰρ αἰσθητὸς γίνεται ταύτῃ καὶ κατὰ τοῦτο, καθὰ πρότερον αἰσθάνεται, καὶ ᾗ καὶ οὗ, γνωστὸς δὲ γινώσκων· ὥστε διὰ τοῦτο καὶ ζῆν ἀεὶ βούλεται ὅτι βούλεται ἀεὶ γνωρίζειν, τοῦτο δὲ ὅτι αὐτὸς εἶναι τὸ γνωστὸν. Cf. also Nicomachean Ethics IX 9.

242

Maria Liatsi

human mind is primarily concerned with something else and only incidentally with its own activity. In contra-distinction to the human mind the divine mind knows itself not incidentally but as its only object. That is what Aristotle is saying by this comparison with the human mind. From this follows the conclusion that the divine mind, defined as pure energeia, is always in the state of self-reference, whereas the human mind does not remain, because it cannot remain, in this state according to its human conditions. The result of all this is that the knowledge of objects and of its own existence has to discern between itself and its objects. Therefore, concerning the Prime Mover’s thought, the first task is not to seek the content of this thought, but to find its structure, instead of maintaining a cryptic identity between the knower and what is known. But also for the Prime Mover there exists the primacy of the object of its thought and not the primacy of the thought. The dignity of the Prime Mover is the highest, not because he is the first ground of the cosmos, but because its thought participates in what it is thinking, only in this way becoming one and the same (Met. Λ 7, 1072b21: ὥστε ταὐτὸν νοῦς καὶ νοητόν).29 Although thinking, like any other cognitive activity, has its object also in itself, we are able to distinguish between the object and the cognitive activity. The mind (νοῦς) thinks itself only κατὰ μετάληψιν τοῦ νοητοῦ (Met. Λ 7, 1072b20), so that the initiative comes from the νοητόν (Met. Λ 7, 1072a30: νοῦς ὑπὸ τοῦ νοητοῦ κινεῖται). The divine mind is the best of all because it is thinking what it is thinking: διὰ γὰρ τοῦ νοεῖν τὸ τίμιον αὐτῷ ὑπάρχει (Met. Λ 9, 1074b21). The attempt to analyse the structure of the thinking of the Prime Mover’s thought, instead of seeking the content of it, shows more precisely what Aristotle really means when he says that the νοῦς, in thinking itself, moreover is thinking some content of more or less distributed elements of thought in the plural. In thinking its objects, each act of thinking can also experience itself. That’s all. The same structure is operating in the Prime Mover’s νοῦς as νόησις νοήσεως. It is the plain consciousness of itself, nothing more, without any particular content. Whoever wishes to contemptuously call this narcissistic and not Aristotelian may do this for his own satisfaction, but he is off track. There is absolutely no need for the Prime Mover’s ontic dignity to take care of the world and what is therein. Many commentators dislike this aristocratic attitude of Aristotle, which is behind his silence, and therefore they prefer the Christian interpretation of the Prime Mover’s being. Aristotle’s attitude is what Nietzsche calls “das Pathos der Distanz”. This is not in accordance with everybody’s taste. I am afraid that there will be never a consensus concerning the Prime Mover’s noe¯sis. 29

Cf. Parmenides 28 B 8, 34 D.-K.: ταὐτὸν δ᾽ ἐστὶ νοεῖν τε καὶ οὕνεκεν ἔστι νόημα.

Aristotle’s Silence about the Prime Mover’s Noe¯sis

243

Bibliography Alexander and Pseudo-Alexander, in: M. Hayduck (ed.), Commentaria in Aristotelem Graeca, vol. I, Berlin 1891. Berti, E.: Unmoved mover(s) as efficient cause(s) in Metaphysics L 6, in: M. Frede/D. Charles (eds.), Aristotle’s Metaphysics Lambda. Symposium Aristotelicum, Oxford 2000b, 181–206. Bordt, M.: Aristoteles’ Metaphysik XII, Darmstadt 2006. Burnyeat, M. F.: Aristotle’s Divine Intellect, Milwaukee 2008. Broadie, S.: Que Fait le Premier Moteur d’Aristote? Sur la théologie du livre Lambda de la « Métaphysique », in: Philosophique de la France et de l’Étranger, 183 1993, 375–411. Brunschwig, J.: Metaphysics L 9: A Short-Lived Thought-Experiment? In: M. Frede/D. Charles (eds.), Aristotle’s Metaphysics Lambda. Symposium Aristotelicum, Oxford 2000, 275–306. Cornford, F. M.: Before and After Socrates, Cambridge 1932. De Filippo, J. G.: Aristotle’s Identification of the Prime Mover as God, in: CQ, 44 1994, 393–409. De Koninck, T.: Aristotle on God as Thought Thinking Itself, in: L. P. Gerson (ed.), Aristotle. Critical Assessments I: Logic and Metaphysics, London 1999, 365–402. Elders, L. (ed.): Aristotle’s Theology, A Commentary on Book L of Metaphysics, Assen 1972. Flashar, H. (ed.): Aristoteles: Fragmente zu Philosophie, Rhetorik, Poetik, Dichtung, übers. und erläut. von Hellmut Flashar, Berlin 2006. Frede, D.: Theophrasts Kritik am unbewegten Beweger des Aristoteles, in: Phronesis, 16 1971, 65–69. Frede, M.: Introduction, in: M. Frede/D. Charles (eds.), Aristotle’s Metaphysics Lambda. Symposium Aristotelicum, Oxford 2000a, 1–52. Freudenthal, J./Fränkel, S.: Die durch Averroes erhaltenen Fragmente Alexanders zur Metaphysik des Aristoteles: Berlin 1885. Gadamer, H.-G.: Aristoteles’ Metaphysik XII, Frankfurt a.M. 1948 Genequand, C. (ed.): Ibn Rushd’s Metaphysics. A translation with Introduction of Ibn Rushd’s commentary on Aristotle’s Metaphysics, Book Lambda, Leiden 1986. Guthrie, W. K. C.: The Development of Aristotle’s Theology I, in: CQ, 27 1933, 162–171. Guthrie, W. K. C.: The Greeks and their Gods, London 1950. Heiberg, I. L. (ed.): Simplicius: In Aristotelis De Caelo Commentaria (CAG VII), Berlin 1984. Horn, C.: In welchem Sinn enthält Metaphysik Lambda eine Theologie? In: Jahrbuch für Religionsphilosophie, 1 2002, 28–49.

244

Maria Liatsi

Horn, C.: Aristoteles in Metaphysik Lambda über das Verhältnis von Mythologie und Rationalität, in: J. Halfwassen/M. Gabriel (eds.), Kunst, Metaphysik und Mythologie, Heidelberg 2008, 19–31. Kahn, C. H.: The Place of the Prime Mover in Aristotle’s Teleology, in: A. Gotthelf (ed.), Aristotle on Nature and Living Things, Bristol–Pittsburgh 1985, 183–205. Kosman, A.: Aristotle’s Prime Mover, in: M. L. Gill/J. G. Lennox (eds.), Self Motion: From Aristotle to Newton, Princeton 1994, 135–153. Kretzmann, N.: Omniscience and Immutability, in: Journal of Philosophy, 63 1966, 409–421. Laks, A.: Metaphysics L 7, in: M. Frede/D. Charles (eds.), Aristotle’s Metaphysics Lambda. Symposium Aristotelicum, Oxford 2000, 207–243. Laks, A./Most, G. W. (eds.): Théophraste, Métaphysique, Paris 1993. Manuwald, B.: Studien zum Unbewegten Beweger in der Naturphilosophie des Aristoteles, Stuttgart 1989. Oehler, K.: Aristotle on Self-knowledge, in: Proceedings of the American Philosophical Society, 118 1976, 463–506. Oehler, K.: Logic of Relations and Inference from Signs in Aristotle, in: Ars Semeiotica, 4 1981, 237–246. Oehler, K.: Die Anfänge der Relationenlogik und der Zeichenschluß bei Aristoteles, in: Zeitschrift für Semiotik, 4 1982, 259–266. Oehler, K.: Der Unbewegte Beweger des Aristoteles, Frankfurt a.M. 1984 Oehler, K.: Die Lehre vom Noetischen und Dianoetischen Denken bei Platon und Aristoteles. Ein Beitrag zur Erforschung der Geschichte des Bewusstseinsproblems in der Antike, Hamburg 21985. Peck, A. L. (ed.): Aristotle. Parts of Animals. The Loeb Classical Library 323, Cambridge, Mass.–London 1937. Politis, V.: Aristotle and the Metaphysics, London–New York 2000. Ross, W. D. (ed.): Aristotle’s Metaphysics. A Revised Text with Introduction and Commentary, Oxford 1924. Ross, W. D./Forbes, F. H. (eds.): Theophrastus. Metaphysics, Oxford 1929. Rosemann, P. W.: Das Aristotelische Problem der Selbstbezüglichkeit des Unbewegten Bewegers in der Kommentierung Ibn Rushds, in: Zeitschrift für philosophische Forschung, 40 1986, 543–560. Sandbach, F. H.: A Transposition in Aristotle’s Metaphysics L 9, 1074 b, in: Mnemosyne, 8 1954, 39–43. Scheibe, E.: Über Relativbegriffe in der Philosophie Platons, in: Phronesis, 12 196, 28–49. Schüssler, W.: Als protestantischer Theologe in philosophischem Material, in: I. Nord/Y. Spiegel (eds.): Spurensuche. Lebens- und Denkwege Paul Tillichs 2001, 139–151.

Aristotle’s Silence about the Prime Mover’s Noe¯sis

245

Schwegler, A. (ed.): Die Metaphysik des Aristoteles. Grundtext, Übersetzung und Kommentar nebst erläuternden Abhandlungen, Tübingen 1847. Sedley, D.: Metaphysics L 10, in: M. Frede/D. Charles (eds.), Aristotle’s Metaphysics Lambda. Symposium Aristotelicum, Oxford 2000, 327– 350. Wieland, W.: Die Ewigkeit der Welt (Der Streit zwischen Joannes Philoponus und Simplicius), in: D. Henrich (ed.), Die Gegenwart der Griechen im neueren Denken. Festschrift H.-G. Gadamer, Tübingen 1960, 291–316. Zeller, E.: Die Philosophie der Griechen in ihrer geschichtlichen Entwicklung, Leipzig 41921.

Cases of Celestial Teleology in Metaphysics Λ ISTVÁN BODNÁR

Aristotle’s remark in Metaphysics Λ about an eternal substance which causes motion in the outermost celestial sphere as ‘that for the sake of which’, or as an object of love does, has received extensive and detailed scrutiny. In this paper I will start out from that central case of celestial teleology. My overarching aim will be to discuss this in the context of several lesser instances of celestial teleology – ones where the goal to which the activity of a celestial entity is directed is within the celestial domain itself. All of these cases of celestial teleology, just by the fact that they involve entities of truly cosmic dimensions, can also be taken as cases of cosmic teleology. But this latter name might turn out to be a misnomer. The teleological structures I am interested in here are of vast extension, but neither do they need to be all-encompassing, nor must they incorporate the operations of the lower regions of the cosmos. Even though the operations of the entities in these teleological structures will have causal relevance to the sublunary domain, and there will be important contrasts to be drawn between these celestial cases and the sublunary ones.

I. Before turning to the cases of intra-cosmic celestial teleology, I shall chart my understanding of the most crucial, extra-cosmic cases of celestial teleology. The very first of these is the causal connection between the first principle, which moves as the object of desire and thought does (kinei de ho¯de to orekton kai to noe¯ton, 1072a26), or, as Aristotle puts it a few lines later, as an object of love does (ho¯s ero¯menon 1072b3). Then, somewhat later, in Chapter 8, it will be submitted that for each component circular motion of the celestial bodies there is a separate celestial sphere performing this revolution, under the causal influence of a dedicated, separate unmoved mover for each one of them.1 1

Note that without this presupposition the task Aristotle sets before himself in this eighth chapter, that of counting up the unmoved, supra-natural substances, could never get off the ground at all.

248

István Bodnár

This is not a statement Aristotle announces only in passing. Apart from the claim that the prime supra-physical entity serves as a goal of the motion of the celestial sphere, we are provided with some details of the mode of existence of such entities: each and every one of them is eternal, a substance and an actuality.2 Furthermore, as Aristotle submits, these entities are substances which do not undergo any change, and have reached the best state according to themselves.3 Very little, however, is provided about the details of their mode of operation apart from the fundamental detail – announced several times over – that they are responsible for the motion as goals. This much, however, is already sufficient to contrast these with other instances of unmoved movers, like the souls. Souls do not cause motion the way some object of desire, of thought, or of love does. To begin with, souls are not external to the entities they move. Instead, they are intimately joined to the movable physical body – to the extent that they are the first actualities of these living organic bodies. Besides the souls of perishable living beings, arts and crafts also qualify as instances of unmoved movers. Unlike souls, an art most often is the cause of motion in an external entity – one different from the practitioner of the art in question. Nevertheless, art and craft will always be intimately connected to the soul, and also to the body of the practitioner of the art, who e.g. administers the cure and causes the requisite changes in the patient. Therefore, also in the case of art and craft there is an intricate connection between the unmoved mover and the particular perishable body of the practitioner, which means that if the exercise of this art involves the motion of the body as a whole, these entities, too, will impart motion, albeit incidentally, to themselves. The unmoved movers of the celestial realm are of a different kind. At least the unmoved mover of the sphere of the fixed stars is exempt from all motion – both motion according to itself, and also incidental motion.4 When in the next lines Aristotle expands this to a characterisation of the unmoved movers of the lower celestial spheres, he only stresses their immobility according to themselves.5 This may, or may not allow for their incidental mo2 3 4 5

†τοίνυν† ἔστι τι ὃ οὐ κινούμενον κινεῖ, ἀΐδιον καὶ οὐσία καὶ ἐνέργεια οὖσα (1072a24–26) πᾶσαν φύσιν καὶ πᾶσαν οὐσίαν ἀπαθῆ καὶ καθ᾽ αὑτὴν τοῦ ἀρίστου τετυχηκυῖαν τέλος εἶναι δεῖ νομίζειν, (1074a19–20) ἡ μὲν γὰρ ἀρχὴ καὶ τὸ πρῶτον τῶν ὄντων ἀκίνητον καὶ καθ᾽ αὑτὸ καὶ κατὰ συμβεβηκός, κινοῦν δὲ τὴν πρώτην ἀΐδιον καὶ μίαν κίνησιν· (1073a23–25). ἐπεὶ δὲ τὸ κινούμενον ἀνάγκη ὑπό τινος κινεῖσθαι, καὶ τὸ πρῶτον κινοῦν ἀκίνητον εἶναι καθ᾽ αὑτό, καὶ τὴν ἀΐδιον κίνησιν ὑπὸ ἀϊδίου κινεῖσθαι καὶ τὴν μίαν ὑφ᾽ ἑνός, ὁρῶμεν δὲ παρὰ τὴν τοῦ παντὸς τὴν ἁπλῆν φοράν, ἣν κινεῖν φαμὲν τὴν πρώτην οὐσίαν καὶ ἀκίνητον, ἄλλας φορὰς οὔσας τὰς τῶν πλανήτων ἀϊδίους (ἀΐδιον γὰρ καὶ ἄστατον τὸ κύκλῳ σῶμα· δέδεικται δ᾽ ἐν τοῖς φυσικοῖς περὶ τούτων), ἀνάγκη καὶ τούτων ἑκάστην τῶν φορῶν ὑπ᾽ ἀκινήτου τε κινεῖσθαι καθ᾽ αὑτὴν καὶ ἀϊδίου οὐσίας. (1073a26–34).

Cases of Celestial Teleology in Metaphysics Λ

249

tion. If the fact that Aristotle does not stress the lack of incidental motion in their case is just an omission, without further significance, the status of these supra-sensible celestial movers will be exactly identical to that of the mover of the first celestial sphere. No doubt, in this case the way the unmoved movers are causally responsible for the motion of the spheres moved by them will also be the same, in so far as they move as objects of love do. On the other hand we may take the omission of the qualification that these further unmoved movers are not said to be unmoved in an incidental manner significative, that they can be taken to be moved incidentally. Nevertheless, this need not change the status of these movers in a fundamental way. Indeed, Aristotle in Physics VIII 6 allowed for the incidental motion of the principles of some celestial entities – of those ones that perform several motions. What he stressed also in that instance was that unlike mortal living beings, whose principle of motion may undergo incidental motion by itself, these principles of celestial motion undergo incidental motion only by something else, presumably by some other celestial body, or by the moving principle of another celestial sphere.6 And we might go a step further. Even if Aristotle were to allow here in the Metaphysics incidental motion by themselves to these movers of planetary spheres – contrary to what he said in the Physics – the status of these further unmoved movers, as far as the bodies moved by them are concerned, need not differ at all from the causal link between the first unmoved mover and the sphere of the fixed stars, the body moved by that unmoved mover. In short: the somewhat vague phrase that these entities also move the spheres as the objects of striving or of love indicates throughout the same kind of causal link between mover and object moved. As a further consideration one should add that, earlier in Metaphysics Λ, Aristotle submitted that the unmoved mover is the cause of motion just as the object of striving, or of love is. Then, at the beginning of Chapter 8 he announced that he was about to investigate whether there are one or more such substances.7 This means that without some specific indication to the contrary we should presume that the type identity between these substances extends also to the crucial points of their mode of operation. This is also borne out by a further crucial consideration. Aristotle after the enumeration of the celestial spheres – first according to Eudoxus’ theory,

6

7

οὐκ ἔστιν δὲ τὸ αὐτὸ τὸ κινεῖσθαι κατὰ συμβεβηκὸς ὑφ᾽ αὑτοῦ καὶ ὑφ᾽ ἑτέρου· τὸ μὲν γὰρ ὑφ᾽ ἑτέρου ὑπάρχει καὶ τῶν ἐν τῷ οὐρανῷ ἐνίαις ἀρχαῖς, ὅσα πλείους φέρεται φοράς, θάτερον δὲ τοῖς φθαρτοῖς μόνον. (259b28–31) Πότερον δὲ μίαν θετέον τὴν τοιαύτην οὐσίαν ἢ πλείους, καὶ πόσας, δεῖ μὴ λανθάνειν. (1073a14 f.).

250

István Bodnár

then according to that of Callippus, and finally according to the Callippan theory supplemented by his own retroactive, or rewinding spheres – submits that the number of celestial motions and their movers cannot be any larger than the tally given by him, with due allowance for alternative accounts. To this end he first submits that any supra-physical, perfect substance has to serve as a goal,8 and then he completes this consideration by deploying a reductio argument: if there were any more additional entities of the same type, that would imply the presence of further celestial motions. In section IV below we shall have to address the impossibility of the presence of these further celestial motions. What is of crucial importance at this stage is that Aristotle submits that these putative further celestial movers would also have ‘to cause change as being an end of movement’.9 What this passage stressed is that these are all natures, or substances without the influence of any external activity, and they are in the best state in virtue of themselves (πᾶσαν φύσιν καὶ πᾶσαν οὐσίαν ἀπαθῆ καὶ καθ᾽ αὑτὴν τοῦ ἀρίστου τετυχηκυῖαν κτλ. 1074a19–20). In this statement, accordingly, excellence is given as the reason why these substances serve as goals.10 Intimately connected to this link between the excellence of these entities and their status as goals is a further bit of information. Just before announcing that the unmoved mover moves as an object of love, he submits that there is a distinction between two types of ‘that for the sake of which’ – the one that is ‘that for the sake of which’ as a beneficiary, and the other which is a goal, a ‘that for the sake of which’ just for the sake of which, i.e. not as a beneficiary. One of these types he submits – the latter – is also present in unmoved entities.11 This means that the striving, or the celestial motion caused by these unmoved entities does not induce any change, or any additional excellence whatsoever in these unmoved entities.12 8

9 10 11 12

ἔτι δὲ πᾶσαν φύσιν καὶ πᾶσαν οὐσίαν ἀπαθῆ καὶ καθ᾽ αὑτὴν τοῦ ἀρίστου τετυχηκυῖαν τέλος εἶναι δεῖ νομίζειν, οὐδεμία ἂν εἴη παρὰ ταύτας ἑτέρα φύσις, ἀλλὰ τοῦτον ἀνάγκη τὸν ἀριθμὸν εἶναι τῶν οὐσιῶν. (1074a19–22). εἴτε γὰρ εἰσὶν ἕτεραι, κινοῖεν ἂν ὡς τέλος οὖσαι φορᾶς· (1074a22–23). Note that this need not exclude that they could also fulfil some other Aristotelian causal role. ὅτι δ᾽ ἔστι τὸ οὗ ἕνεκα ἐν τοῖς ἀκινήτοις, ἡ διαίρεσις δηλοῖ· ἔστι γὰρ τινὶ τὸ οὗ ἕνεκα )καὶ* τινός, ὧν τὸ μὲν ἔστι τὸ δ᾽ οὐκ ἔστι. (1072b1–3) This is borne out by Aristotle’s characterisation of these entities as “natures and substances without the influence of any external activity, and they are in the best state in virtue of themselves” (πᾶσαν φύσιν καὶ πᾶσαν οὐσίαν ἀπαθῆ καὶ καθ᾽ αὑτὴν τοῦ ἀρίστου τετυχηκυῖαν κτλ;, 1074a19–20). One could even submit that these eternal, unmoved entities are what they are without regard to any effect they have on the celestial spheres. For an account along these lines see Bordt, M. 2011: Why Aristotle’s God is Not the Unmoved Mover, in: Oxford Studies in Ancient Philosophy, 40, 91–109.

Cases of Celestial Teleology in Metaphysics Λ

251

Accordingly, the effect of these movers as goals exerts its influence only in the moving celestial spheres. Somehow they constitute the goal that the motion induced in these spheres tracks. Or, alternatively, there is some other goal-oriented activity these celestial spheres perform, and their circular motion is a consequence of this goal-oriented activity they have.13 What needs to be stressed is that even if there is, or even if there were, such an additional, distinct goal-oriented activity, Aristotle’s description does not allow for a wedge between this activity and the motion these celestial spheres perform. The motion belongs to them essentially; it cannot be just a casual spin-off of some more fundamental activity.14 The celestial sphere can pursue the goal set before itself only by carrying out its circular motion. Such circular motion, as Aristotle stressed earlier in Metaphysics Λ, has a distinguished status because these celestial motions are (1) the only motion which is continuous – other motions inherently possess starting-points and end-points;15 (2) they are eternal, and manifestly so;16 and (3) locomotion, and within it this circular motion induced by the unmoved mover, is the primary motion.17 In other words, one can submit that this kind of motion, performed by the celestial sphere, with its elevated status, is congruent with the mover which induces it as the goal of this celestial sphere, or the goal of the motion performed by this celestial sphere.18 13

14

15 16 17 18

For the distinction see Broadie, S. 2009: Heavenly Bodies and First Causes, in: G. Anagnostopoulos (ed.), A Companion to Aristotle, Oxford, 240: “Why does the sphere’s love of its unmoved mover give rise to rotation? Are there two sphere-acitivities, rotation and loving the unmoved mover, or could these be different aspects of a single activity?” With this compare Theophrastus’ aporetic remark in his Metaphysics, εἰ δ᾽ οὖν τῆς κυκλικῆς αἴτιον τὸ πρῶτον, οὐ τῆς ἀρίστης ἂν εἴη· κρεῖττον γὰρ ἡ τῆς ψυχῆς, καὶ πρώτη δὴ καὶ μάλιστα ἡ τῆς διανοίας, ἀφ᾽ ἧς καὶ ἡ ὄρεξις. (5b7–10) Note, however, a fundamental difference between the presentation in my main text, and Theophrastus’ query: Theophrastus speaks about two motions, the circular motion on the one hand, and the motion of the soul, which induces this circular motion. This move is inadmissible by what Aristotle submits here. By his lights the fundamental motion is the circular motion of the celestial spheres. κίνησις δ᾽ οὐκ ἔστι συνεχὴς ἀλλ᾽ ἢ ἡ κατὰ τόπον, καὶ ταύτης ἡ κύκλῳ. (1071b10f) καὶ ἔστι τι ἀεὶ κινούμενον κίνησιν ἄπαυστον, αὕτη δ᾽ ἡ κύκλῳ (καὶ τοῦτο οὐ λόγῳ μόνον ἀλλ᾽ ἔργῳ δῆλον), ὥστ᾽ ἀΐδιος ἂν εἴη ὁ πρῶτος οὐρανός. (1072a21–23) φορὰ γὰρ ἡ πρώτη τῶν μεταβολῶν, ταύτης δὲ ἡ κύκλῳ· ταύτην δὲ τοῦτο κινεῖ. (1072b8–10) Note that Aristotle’s very meagre pointers allow for both formulations. When he submits that the unmoved mover moves as the object of striving, the object of thought, or the object of love does, one could suppose that the mover stands in this relationship to the entity that performs the motion. In the parallel cases it is such an entity which does the striving, which thinks, or which loves the object of love and so is under the causal influence of this goal. The other formulation is present in the passage 1074a14–23. There the unmoved movers are presented as the goals of the motions induced by them. Needless to say, there need not be any tension between the two formulations, as the celestial objects are pursuing these goals by performing their circular motions, hence their circular motions are also tracking this very same goal.

252

István Bodnár

This congruence between the goal and the motion performed remains in force even under two further restricting considerations. One is that the motion performed by a celestial sphere is apparently natural: it is the expression of the nature of the corporeal element of this entity.19 The other consideration is that this should not obliterate the gulf between the activity of the supernatural unmoved mover, and that of the celestial sphere. Indeed, without this fundamental divide the celestial entity could literally attain the excellence that is set before it as an object of striving and love. In that case it could be fully assimilated to the object of its striving. This, then should suggest that this relationship of striving and love (or the one analogous to striving and love) is not one of assimilation, even though the two entities involved have some essential connections as far as their activity is concerned. This connection, forged by the goal-directedness of the motion, is what can be expressed with the claim, in the traditional parlance, that the celestial spheres imitate – as much as they can – the excellence, the eternal activity of their supra-physical movers. Nevertheless, at this point one should note that Aristotle never submits in so many words that there is a relation of imitation between these two entities. Whenever he speaks about imitation, it is about the motion of perishable physical entities imitating the eternal motion of celestial beings.20 Perhaps the fact that Aristotle does not speak about the imitation of the activity of the unmoved mover in the case of the motion of the celestial sphere may underline that there is an even more unbridgeable gulf between the two vastly different domains, of vastly different worth and excellence than the one that exists between the eternal celestial realm and the perishable sublunary entities. This, however, will not remove the original claim, that there is some congruence between the unmoved mover – i.e., the goal of the moving enti19

20

At the very least this corporeal element should not present an obstacle to this eternal motion. But we can be confident that Aristotle requires more from the material frame of the celestial spheres here: at Metaphysics 1073a31f he explicitly refers to the proofs propounded in the physical works (presumably the De caelo is meant), and submits that the body moving in a circle is eternal and without rest (ἀΐδιον γὰρ καὶ ἄστατον τὸ κύκλῳ σῶμα· δέδεικται δ᾽ ἐν τοῖς φυσικοῖς περὶ τούτων). Here apart from the actual doctrinal statement the very terminology that Aristotle uses, namely the telling phrase ‘the body that moves in a circle’ (τὸ κύκλῳ σῶμα), is also significant. On Aristotle’s programmatic statement in Chapter 1 of Metaphysics Λ, that the elements of eternal perceptible substance – or the elements of both eternal and perishable perceptible substance – have to be grasped, see Fazzo, S. 2013: Heavenly Matter in Aristotle: Metaphysics Lambda 2, in: Phronesis, 58, 160–175. See Broadie (see note 13), 241: “It is worth noting that Aristotle does not state that the spheres by their motion imitate the activity of the unmoved movers, although he is commonly interpreted as holding this. His explicit remarks about cosmic imitation ascribe it to perishables.”

Cases of Celestial Teleology in Metaphysics Λ

253

ty – and the motion performed by the moving entity under the influence of this goal. I think we are even entitled to submit that this congruence is supported by the teleological structure involved. The object of striving and love, exactly by being in this very role, elicits this congruence from the celestial sphere. There are two further remarks to be made at this juncture. Were the congruence elicited by the unmoved mover an instance of the identity of a causally efficacious property, we could speak about a standard case of efficient causation, one where the requirement of causal synonymy would fully apply. The restriction – that it is not the case that a causally efficacious property would be transmitted in the process of causation – is sufficient reason to doubt that the unmoved mover would be an efficient cause, without further qualifications. But – unlike some other objects of striving – this goal does bring about the congruence between mover and moved we have been talking about. This may very well allow for taking this mover as a moving cause, albeit only in this rather restricted sense. The other consideration pertains to the role of celestial souls. Specifically, it is the question of whether it is legitimate to speak about the celestial bodies as ones having souls in the first place. The terms of comparison – that the unmoved mover is operative as an object of striving, or of thought, or of love – suggest activities that only ensouled beings can perform.21 So it is quite natural to presume that the celestial bodies are ensouled, and their souls are crucially involved in this eternal striving, and love.22 Nevertheless, what the lack of any further specification or indication besides these comparisons suggests is that all the psychological setup of the celestial entities contribute is to perform this striving, which issues in the celestial revolution. In some sense the case is parallel with that of the role of the elemental tendency of the constitutive material frame of the celestial bodies. As I suggested above, the elemental tendency for circular motion of the celestial bodies remains in force also in this book of the Metaphysics. But this is not a tendency which could be responsible for the motion of the celestial bodies in the absence of the object of striving and love, the goal tracked by this motion. It is important to stress that this does not rule out that material frame and soul have an important, or even indispensable, role to play. What Metaphysics Λ submits – without stressing it in so many words – is that neither could the celestial bodies perform their motion as their natural motion, so

21 22

This is true also of orekton. Orexis is a term that in its literal sense Aristotle employs about living beings. Here again it is instructive to refer to Theophrastus’ aporia in his Metaphysics, already mentioned in n. 14 above.

254

István Bodnár

that the only entity responsible for this motion would be their nature; nor should the souls of the celestial bodies bear the brunt of causal efficacy on their own for these movements.

II. As Aristotle explicitly connects the discussion in Metaphysics Λ to his physical writings, we should check how the details of the physical accounts fit in with what we learn about the celestial case in Metaphysics Λ. There are at least two discussions that should be taken into account. The first is Physics II 7, 198a21–27, where Aristotle submits that the physical investigation into causes will have to rely on all four causes. Nevertheless, three of these are intimately interrelated, to the extent that two – the ‘what’ and ‘that for the sake of which’, or to put it otherwise, the form and the end – are often one and the same.23 This form and goal, as it turns out, is the nature of this entity. This already suggests that these considerations will not be transferable to the celestial case without significant readjustment. Even though Metaphysics Λ is silent about the status of the forms, or the souls of the celestial spheres, it is clear that the supra-physical unmoved movers cannot serve in such a dual function, of being in addition to the goals also the forms of the moving physical bodies. This suggests that the role of these supra-physical entities is rather closely analogous to the cases of goal-setting in activities. A standard case of such a teleological explanation could be e.g. that of taking a walk for the sake of strengthening or recovering our health. In such cases there is a mover, the soul, the internal principle of motion and rest, and this soul produces motion which is under the sway of the aim and objective of strengthening or recovering the health of the person doing the exercise. But even this is different in crucial respects from the case of celestial movers. In the case of health the activity of the human person will result in acquiring or preserving health. Celestial movers on the other hand are not something to be attained by the celestial spheres. In order to present an analogous case we should find some activity which successfully tracks a goal that is never to be attained by the entity performing the activity.

23

ὅτι μὲν οὖν τὰ αἴτια ταῦτα καὶ τοσαῦτα, φανερόν· ἐπεὶ δ᾽ αἱ αἰτίαι τέτταρες, περὶ πασῶν τοῦ φυσικοῦ εἰδέναι, καὶ εἰς πάσας ἀνάγων τὸ διὰ τί ἀποδώσει φυσικῶς, τὴν ὕλην, τὸ εἶδος, τὸ κινῆσαν, τὸ οὗ ἕνεκα. ἔρχεται δὲ τὰ τρία εἰς [τὸ] ἓν πολλάκις· τὸ μὲν γὰρ τί ἐστι καὶ τὸ οὗ ἕνεκα ἕν ἐστι, τὸ δ᾽ ὅθεν ἡ κίνησις πρῶτον τῷ εἴδει ταὐτὸ τούτοις· ἄνθρωπος γὰρ ἄνθρωπον γεννᾷ – καὶ ὅλως ὅσα κινούμενα κινεῖ (Phys. II 7.198a21–27).

Cases of Celestial Teleology in Metaphysics Λ

255

The closest parallels to such a goal are apparently the cases where some perishable entity pursues the eternal activity of some imperishable being. Such claims are not restricted in Aristotle to beings endowed with life. The eternal cycle of elemental transformations constitutes a prime example.24 But even more pertinent to the case of celestial striving could be Aristotle’s description of the reproductive function of the vegetative soul. The aim of this activity is to ensure that the living being participates in the eternal and divine in so far as it is capable to do so. Aristotle even submits that every living being desires this eternity and divine status, and all the natural operations of the living being are performed with a view to this aim. After the introduction of the two kinds of goals – one which is ‘for the sake of which’ as a beneficiary, and the other which is a goal, a ‘that for the sake of which’ just for the sake of which – Aristotle emphasizes that the goal set and pursued by the living organism as its highest goal is one which cannot be attained. Nevertheless, by pursuing this goal these living beings are able to participate in this unattainable goal – some less, some more.25 The first thing to note here is that in a manner somewhat similar to the case of the celestial unmoved movers, eternity and the divine serve here as causes of motion for the sublunary living beings. The soul of the living being, with every one of its actions pursues the very same goal throughout, and this determines all of its actions teleologically. One should add that even though this striving is only successful on the level of the species – there will always exist, and there have always existed living beings of the type that do this striving now – the different types of living beings partake of the divine to different degrees: some more, some less. This means that as far as these entities are concerned, we can establish differences between the different kinds, even though all animal and plant species are eternal, and to that extent they should be equally successful in this imitation of the divine. We could formulate various guesses about what this difference might consist in – the obvious ones being differences in the span of life of the individuals of different species, on the one hand, and on the other hand the extent to which the offspring they produce is developed at birth. On this

24 25

Metaphysics Θ 8.1050b28–30. μιμεῖται δὲ τὰ ἄφθαρτα καὶ τὰ ἐν μεταβολῇ ὄντα, οἷον γῆ καὶ πῦρ. καὶ γὰρ ταῦτα ἀεὶ ἐνεργεῖ· καθ᾽ αὑτὰ γὰρ καὶ ἐν αὑτοῖς ἔχει τὴν κίνησιν. τὸ ποιῆσαι ἕτερον οἷον αὐτό, ζῷον μὲν ζῷον, φυτὸν δὲ φυτόν, ἵνα τοῦ ἀεὶ καὶ τοῦ θείου μετέχωσιν ᾗ δύνανται· πάντα γὰρ ἐκείνου ὀρέγεται, καὶ ἐκείνου ἕνεκα πράττει ὅσα πράττει κατὰ φύσιν (τὸ δ᾽ οὗ ἕνεκα διττόν, τὸ μὲν οὗ, τὸ δὲ ᾧ). ἐπεὶ οὖν κοινωνεῖν ἀδυνατεῖ τοῦ ἀεὶ καὶ τοῦ θείου τῇ συνεχείᾳ, διὰ τὸ μηδὲν ἐνδέχεσθαι τῶν φθαρτῶν ταὐτὸ καὶ ἓν ἀριθμῷ διαμένειν, ᾗ δύναται μετέχειν ἕκαστον, κοινωνεῖ ταύτῃ, τὸ μὲν μᾶλλον τὸ δ᾽ ἧττον, καὶ διαμένει οὐκ αὐτὸ ἀλλ᾽ οἷον αὐτό, ἀριθμῷ μὲν οὐχ ἕν, εἴδει δ᾽ ἕν. (De anima II 4.415a28–b 7)

256

István Bodnár

latter count vivipara would partake the most in the eternal and the divine, whereas ovipara, and the animals reproducing through grubs, and finally plants would be further down the scale. Be that as it may, again in contrast to the celestial case the obvious further step down is to compare the cycle of elemental transformations. There – by its very own destruction, and providing for the generation of an element of a different kind – the element makes possible the further links in the chain, among them those in which, later, an elemental mass of the same kind as itself will be generated. Elements – unlike living beings – do not leave lumps of the same element behind after their destruction. But whatever they leave behind will be a link in the chain leading up to the generation of some lumps of the same element as they were. In the end, in both of these types of cases – the imitation of the eternal celestial motion by inanimate elements, and the striving of living beings for eternal and divine existence – the cardinal divide will be between the eternal entity, as a goal, and the perishable beings, pursuing that goal with limited success in the individual case, but achieving some kind of success in the long run, by extending their own existence into posterity. The case of Metaphysics Λ, however, suggests a further divide, this time between two types of eternal entities. In order to set in relief this further divide it is instructive to turn to the difficulties tackled in De caelo II 12. In that chapter, Aristotle announces two difficulties of celestial theory, and provides answers to them. The second of these – which I shall touch upon only in passing – is why it is the case that one of the motions – the motion of the fixed stars – moves an immense multitude of celestial objects, whereas in the case of the planets there are several spheres carrying each of them. The other difficulty is why it is the case that the fixed stars move with a single revolution, while some of the planets move with more circular motions, and finally as we proceed towards the centrally positioned Earth, the Sun and the Moon move with fewer movements than some of the planets.26 An arrangement in 26

Note that something similar might be the case also in the Metaphysics Λ account: the counteracting – or rewinding – spheres do not add a further motion to the tally for the embedded spheres. Rather, they remove a component motion, provided by a carrying sphere. Hence the number of motions performed by the sphere carrying a lower planet is emphatically not the sum of the number of all the carrying and rewinding spheres above this sphere. This is so both on the traditional – and problematic – understanding of how the motions of the planetary system above are factored into those of the lower planetary system, and also on Jonathan Beere’s proposal, according to which between two planetary systems there is no transmission of motion at all, because Aristotle effects complete decoupling between these systems (Beere, J. B. 2003: Counting the Unmoved Movers: Astronomy and Explanation in Aristotle Metaphysics XII.8, in: Archiv für Geschichte der Philosophie, 85, 1–20). For a detailed discussion of these issues see my ‘Aristotle’s rewinding spheres: Three options and their difficulties’ (Bodnár, I. 2005: Aristotle’s rewinding spheres: Three

Cases of Celestial Teleology in Metaphysics Λ

257

which there are more and more component motions performed as one proceeds inwards might have seemed to be more intuitive instead. It is at this point that Aristotle submits that the tenor of the previous discussion needs an important corrective. The celestial entities should not be treated as mere bodies. They are not inanimate entities: they partake of activity and life. And the activity they perform is remarkably similar to that of the perishable living beings: they aim at performing the actions that bring them into the best state. What is clear from Aristotle’s discussion is that this is not just a best state relative to them. Instead, the answer to the aporia submits that there is some best state that some entities possess without any specific action, whereas others need to exert themselves: they need to set this best state before themselves as a goal, and to be in action accordingly. This will require that some of these perform a single motion.27 Others, however, require more exertion; they will need to perform several motions – like someone who needs several types of exercise in order to achieve a healthy state, Aristotle submits. There are two further claims Aristotle makes in the course of this argument. Of central importance for the solution of the aporia is the claim that compared to some entities that perform quite a few activities, some less perfect ones will not even be able to reach that perfect state, and hence it is better for them not to take aim at that best possible state, which is unattainable for them. Instead, they should aim at some related goal, attainable by them through performing fewer activities. What needs also to be taken into account is that Aristotle apparently submits that the most elevated entity, in the best state without any activity, is also a ‘that for the sake of which’: “That which is as being in the best state does not need any activity. For that is ‘that for the sake of which’. Action, on the other hand, consists always of two things: when there is both ‘that for the sake of which’ and ‘that which is for the sake of it’.”28

27

28

options and their difficulties, in: Apeiron, 38, 257–275). Note furthermore that Beere’s proposal is analogous to Ptolemy’s solution in Book II of his Planetary Hypotheses, that (in a paraphrase by Andrea Murshel) “a spherical body will move another spherical body lying within it if their mathematical axes are not collinear” (Murshel, A. 1995: The structure and function of Ptolemy’s physical hypotheses of planetary motion, in: Journal for the History of Astronomy, 26, 40). Performing a single motion could, in principle, mean that each such entity performs the motion specific to itself. This, however, is not the case: fixed stars perform a single motion which is the same for each and every one of them. Indeed, this is where the two aporias of the chapter are conjoined: the multitude of fixed stars perform a single motion in the sense that they all are attached to a single sphere, moving with a single diurnal revolution. Τῷ δ᾽ ὡς ἄριστα ἔχοντι οὐθὲν δεῖ πράξεως· ἔστι γὰρ αὐτὸ τὸ οὗ ἕνεκα, ἡ δὲ πρᾶξις ἀεί ἐστιν ἐν δυσίν, ὅταν καὶ οὗ ἕνεκα ᾖ καὶ τὸ τούτου ἕνεκα. De cael. II 12, 292b4–7.

258

István Bodnár

Here in principle one could propose a deflationary reading, according to which the pronoun αὐτὸ (that) in the clause ‘for that is ‘that for the sake of which’’ (ἔστι γὰρ αὐτὸ τὸ οὗ ἕνεκα) might refer to the best state, and not to the entity having that best state without any activity. Aristotle’s wording in his closing remarks, however, makes it clear that even if the striving of the other entities involved is aiming at the best possible state, that state will be connected to an excellent entity, the ‘most divine principle’, which some of the celestial bodies do attain, while others are unable to do so: “And for this reason the Earth does not move at all, what are near perform few motions – for they do not arrive at the last [goal], but only in so far as they are able to attain to the most divine principle. The first heaven attains to it immediately through a single motion, the ones in between the first [heaven] and the last [heavens]29 arrive at it, but they arrive at it through several motions.”30 This, then is surprisingly similar to what we have in Metaphysics Λ, with goal-oriented activity on the part of the celestial spheres, and some unmoved entity of utmost excellence serving as the goal of these motions. Nevertheless, there are significant differences between the two accounts. Instead of stressing the plurality of such excellent unmoved entities, De caelo emphasises the different ways in which such excellence can be emulated. There can be a difference of motions, and indeed, even within the celestial domain, there can be a difference of achievement. To put it in other words, the homogenisation of the activity of celestial bodies – that each and every one of them pursues its goal with the same success – is balanced in Metaphysics Λ by the explicit introduction of a plurality of such excellences, each of them the goal of a different celestial sphere. Or, to express it in another way: on the De caelo account the planets above the Sun and the Moon can fully partake in divine excellence. The Sun and the Moon cannot – they are eternal, and perform fully regular motion; nevertheless, the very fact that their motion is of fewer components suggests that there is no reason for them to try any harder. Their motions attain as closely to the most divine principle as possible, but this is a limited success

29

30

Note that the ‘last heavens’ here are in plural, not a single celestial sphere is meant but at least the ones to which Sun and Moon are attached, or the ones which carry these two celestial luminaries. Καὶ διὰ τοῦτο ἡ μὲν γῆ ὅλως οὐ κινεῖται, τὰ δ᾽ ἐγγὺς ὀλίγας κινήσεις· οὐ γὰρ ἀφικνεῖται πρὸς τὸ ἔσχατον, ἀλλὰ μέχρι ὅτου δύναται τυχεῖν τῆς θειοτάτης ἀρχῆς. Ὁ δὲ πρῶτος οὐρανὸς εὐθὺς τυγχάνει διὰ μιᾶς κινήσεως. Τὰ δ᾽ ἐν μέσῳ τοῦ πρώτου καὶ τῶν ἐσχάτων ἀφικνεῖται μέν, διὰ πλειόνων δ᾽ ἀφικνεῖται κινήσεων. De cael. II 12.292b19–25.

Cases of Celestial Teleology in Metaphysics Λ

259

story only. Their activity is not fully determined, fully controlled by this most divine excellence. They have that entity as their goal, but owing to their limitations they have to calibrate their activity to a considerable extent. And this is true not only of those striving unsuccessfully. Perhaps the activity of the first celestial entity – attaining to the most divine principle immediately, through a single motion – is fully controlled by this most divine principle. This however cannot possibly be the case for any of the other celestial bodies: neither for the lower celestial bodies – on which see below –, nor for the intermediate ones, which perform several motions. Their several motions are conditioned by their own limitations, and not just by the excellence of the divine principle which they pursue. As we move further down, this feature of the De caelo account will be even further articulated: Those planets which move with a fewer number of motions than the ones above them, due to their even more pronounced limitations, will pursue a limited version of the goal only. Note that Aristotle’s formulation – ‘they do not arrive at the last [goal], but only in so far as they are able to attain to the most divine principle’ – does not suggest that this limited version of the goal would be a distinct goal in its own right. Instead, this further restriction describes a case where the complex motion of the celestial body corresponds even less to the excellence the celestial body pursues, and it is even less under the causal influence and control of ‘the divine principle’ due to the endemic inherent limitations of these lower celestial entities. Note, however, that this difference between the first aporia of De caelo II 12 and Metaphysics Λ is also one of the subjects of motion, an issue I shall broach in section IV in some detail. Indeed, on this very score we can also contrast the first aporia of De caelo II 12 to the second one. The first aporia stresses the differences in how many motions each celestial luminary performs, and arranges them in a grand scheme of a scala naturae of natural excellences, reaching down all the way to the Earth. In contrast, the second aporia takes the spheres as the centres of activity, and does not address the issue of whether they perform their motion under the causal influence of a divine principle that the celestial luminaries were said to be pursuing in their different ways. Indeed, the exposition of the aporia starts out neither from the spheres, nor from the individual celestial luminaries, but from the celestial motions. It is concerning these celestial motions that the aporia asks why it is the case that one of them carries a vast multitude of bodies, whereas no other carries more than a single celestial luminary. It is in the answer to this difficulty that the spheres take centre stage. First the answer will be analogous to what we had in the first aporia: the very first celestial sphere vastly exceeds all the rest in excellence, hence the difference in the number of luminaries carried by it, and by the other celestial spheres. Then, however, Aristotle also adds that

260

István Bodnár

“The other motions have one body also for this reason, that the ones which come before the last, the one which has the luminary,31 move many bodies, because the last sphere moves being set in many spheres, and each of the spheres happens to be a body.”32 This, then, means that after submitting that the first sphere vastly exceeds the rest, here we have a suggestion according to which there need not be such a fundamental divide between the sphere of the fixed stars, and at least some of the planetary moving spheres. Both aporias stress instances of vast disparity between the capacities and works of different entities in the celestial domain. But as it turns out, the disparities revealed do not match across the two accounts. Actually, this was only to be expected: before providing his answers to these difficulties, Aristotle also submitted that we can expect only limited and tentative results in this distant domain, which is fundamentally different from our sublunary cosmos.33

III. This suggests that the homogenisation of the celestial entities, then, produces a clear-cut contrast between celestial and sublunary entities: each celestial entity is fully under the sway of the goal it pursues, whereas even if according to pronouncements at other places in the Metaphysics the sublunary ones strive for the excellence of these entities, they remain hopelessly distant from attaining in their own particular case the perfection they emulate. I submit that the difference between the types of influence that ‘the eternal and the divine’ exert on the two domains probably has to do with this contrast. In the case of the celestial entities, accordingly, the fact that their striving in some sense results in attaining the perfection they aim at (or in some other cases at least attaining some closely correlated perfection) may be a reason why this excellence is at the same time their mover.34 31 32

33

34

The Greek has ἄστρον ‘star’. If we were to use ‘star’ in this instance, we should keep reminding ourselves that planets are wandering stars, in contrast to the fixed ones. Καὶ ἔτι διὰ τόδε ἓν ἔχουσι σῶμα αἱ ἄλλαι φοραί, ὅτι πολλὰ σώματα κινοῦσιν αἱ πρὸ τῆς τελευταίας καὶ τῆς ἓν ἄστρον ἐχούσης· ἐν πολλαῖς γὰρ σφαίραις ἡ τελευταία σφαῖρα ἐνδεδεμένη φέρεται, ἑκάστη δὲ σφαῖρα σῶμά τι τυγχάνει ὄν. De cael. II 12.293a4–8. This can also underline that the results Aristotle introduces here are only tentative and preliminary ones, see Mariska Leunissen 2010: Explanation and Teleology in Aristotle’s Science of Nature, Cambridge, ch. 5. Note that from this it does not follow that ‘the eternal and the divine’ pursued by sublunary perishable animals is not their mover. Rather, the fact that they are unable to attain this excellence themselves leaves the case open.

Cases of Celestial Teleology in Metaphysics Λ

261

The homogeneity of the celestial domain as far as the success of striving is concerned has an important counterpart: Although De caelo II 12 does not exclude that there are several different excellent entities, each of them due to their own perfection being a goal of emulation for other entities, the fact that different entities can strive for their perfection in different ways and with different results allows that a single such goal of striving is sufficient for all the celestial entities. Not so in Metaphysics Λ 8: Here Aristotle stresses that each motion has to be moved by a dedicated transcendent mover, by a different goal of its own (1073a26–34). But – besides the fact that the first mover has to be the principle, and the first item of all the entities (as submitted at 1073a23–2535) – Aristotle does not specify or even hint at differences among the transcendent celestial movers. Granted, the initial statement about the first mover of the sphere of the fixed stars requires that the first transcendent mover must be not only first among all the transcendent movers there are, but also a principle, if not of them, then at least in contrast to them.36 Nevertheless, we are left in the dark as to what this status of being a principle consists in. Most importantly, the sentence characterising these supra-physical entities at 1074a19–23 speaks about every celestial mover in terms of ‘having attained to the best in virtue of itself’.37 So, whatever the principality of the first unmoved mover consists in, the other unmoved movers have to attain their own perfection on their own, and not in virtue of something that is provided for them by this first entity. This need not exclude that there are some crucial connections between the first unmoved mover and the other unmoved movers that are not this first one. Nevertheless, even if this is so, Aristotle’s formulation in lines 1074a19–23 does not allow that this connection could also include that the first unmoved mover serves as an object of striving for the other unmoved movers in the same way that it is the object of striving for the sphere of the fixed stars. That way these other unmoved movers could not have attained to the best by virtue of themselves.

35 36

37

ἡ μὲν γὰρ ἀρχὴ καὶ τὸ πρῶτον τῶν ὄντων ἀκίνητον καὶ καθ᾽ αὑτὸ καὶ κατὰ συμβεβηκός. Note furthermore 1074a31–38, where the unicity of the heaven is argued for from the consideration that if there were a plurality of heavens, each of them should have a unique principle of its own. (This plurality of dedicated principles would then imply the absurdity that in this case they should also involve matter.) πᾶσαν φύσιν καὶ πᾶσαν οὐσίαν ἀπαθῆ καὶ καθ᾽ αὑτὴν τοῦ ἀρίστου τετυχηκυῖαν τέλος εἶναι δεῖ νομίζειν.

262

István Bodnár

IV. In concluding I should return to an issue I have mentioned but have not discussed. This is the issue of whether in addition to the goals that are not beneficiaries there are also goals as beneficiaries in the celestial region. In some sense such beneficiaries should be trivially present also in that domain. Once it is admissible to speak about an object of striving, whatever entity does this striving will also qualify as a beneficiary of this activity provided the activity is not unsuccessful. And if this striving is eternal, and if it is not eternally unsuccessful, then it should bring eternal benefits for the entity involved. Unless we accept this consideration, we should countenance the very unpalatable alternative that these entities are engaged in eternal striving which is at least futilely indifferent to them, if not a downright stressful letdown.38 Accordingly, even though Aristotle never makes this explicit, we may be entitled to assert that each celestial sphere performs its proper motion for its own sake, for its own benefit.39 Nevertheless, Aristotle does make a further, more momentous claim, that “no movement will be for the sake of itself, nor for the sake of some other movement, but for the sake of the stars. For if one movement were for the sake of another movement, that movement also would have to be for the sake of something else. So – since it is not possible to proceed into infinity, the end of every motion will be one of the divine moving bodies moving through the heaven.”40

38

39

40

Cf. Aristotle’s considerations at De caelo II 1, 284a27–35 against the celestial spheres performing their revolution contrary to their nature, their intrinsic drive. Such an existence could not be without travail and could not be blessed, it would be rather like an eternal confinement to a fate like Ixion’s. The alternative would be to say that each celestial sphere performs its own motion for the sake of some crucial component or part it has. This latter case would be analogous to the claim Aristotle made at Physics II 7, 198a21–27 that ‘what something is’, the form, and ‘that for the sake of which’, the goal often coincide. Or, in less general parlance, the case could be analogous to what Aristotle claims about souls, namely that they are not only the first actuality of the living being, they also constitute the goal of the bodily frame of the living being. (De anima II 4, 415b15–21) οὐδεμία φορὰ αὑτῆς ἂν ἕνεκα εἴη οὐδ᾽ ἄλλης φορᾶς, ἀλλὰ τῶν ἄστρων ἕνεκα. εἰ γὰρ ἔσται φορὰ φορᾶς ἕνεκα, καὶ ἐκείνην ἑτέρου δεήσει χάριν εἶναι· ὥστ᾽ ἐπειδὴ οὐχ οἷόν τε εἰς ἄπειρον, τέλος ἔσται πάσης φορᾶς τῶν φερομένων τι θείων σωμάτων κατὰ τὸν οὐρανόν. (1074a26–31) Note that similar to other translations my translation does not stress the terminological difference between motion in general (kinein, kineisthai) used elsewhere, and locomotion (pherein, pheresthai). I shall return to this issue below.

Cases of Celestial Teleology in Metaphysics Λ

263

Clearly, here the individual stars and planets are specified as the beneficiaries of the motions of some larger celestial bodies. In the case of the fixed stars this is the motion of the huge sphere containing all the innumerable multitude of these fixed stars, whereas in the case of the planets this is the ordered system of movements that is performed by the nested spheres carrying this celestial luminary. In all of these cases, every single moving sphere has – in addition to the benefit it derives from its own motion – two goals: one, its mover, which is a goal, but not a beneficiary, the other, the celestial luminary which is carried by the motion of this sphere, as the beneficiary. It is important to stress that this latter aspect is announced in the course of an argument where Aristotle submits that the number of celestial movers is any one of the results he has introduced in the ‘astronomical’ section of the chapter.41 In order to fully appreciate this second case of teleology, we would need to know far more than we do about the relationship of the carrying spheres, the motions provided by them, and the individual celestial body which is set in the innermost one of these spheres. We would need to know, e.g. whether these spheres are several independent entities, each of them pursuing its own goal, or whether they, in a way, constitute a single entity – a planetary system of spheres – with components which pursue their own goals, but nevertheless are calibrated or linked in some sense to the overarching goal of the whole system, that these celestial spheres together should carry the planet. On the second alternative the case would be analogous to different bodily constituents providing for the whole of the organism, or for its form, the soul of the organism concerned, whereas the first alternative would be similar to those cases where in some way one item contributes to the functioning of another one. Even though we are in no position to address this fundamental issue, we can at least try to situate the relationship of this beneficiary, the celestial luminary, to the carrying spheres providing its motions, with regard to other analogous structures which are cases of hypothetical necessity. What seems to be fairly certain is that this relationship between the carrying spheres providing the benefit to the celestial luminary, and the beneficiary, the celestial luminary itself cannot be a case of straightforward hypothetical necessity. Nevertheless, there is no denying that the celestial case is analogous to such 41

In note 1 above I stressed the indispensability of the status of the unmoved movers as goals; if these entities were unrelated to the physical universe, we could not start this project of counting them up. Similarly, without tying the individual instances of pursuing such a goal to the component motions of a celestial luminary, which is the beneficiary of these component motions, and also to the complex one emerging from them, the project would lack necessary empirical footing.

264

István Bodnár

cases where e.g. if one is to build a house, one needs to construct functionally adequate foundations, the requisite walls, and the right kind of roof. Similarly, one could argue that in the case of each of the celestial luminaries, for that luminary to be exactly what it is, it needs to perform the exact number of precisely specified celestial motions. Hence, for the purposes of the complex motion of this celestial luminary, it is exactly those spheres of its planetary spherical system to which it is attached that are necessary.42 To this extent we are very close to an instance of hypothetical necessity, one centred on the celestial luminary itself. Each individual sphere of the system of planetary spheres of this planet performs the motion it does under the influence of its unmoved mover, but then this activity, and the complex result of all these activities, would be subservient to the needs of a specific entity.43 If to this extent this is a case that is analogous to a case of hypothetical necessity, then the order of causation and explanation should also proceed from establishing what is necessary for the beneficiary of the whole complex operation, to seeing to what extent the complex entity providing for this beneficiary can carry out this task. Such considerations would not be foreign to the characterisation of De caelo II 12, where the different planets reach their perfection in different ways, and then to different degrees. Metaphysics Λ, however, does not present this kind of gradation, so it is doubtful that there would be conceptual room for a distinction between the actual contribution of the celestial spheres, the support mechanism so to say, and the needs of the individual luminary, the beneficiary. Therefore, the order of causation and explanation need not be fixed on the side of the beneficiary in the way that a scheme of hypothetical necessity would require. Moreover, allowing even for something analogous to a case of hypothetical necessity here would require that not only the spheres themselves stand in this relationship to the benefit provided for the celestial luminary, but also 42

43

Indeed, if we were allowed to speak about the system of celestial spheres which provide these motions for the celestial luminary as a single, unified entity, one might wish to submit that the relevant parallel case is not one of hypothetical necessity, but rather one where having some part, and the operation of that part, are necessary because these follow directly from what it is to be that single, unified celestial entity. (For this contrast between direct consequences of what it is to be something, and cases of hypothetical necessity see Balme, D. M. 1972: De Partibus Animalum I and De Generatione Animalum I, Oxford, 87 and Cooper, J. M. 1987: Hypothetical necessity and natural teleology, in: A. Gotthelf and J. G. Lennox (eds.): Philosophical issues in Aristotle’s biology, Cambridge, 254.) It is pertinent to see that the two legs of the explanation of this sentence here can be set in stark contrast. Bona fide cases of Aristotelian hypothetical necessity are contrasted to those cases that are necessary unconditionally, without any further restriction (cf. GC II 11, 337b7–29). Here, on the other hand, the operation of the celestial spheres and their movers are necessary in their own right, and even so they are also necessary for the benefit of the celestial luminary attached to the innermost sphere of the celestial system.

Cases of Celestial Teleology in Metaphysics Λ

265

those items which are necessary for the activity of the individual spheres within this planetary spherical system. This comes, however, dangerously close to claiming that the unmoved movers also stand in this relationship, albeit at a further remove, to the celestial luminary. With this we would run the risk of demoting the supra-natural unmoved movers to the status of ancillaries in a complex web of what is necessary for the benefit of the celestial luminary, One could retort to this that in the course of these considerations Aristotle actually submits the general claim that every mover of a locomotion is for the sake of the moving object performing the locomotion.44 There are two considerations, though, that tell against taking this claim to include or involve a reference to the supra-physical celestial movers. One is the terminology used: elsewhere throughout the chapter, each of these movers is referred to as a kinoun, an entity which causes any kind of motion, not necessarily locomotion. True, the celestial motions caused by them will be locomotions. Nevertheless, if Aristotle here were referring to the unmoved movers with the word pheron it would be a single, unprecedented instance.45 The other consideration is the actual course of the argument: Aristotle does not pursue a line of thought that would establish that the unmoved movers are for the sake of the celestial luminaries moved by them. Rather, immediately after announcing his initial claim, he switches to a different formulation, that it is the motions that are for the sake of the celestial luminaries. Presumably this is a legitimate move because what counts as a mover of a celestial locomotion here is nothing but the celestial sphere carrying some other item, and thereby providing for it its own locomotion. Finally, we would need to be clear about whether the celestial luminary is part of the innermost sphere, or a separate entity which is set in this sphere.46 The best example for testing this claim is the sphere of the fixed stars. There we do not need to factor into our account the contribution of further carrying spheres. Fixed stars are carried only by the single sphere they are attached to. The most important passage about the relationship between carrying spheres and their celestial luminary is when Aristotle discusses the motion of the fixed stars at De caelo II 8. There he submits that they do not move with

44 45 46

εἰ γὰρ πᾶν τὸ φέρον τοῦ φερομένου χάριν πέφυκε καὶ φορὰ πᾶσα φερομένου τινός ἐστιν, οὐδεμία φορὰ αὑτῆς ἂν ἕνεκα εἴη οὐδ᾽ ἄλλης φορᾶς, ἀλλὰ τῶν ἄστρων ἕνεκα. (1074a25–28). Cf. the translation of these lines in the Revised Oxford Translation, glossing over this distinction: “for if everything that moves is for the sake of that which it moves”. This query is analogous to the one I have voiced above, whether the spheres of a single planetary system of spheres constitute some kind of unity, or remain distinct entities even though they cooperate in producing a joint activity.

266

István Bodnár

their own motion, rather the motion is provided to them by the sphere. Aristotle uses the word endeisthai to describe the relationship of the sphere and the individual celestial body,47 and this suggests the tying together of two distinct entities. However, this ontological commitment may be mitigated by a further consideration, namely that the celestial luminaries carried along are substances of the same celestial matter. The motion of the sphere cannot possibly be communicated to them as something completely external, but they may also partake of this motion as any other constituent part or section of this sphere does. All in all, even though some of the overriding theoretical concerns may be manifest in well recognisable contours, the vast magnitude of further details eludes us. Among all this uncertainty, however, we can make out from the discussion of these chapters that the celestial domain is a whole of complete, purposeful order, a conviction that will be crucial for the considerations about pervasive cosmic order in the concluding chapter of Metaphysics Λ.

Bibliography Balme, D. M.: Aristotle, De Partibus Animalium I and De Generatione Animalium I, Oxford 1972. Beere, J. B.: Counting the Unmoved Movers: Astronomy and Explanation in Aristotle Metaphysics XII.8, in: Archiv für Geschichte der Philosophie, 85 2003, 1–20. Bodnár, I.: Aristotle’s rewinding spheres: Three options and their difficulties, in: Apeiron, 38 2005, 257–275. Bordt, M.: Why Aristotle’s God is Not the Unmoved Mover, in: Oxford Studies in Ancient Philosophy, 40 2011, 91–109. Broadie, S.: Heavenly Bodies and First Causes, in: G. Anagnostopoulos (ed.), A Companion to Aristotle, Oxford 2009, 230–241. Cooper, J. M.: Hypothetical necessity and natural teleology, in: A. Gotthelf and J. G. Lennox (eds.), Philosophical issues in Aristotle’s biology, Cambridge 1987, 243–274. Fazzo, S.: Heavenly Matter in Aristotle: Metaphysics Lambda 2, in: Phronesis, 58 2013, 160–175.

47

Note that even though in this chapter at 290a18–20 Aristotle contrasts the fixed (endedemenoi) stars to the planets, elsewhere he can describe the planet as being fixed (endedemenon) to the motion it performs (II 12, 292a12–14), and indeed a sphere as being fixed to several other ones (II 12, 293a6f).

Cases of Celestial Teleology in Metaphysics Λ

267

Leunissen, M.: Explanation and Teleology in Aristotle’s Science of Nature, Cambridge 2010. Murshel, A.: The structure and function of Ptolemy’s physical hypotheses of planetary motion, in: Journal for the History of Astronomy 26 1995, 33–61.

The Unity of the World-order According to Metaphysics Λ 10 CHRISTOPH HORN

In Lambda 10, Aristotle describes the universe as a unified and well-ordered whole. As he strongly emphasizes, the order of the cosmos is that of an organized totality. It should not be seen, we are told, as a mere contingent type of match. As Aristotle admits, the world shows some arbitrariness, but this is confined to its inferior parts: the higher an entity, the more its conduct is determined by regularity. The cosmos thereby possesses, he says, “the good and the best”, and this is traced back to the fact that all entities do not exist independently from one another, but in a close interrelation. Moreover, we learn how this interrelatedness has to be understood: namely in the sense that everything is directed towards one center (πρὸς μὲν γὰρ ἓν ἅπαντα συντέτακται: 1075a18–9). The order goes back, according to Aristotle, to a divine entity which is at the center (or at the top) of the universe (of course not in a spatial sense). This deity is compared with the general of an army, and a little later it is equated to a king. Aristotle even ascribes to the entities in the universe the “desire not to be governed badly” (οὐ βούλεται πολιτεύεσθαι κακῶς: 1076a3–4), before he concludes the chapter by quoting the Homeric phrase οὐκ ἀγαθὸν πολυκοιρανίη· εἷς κοίρανος ἔστω (1076a4: “The rule of many is not good; let there be one ruler”, following Iliad II 204) which reinforces the idea of a divine world government. Doubtlessly, the most natural way to deal with our text is to read it as a strong sort of teleology connected with a theological account of the universe: a divine design-argument. Nevertheless, not so many scholars are willing to accept this sort of reading, and very understandably so. Given that Aristotle is generally reluctant in conceding overarching forms of cosmic causality, one fails to see how the universe can be regarded as an organized or structured totality. Moreover, Aristotle does usually not seem to go for the version of teleology that is presupposed in a divine design-argument. And finally, he normally does not allow for a theology that attributes such a strong ordering role to a God showing (quasi-)personal attributes, an idea that is rather typical for a creationist account of the cosmos. In what follows, I would nevertheless like to defend this divine designreading, as I would like to call the teleological-theological interpretation. In

270

Christoph Horn

order to do so, I will have to dispel three major obstacles. First, we need an explanation of how Aristotle can justify the idea of an overarching causality (I.). Then, I must turn to the question of how Aristotle’s account of cosmic order and unity can be reconciled with his general idea of teleology (II.). Finally, I will deal with the question of how his standard view of theology (which seems quite deflationary) can be reconciled with what is said in Lambda 10 (III.). The biggest challenge I will have to face is the ‘coordination problem’.

I. When reading the text closely, one cannot avoid the impression that Lambda 10 should be interpreted in the sense of a divine design-reading. But how can the demanding claims on the divine unity and teleological order of the cosmos be reconstructed on an Aristotelian basis? The philosophical problem addressed in Lambda 10 is formulated in its very first sentence: In which sense does the nature of the universe (ἡ τοῦ ὅλου φύσις) possess the good or the best? In this question, it is clearly presupposed that the nature of the cosmos should rightly be described as having the good or the best. Aristotle apparently considers the world as good, even in a sense as the best possible. Similarly to the Leibnizian idea of le meilleur des mondes possibles Aristotle seems to think that not every individual being is the best, but only the world taken as a whole and its divine ruler.1 We are told that the good (τὸ εὖ) lies both in the order and in the ordering general (ὁ στρατηγός), but is more related to the general since it is he who brings about order (and not the other way round). Aristotle sticks here to the principle that the cause must possess to a higher degree what it gives than the caused. But how can Aristotle describe the universe as the best? Here are the relevant steps (1–7) taken in Lambda 10: (1) Aristotle offers us three ways of understanding how the universe can possess “the good and the best”: either (a) by something separate which exists per se (κεχωρισμένον τι καὶ αὐτὸ καθ᾽ αὑτό,) or (b) by its order (ἢ τὴν τάξιν) or (c) by both (1075a11–3). (2) He opts for (c) by pointing out that both ways of understanding the excellence of the cosmos can simultaneously be true; to illustrate this, he compares the universe to an army (στράτευμα): an army too can be good both by its order (τάξις) and by its general (στρατηγός), even if the 1

Of course, Aristotle neither expresses his thought in the terminology of modal logic, nor does he support the idea of a divine pedagogy causing the moral improvement of those who experience suffering.

The Unity of the World-order According to Metaphysics Λ 10

271

second contributes much more to its goodness than the first (1075a14–5). (3) The reason for the prevalence of the general (and analogously the divine ruler) is that he caused the order of the army (and the order of the universe respectively). But he (the divine ruler) did not arrange everything in an equal way (οὐχ ὁμοίως) – even if (or precisely since) his order comprises also aquatic and winged animals and plants (1075a16–7).2 (4) At any rate, it is not the case that these entities existed unconnectedly or independently from one another; the universe is jointly structured in relation to one single entity. The difference in how the entities participate in the world-order is then explained by a comparison with a household (ὥσπερ ἐν οἰκίᾳ) where the freemen have the least extent of latitude to act arbitrarily, but have to strictly follow the order, while the slaves and the animals contribute less to the common good (μικρὸν τὸ εἰς τὸ κοινόν) since they are acting mostly by chance (τὸ δὲ πολὺ ὅ τι ἔτυχεν) (1075a18–22). (5) The principle (ἀρχή) that constitutes the difference between the higher entities behaving regularly and the lower ones acting arbitrarily is either the nature in general or their respective sort of nature (φύσις). The lower ones are entities that are dissolved (τὸ διακριθῆναι) by necessity; they are followed and replaced by succeeding beings which participate in the whole (1075a22–5). (6) In the text then follows the extended part where Aristotle criticizes and rejects the theories of first principles as they have been advanced by (some of) his predecessors, especially Empedocles, Anaxagoras, Plato, and Speusippus (1075a25–1076a3; on this see below III). (7) The chapter closes with the remark (already mentioned) that the entities in the universe do not want to be governed badly, followed by the quotation of Homer. Two major points should be retained: First, Aristotle does not claim that each entity in the universe, as an individual, is optimal (not even for what it ideally can be), but only the order in general and the divine ruler. And second, he apparently maintains that it is the divine ruler who caused the order. At first glance, passages (1) and (2) seem to leave open whether or not the divine ruler is the cause of the universal order. The alternative as formulated here – that between the goodness coming from a separate entity and the goodness coming from the order – seems to presuppose that the cosmic excellence can either consist in the perfection of its highest entity or in the perfection of its structure, without implying the idea that the structure is caused by the divine ruler. But exactly this type of causality is explicitly maintained in (3): the excellence of the universe is both that of its ruler and that of its structure precisely since it was he who caused it. In passages (3), (4)

2

The qualification ἀλλ᾽ οὐχ ὁμοίως is clearly related to water animals (πλωτά), birds (πτηνά) and plants (φυτά).

272

Christoph Horn

and (5), however, Aristotle differentiates between the manifold ways in which the world-order possesses the good. The underlying idea obviously is that there exist degrees of regularity (or of contingency, to put it the other way round), and the household example is meant to illustrate how different entities can have diverse ways of fulfilling their natures. Fulfilling their natures immediately implies a regularity of their conduct. And the higher an entity is, the more regular and less arbitrary is its behavior. Thus, a well-ordered universe of different layers can emerge. Within the Lambda, this doctrine does not come as a surprise. We already know that the universe is a layered or tiered one: in ch. 1, Aristotle differentiates three types of substances, the terrestrial, the celestial and the immaterial ones and characterizes them as being moved and perishable, moved but imperishable, and unmoved and imperishable. This order of substances clearly forms some sort of ontological hierarchy, characterized by an increasing dignity and regularity, and the divine level plays some causal role (which is yet to be explained) for the rest of the cosmos. Furthermore, the Prime Mover is repeatedly characterized as “principle and cause” (ἀρχὴ καὶ αἰτία: already in ch. 4, 1070b34–5).3 Everything in the universe is included into the order which is directed towards one single entity, the highest deity. But how can degrees of participation in this order be formulated? As David Charles concluded from other Aristotelian writings, he distinguishes between at least four types of matter and hence four types of change or motion (κίνησις).4 Following this idea, we are faced with five different degrees of regularity. For Aristotle, there seem to exist: 1. entities which are unchangeable (and hence imperishable), having no matter and hence no κίνησις, 2. entities which are (otherwise) unchangeable, but in locomotion, 3. entities which are (otherwise) unchangeable, but in locomotion and in qualitative change, 4. entities which are (otherwise) unchangeable, but in locomotion, in qualitative change, and in quantitative change, 5. entities which are changeable in a substantial sense, namely perishable. If Charles’ reading is correct, then the universe is regularly structured, and also in Lambda 2, the concept of regularity is combined with the idea that 3 4

There is an important remark in Metaphysics A 2 where Aristotle claims that God is for all (traditional philosophers) one of the causes and principles (983a8 f.) See Charles (Charles, D. 2000: Aristotle on Matter and Change: a study of Lambda 2, in M. Frede/D. Charles: Aristotle’s Metaphysics Lambda: Symposium Aristotelicum, Oxford, 89–110) with regard to Lambda 2.

The Unity of the World-order According to Metaphysics Λ 10

273

there is some inhomogeneity: Whereas layer 1 is the level taken up by the unmoved mover(s), level 2 is that of the fixed stars, level 3 that of the planets, and level 4 that of sun and moon. On level 5, living beings exist in the sublunary world, namely humans, animals, and plants. We should certainly add a level 6, that of inorganic things, namely the four elements (I will come to that later). The motion of 1 is an ideal circular one, while the motions of 2, 3 and 4 are still regular, but less ideal. The motions of the entities in 5 are even less ideal, except in the sense that perishable things are coming into being and passing away on a regular basis. The motion of elements is that of regularly going upwards or falling down. Even if it is not cogent, much can be said in favour of such a reading of Lambda 2. Now we are told in passage (5) that the principle (ἀρχή) which causes these different ways of participating in the world-order (i.e. the different degrees of regularity) is the nature (φύσις) of all entities. Aristotle adds this remark which can be read in two different ways: If one follows D. Sedley and the manuscripts, there exists one single nature which is the principle of all entities in the universe.5 According to the reading of W. D. Ross and Werner Jaeger, every item has its own nature.6 Following the first reading, the universe too has a φύσις, and not only those entities to which Aristotle ascribes a regular sort of conduct. Following the second one, every entity has a φύσις of its own. This problem will be discussed in the next section (II.). What we saw so far is that, according to Aristotle, the cosmic order is a multi-layered phenomenon: on each level, the entities attain the perfection that it is possible for them to reach. But is it possible to regard all of these different states of perfection as brought about by a comprehensive causality that is present in the universe? As is well known, Aristotle accepts neither a creator nor a Demiurge of the heavenly bodies or the sublunary world. The Prime Mover does not generate lower entities, nor does he actively shape them. Both the heavenly bodies and the sublunary world exist eternally. The stars, the planets, the sun and the moon are everlasting material individuals, whereas entities in the sublunary world obtain a sort of quasieternity by their intergenerational transfer of forms based on reproduction. Even if the Prime Mover might be characterized, to some extent, in personalist terms, it shows no perception of or interest in the material world. What the Prime Mover in fact does, is simply to give a first impulse which 5

6

Sedley: “For that is the kind of principle that nature is for each of them” (τοιαύτη γὰρ ἑκάστου ἀρχὴ αὐτῶν ἡ φύσις ἐστίν) (Sedley, D. 2000: Metaphysics Λ 10, in: M. Frede/D. Charles (eds.), Aristotle’s Metaphysics Lambda. Symposium Aristotelicum, Oxford, 327– 350). Ross: “For the nature of each of them is such a principle” (Jaeger: τοιαύτη γὰρ ἑκάστου ἀρχὴ αὐτῶν ἡ φύσις ἐστίν).

274

Christoph Horn

sets the heavens in motion, and even here strictly speaking only the outermost sphere of heaven is directly affected. Additionally, the assumption of an overarching causality seems to be excluded on the basis of three textual evidences. The first is that Aristotle, in Metaphysics Lambda, famously defends the idea that causality is individually relative. There is “no general man”, but it is “Peleus who begot Achilleus and your father you” (5, 1071a22). For Aristotle, it is the father who generates the son, not the Prime Mover (in Creationist theories of the Judaeo-Christian tradition, it is thus always difficult to separate the contributions to causality which are made by horizontal cause from those advanced by vertically operating causes). Causal relations have to be located within a single species: one might call that the ‘homoeidetic’ concept of causation. A human being begets a human being (ἄνθρωπος γὰρ ἄνθρωπον γεννᾷ: 1070a8.27–8). The term ὁμοειδές – well known from Metaphysics Zeta 7, 1032a24 – is in the Lambda explicitly used in ch. 5, 1071a17. It refutes the idea that universal causes may exist. So whereas Aristotle praises Plato for individualizing natural kinds and for his acceptance of an independent Form for each of them, he blames him for the more basic fact that Plato adopted a general type of causality instead of an individual relative one. Some F is brought about, according to Aristotle, by an entity which possesses F in a synonymous way, not, like a Platonic Form, homonymously. Aristotle thinks that homoeidetic causality is sufficient to explain how individuals are generated. In the same sense, the doctor applies medical knowledge to the patient and the constructor his technical knowledge to the house.7 The second aspect that seems to exclude an overarching causality is the description of elements, causes, and principles in Lambda 4. Aristotle declares that there are, in general, three elements (στοιχεῖα), and four causes and principles (αἰτίαι καὶ ἀρχαί) (1070b25–6). The elements are form, privation, and matter (1070b18–9), and the same are also causes and principles, plus the efficient cause. But their generality as elements, causes, and principles is only by analogy. It is not the case that the same entities, e.g. again Platonic Forms, can be taken as general principles of everything. As Aristotle emphasizes, it would be absurd (ἄτοπον) to assume that elements, causes and principles are numerically identical, but only analogically; between the entities in the world, there is nothing common (οὐδέν ἐστι κοινόν) except the structure of substance and the categories (1070a35–b4). The third difficulty for an overarching type of causality is due to the description of how the Prime Mover sets the cosmos in motion. Aristotle 7

Sexual reproduction is thus described as a sort of τέχνη in which a form is transferred.

The Unity of the World-order According to Metaphysics Λ 10

275

does not possess an equivalent to Plato’s Form of the Good, an entity which is, according to Republic VI, both the ultimate formal, efficient and final cause of all other beings. (On the contrary, he famously criticizes it in Nicomachean Ethics I 2 (4).) The causality of the Prime Mover can only be described in the sense of some chain reaction, regardless of whether one sees final or efficient causality at work.8 His causality goes as follows: by exerting an effect on the outermost celestial sphere, the Prime Mover indirectly causes the movement of all other spheres and finally sets even all sublunary things in motion. Thus, only the sphere of the fixed stars is immediately moved by the impulse of the Prime Mover, and despite the fact that, according to Lambda 8, the motion of one sphere influences the motion of the next one, the role of the Prime Mover is not a prominent one. The Prime Mover gives the first impulse, but this is only one of 56 impulses of the same kind. Each sphere passes the impulse received by one of the movers or by other spheres down to the next one. As we also know from ch. 8, Aristotle describes the heavenly motion as the result of a cooperation of the Prime Mover and 55 additional movers. To be sure, the motion of the sphere of fixed stars caused solely by the Prime Mover is, to a certain degree, an excellent one since it is a constant and ideally circular motion. All other motions are more complex. But this is due to the fact that all other motions are resultants from different constitutive factors. We can clearly see that this model does not describe the motion of the spheres in a subordinative way, but as an addition of partial impulses. Even the expression ‘chain reaction’ which I just used is somewhat exaggerated since the first movement does not cause the other first impulses. On the contrary, the 56 impulses which we see at work here are independent from one another, and they bring about 56 different motions. (It might be that the 55 additional movers are striving for the Prime Mover or that he exerts an influence on them, but Aristotle remains silent on this point.) The ‘chain reaction’ we are confronted with is rather that of the additive effect caused by the fact that each sphere influences the subsequent spheres. We might be tempted to conclude that the Prime Mover does not play a highly privileged role in the scenario advanced by Aristotle. Moreover, he has only a quite indirect effect on motion and change in the sublunary world. Given these difficulties, how can there be a cosmic order as described in Lambda 10? Can we identify any overarching form of causality? To be sure, celestial motions have a strong impact on many changes happening on earth. But they do not seem to play a causative role for all of them in general. Whereas it is only locomotion which is the topic of change and alteration in the celestial sphere, we are confronted, in the sublunary world, with four 8

On the controversy, see Alberto Ross in this volume.

276

Christoph Horn

types of motion: locomotion, qualitative change, quantitative change and substantial change (coming about and passing away). It seems implausible to think that the Prime Mover could be the principle of all of these changes, simply by moving the outermost sphere of the cosmos. But apparently, there must be, for Aristotle, this type of overarching causative force: the Prime Mover is described as the first and infinite source of all actual being. We find three passages in the Lambda which are unambiguously pointing in this direction. The first is the final sentence of ch. 4: “Additionally, there exists something of the first rank of all which moves everything” (ἔτι παρὰ ταῦτα τὸ ὡς πρῶτον πάντων κινοῦν πάντα: 1070b34–5). The second is Aristotle’s claim, in ch.7, that “the heavens and the nature depend of such a principle”, i.e. of the Prime Mover (ἐκ τοιαύτης ἄρα ἀρχῆς ἤρτηται ὁ οὐρανὸς καὶ ἡ φύσις: 1072b13–4). And the third is the passage on the sun and its ecliptic course in ch.5 (παρὰ ταῦτα ὁ ἥλιος καὶ ὁ λοξὸς κύκλος: 1071a15–6) which contains the concession that there is at least a certain cosmic causality in addition to homoeidetic causation according to the synonymy principle. At a closer look, homoeidetic causality is insufficient to explain the emergence of individual beings. For it does not suffice to explain how the everlasting process of causation in an imperishable universe is going on. The fact that celestial bodies are in eternal movement as well as the fact that, in the sublunary world, an eternal process of emerging and deceasing occurs, needs some further explication. To give an account of this requires in addition the assumption of an entity that is in eternal actuality. In the background of the Lambda, there seems to be a consideration known from Physics VIII 10 (266a12–b24). There, Aristotle develops the following argument concerning the infinitely extended motion of the universe: [1] Nothing that has a finite size can cause an infinite motion. [2] In order to cause an infinite motion, it takes an infinite δύναμις, but nothing of a finite size can possess such a infinite δύναμις. [3] The assumptions [1] and [2] imply: The cause of an infinite process cannot be something finite. [4] There is nothing of an infinitely extended size. [5] It follows from [3] and [4] that the cause of an infinitely extended motion cannot be from something with a finitely or infinitely extended size. [6] Hence, the cause must be something with an infinite δύναμις, but without extension. The Unmoved Mover here as well as in the Lambda fills the gap which is left open by homoeidetic causality. Following Plato’s simile of the sun from Republic VI, at least to some extent, Aristotle describes the sun and the Prime Mover in Lambda 5 as parallel in their function as overarching causes, the former of the sublunary world, the latter of the entire cosmos. In the Corpus Aristotelicum, we find additional evidence for the role of the sun as a general cause behind individual causes. In the Meteorology, e.g., we are told that the characteristic cycle

The Unity of the World-order According to Metaphysics Λ 10

277

of the sun causes generation and destruction by its regular approximations and withdrawals (I 9, 346b20): “The efficient and chief and first cause is the circle in which the sun moves. For the sun as it approaches or recedes, obviously causes dissipation and condensation and so gives rise to generation and destruction.” “ἡ μὲν οὖν ὡς κινοῦσα καὶ κυρία καὶ πρώτη τῶν ἀρχῶν ὁ κύκλος ἐστίν, ἐν ᾧ φανερῶς ἡ τοῦ ἡλίου φορὰ διακρίνουσα καὶ συγκρίνουσα τῷ γίγνεσθαι πλησίον ἢ πορρώτερον αἰτία τῆς γενέσεως καὶ τῆς φθορᾶς ἐστι.” (ROTA) What the text apparently means is that the seasons determine the life span of plants and animals. Living beings emerge and perish within an order dominated by the course of the sun. In the De Generatione Animalium IV 10, Aristotle does also bring in the periodical motion of the moon as a cause of the rise and fall of all living beings (777b16–22): “We find, as we might expect, that in all animals the time of gestation and development and the length of life aims at being measured by naturally complete periods. By a natural period I mean, e.g. a day and night, a month, a year, and the greater times measured by these, and also the periods of the moon, that is to say, the full moon and her disappearance and the halves of the times between these, for it is by these that the moon’s orbit fits in with that of the sun [the month being a period common to both].” “Εὐλόγως δὲ πάντων οἱ χρόνοι καὶ τῶν κυήσεων καὶ γενέσεων καὶ τῶν βίων μετρεῖσθαι βούλονται κατὰ φύσιν περιόδοις. Λέγω δὲ περίοδον ἡμέραν καὶ νύκτα καὶ μῆνα καὶ ἐνιαυτὸν καὶ τοὺς χρόνους τοὺς μετρουμένους τούτοις, ἔτι δὲ τὰς τῆς σελήνης περιόδους. εἰσὶ δὲ περίοδοι σελήνης πανσέληνός τε καὶ φθίσις καὶ τῶν μεταζὺ χρόνων αἱ διχοτομίαι· κατὰ γὰρ ταύτας συμβάλλει πρὸς τὸν ἥλιον· ὁ γὰρ μεὶς κοινὴ περίοδός ἐστιν ἀμφοτέρων.” (ROTA) There exists an argumentative structure that is quite parallel to Metaphysics Lambda also in the De Generatione et Corruptione. Aristotle claims in this work, on the one hand, that causation is homoeidetic and, on the other hand, that the everlasting circle of being generated and passing away needs an additional explication. Let us look first at some evidence for individual causation. In chapter I 5, we are told that causation is basically homogeneous or homoeidetic (320b17–21): “Therefore something comes-to-be in an unqualified sense out of another thing, as we have also established elsewhere. And the cause of its comingto-be is either something in actuality which is homogeneous or homoei-

278

Christoph Horn

detic as e.g. fire is the cause of fire or one human being the cause of another human being, or by actuality, for what is hard does not cometo-be by something hard.” “γίγνεται μὲν οὖν ἁπλῶς ἕτερον ἐζ ἑτέρου, ὥσπερ καὶ ἐν ἄλλοις διώρισται, καὶ ὑπό τινος δὲ ἐντελεχείᾳ ὄντος ἢ ὁμογενοῦς ἢ ὁμοειδοῦς (οἷον πῦρ ὑπὸ πυρὸς ἢ ἄνθρωπος ὑπ’ ἀνθρώπου) ἢ ὑπ’ ἐντελεχείας [σκληρὸν γὰρ οὐχ ὑπὸ σκληροῦ γίνεται]·” (ROTA) Although the passage contains a textual problem,9 its fundamental intention is pretty unambiguous: causation must be explained on the basis of homogeneous or homoeidetic factors: fire causes fire, and a man generates a man. But this picture is not sufficient since it cannot account for growth and diminution. Growth and diminution must be understood in terms of the realization of a potentiality. What we therefore need is something that explains the eternal process of actualization. The solution is presented by Aristotle at the end of the De Generatione et Corruptione II in chapter 10. In this text, he starts from the insight that motion and becoming must be seen as everlasting. But nevertheless, he continues, one must provide an explanation for the fact that there is both coming-to-be and passing-away, hence two opposite motions. What can account for this is only the peculiar cycle of the sun that comes closer and moves back; otherwise, it would be impossible to explain the opposite effects of the same motion (336a31–b15): “This explains why it is not the primary motion that causes coming-tobe and passing away, but the motion along the inclined circle: for this motion not only possesses the necessary continuity, but includes a duality of movements as well. For if coming-to-be and passing-away are always to be continuous, there must be some body always being moved (in order that these changes may not fail) and moved with a duality of movements (in order that both changes, not one only, may result). Now the continuity of this movement is caused by the motion of the whole: but the approaching and retreating of the moving body are caused by the inclination. For the consequence of the inclination is that the body becomes alternately remote and near; and since its distance is thus unequal, its movement will be irregular. Therefore, if it generates by approaching and by its proximity, it – this very same body – destroys by retreating and becoming remote: and if it generates by many successive approaches, it also destroys by many successive retirements. For contrary effects de-

9

Cf. Rashed ad locum. (Rashed, M. 2005: Aristotle. De la generation et la corruption, Paris)

The Unity of the World-order According to Metaphysics Λ 10

279

mand contraries as their causes; and the natural processes of passingaway and coming-to-be occupy equal periods of time. Hence, too, the times – i.e. the lives – of the several kinds of living things have a number by which they are distinguished: for there is an Order controlling all things, and every time (i.e. every life) is measured by a period. Not all of them, however, are measured by the same period, but some by a smaller and others by a greater one: for to some of them the period, which is their measure, is a year, while to some it is longer and to others shorter.” “διὸ καὶ οὐχ ἡ πρώτη φορὰ αἰτία ἐστὶ γενέσεως καὶ φθορᾶς, ἀλλ’ ἡ κατὰ τὸν λοξὸν κύκλον· ἐν ταύτῃ γὰρ καὶ τὸ συνεχὲς ἔνεστι καὶ τὸ κινεῖσθαι δύο κινήσεις· ἀνάγκη γάρ, εἴ γε ἀεὶ ἔσται συνεχὴς γένεσις καὶ φθορά, ἀεὶ μέν τι κινεῖσθαι, ἵνα μὴ ἐπιλείπωσιν αὗται αἱ μεταβολαί, δύο δ’, ὅπως μὴ θάτερον συμβαίνῃ μόνον. τῆς μὲν οὖν συνεχείας ἡ τοῦ ὅλου φορὰ αἰτία, τοῦ δὲ προσιέναι καὶ ἁπιέναι ἡ ἔγκλισις. συμβαίνει γὰρ ὁτὲ μὲν πόρρω γίνεσθαι ὁτὲ δ’ ἐγγύς, ἀνίσου δὲ τοῦ διαστήματος ὄντος ἀνώμαλος ἔσται ἡ κίνησις, ὥστ’ εἰ τῷ προσιέναι καὶ ἐγγὺς εἶναι γεννᾷ, τῷ ἀπιέναι ταὐτὸν τοῦτο καὶ πόρρω γίνεσθαι φθείρει, καὶ εἰ τῷ πολλάκις προσελθεῖν γεννᾷ, καὶ τῷ πολλάκις ἀπελθεῖν φθείρει· τῶν γὰρ ἐναυτίων τἀναντία αἴτια, καὶ ἐν ἴσῳ χρόνῳ καὶ ἡ φθορὰ καὶ ἡ γένεσις ἡ κατὰ φύσιν. διὸ καὶ οἱ χρόνοι οἱ βὶοι ἑκάστων ἀριθμὸν ἔχουσι καὶ τούτῳ διορίζονται. πάντων γὰρ ἐστι τάξις, καὶ πᾶς χρόνος καῖ βίος μετρεῖται περιόδῳ, πλὴν οὐ τῇ αὐτῇ πάντες, ἀλλ’ οἱ μὲν ἐλάττονι οἱ δὲ πλείονι· τοῖς μὲν γὰρ ἐνιαυτός, τοῖς δὲ μείζων, τοῖς δὲ ἐλάττων ἡ περίοδός ἐστι, τὸ μέτρον.” (ROTA) In order to explain the emergence and the death of living beings, one needs to have an explicatory basis that is able to account for opposite effects: this precisely is the sun and its ecliptic course. Note that Aristotle uses here almost the same words as in Lambda 5, 1071a15–6: διὸ καὶ οὐχ ἡ πρώτη φορὰ αἰτία ἐστὶ γενέσεως καὶ φθορᾶς, ἀλλ’ ἡ κατὰ τὸν λοξὸν κύκλον· (De gen. et corr. II 10, 336a31–2). Aristotle, in the De Generatione et Corruptione II 10, does not claim that there is a direct sort of overarching causality, but only an indirect one: that of a permanent actualization. It can be characterized as a co-efficient causality rather than as efficient causality. The Prime Mover’s ἐνέργεια is in a parallel way the principle behind the causal efficacy of any cause. The text continues (336b15–9): “And there are facts of observation in manifest agreement with our theories. Thus we see that coming-to-be occurs as the sun approaches and decay as it retreats; and we see that the two processes occupy equal times. For the duration of the natural processes of passing-away and coming-to-be are equal.” “φαίνεται δὲ καὶ κατὰ τὴν αἴσθησιν ὁμολογούμενα τοῖς παρ’ ἡμῶν λόγοις· ὁρῶμεν γὰρ ὅτι προσιόντος μὲν τοῦ ἡλίου γένεσίς ἐστιν, ἀπιόντος δὲ φθίσις,

280

Christoph Horn

καὶ ἐν ἴσῳ χρόνῳ ἑκάτερον· ἴσος γὰρ ὁ χρόνος τῆς φθορᾶς καὶ τῆς γενέσεως τῆς κατὰ φύσιν.” (ROTA) In my view, there can be no doubt that Aristotle acknowledges some sort of general causality: that exerted by the sun (and the moon) and similarly the Prime Mover. They have an impact on all entities in the universe in that they actualize the potential being and so bring about the cycles of life and death. This is a first, important step in favour of a divine design-reading.

II. It is highly persuasive to ascribe to Aristotle some sort of encompassing, overarching causality in the universe, even if it is not like that of Plato’s Forms. Now the question arises how close or far he is, in Lambda 10, from the Platonic model of teleology as advanced in the Phaedo and the Timaeus. On a closer look, it is no exaggeration to say that Aristotle follows in principle the model spelled out in these two dialogues, despite his famous criticism, especially in the De Caelo, of the Platonic world-soul and the idea of selfmotion.10 Aristotle reformulates the concept of a natural teleology without making use of these problematic elements; he shares, however, the core idea that reason (νοῦς) is the cause of everything in nature and that it produces everything according to the best possibility available. In the Phaedo, this approach to explain natural causality is famously described when Socrates tells us the story of his misunderstanding of Anaxagoras whom he believed to have defended the thesis that νοῦς “ordered everything and is the cause of everything” (νοῦς ἐστιν ὁ διακοσμῶν τε καὶ πάντων αἴτιος: 97b9–10). Aristotle twice gives a very similar criticism of the misleading position of Anaxagoras.11 He fully accepts the Platonic position expressed by Socrates that reason brings everything into an optimal order and locates all things to their best (τόν γε νοῦν κοσμοῦντα πάντα κοσμεῖν καὶ ἕκαστον τιθέναι ταύτῃ ὅπῃ ἂν βέλτιστα ἔχῃ: 97c4–5). Thus, the universe is optimal in the double sense of the ordering νοῦς and the order of the structured multitude. For Aristotle as well as for Plato, this implies two directions of teleological explanations within their philosophy of nature: If we know of certain

10

11

This critique is discussed in some detail by Johansen, T. 2009: From Plato’s Timaeus to Aristotle’s De Caelo: The Case of the Missing World Soul, in: C. Wildberg/A. C. Bowen (eds.): New Perspectives on Aristotle’s De Caelo, Brill, 9–28. Anaxagoras’ theory of νοῦς is rejected both in Met. I 3, 984b8–22 and in I 4, 985a18–22.

The Unity of the World-order According to Metaphysics Λ 10

281

empirical facts, we can come to a sufficient explanation by interpreting them as the best possible solution to a given problem; and if we do not know the details of a given situation, we can identify them by using the criterion of how something would be optimally arranged. The first strategy is the predominant one found within the Corpus Aristotelicum. It is not restricted to biological cases (even if these provide the best and most persuasive examples). On Aristotle’s view of nature, we have to differentiate between (a) heavenly spheres and celestial bodies, (b) human beings, (c) animals, (d) plants, and (e) elements. Aristotle adopts the second strategy especially in the context of his astronomy, e.g. in the De caelo II 12. He says that the great distance between us and the celestial bodies makes it difficult to identify something precise about them; if we consider them, he tells us, not as mere bodies, but as ensouled and living beings, then we have the chance to grasp the perfection of their movements (292a14–22): “On these questions it is well that we should seek to increase our understanding, though we have but little to go upon, and are placed at so great a distance from the facts in question. Nevertheless there are certain principles on which if we base our consideration we shall not find this difficulty by any means insoluble. We may object that we have been thinking of the stars as mere bodies, and as units with a serial order indeed but entirely inanimate; but should rather conceive them as enjoying life and action. On this view the facts cease to appear surprising.” “περὶ δὴ τούτων ζητεῖν μὲν καλῶς ἔχει καὶ τὴν ἐπὶ πλεῖον σύνεσιν, καίπερ μικρὰς ἔχοντας ἀφορμὰς καὶ τοσαύτην ἀπόστασιν ἀπέχοντας τῶν περὶ αὐτὰ συμβαινόντων· ὅμως δ’ ἐκ τῶν τοιούτων θεωροῦσιν οὑδὲν ἄλογον ἂν δόξειεν εἶναι τὸ νῦν ἀπορούμενον. ἀλλ’ ἡμεῖς ὡς περὶ σωμάτων αὐτῶν μόνον, καὶ μονάδων τάξιν μὲν ἐχόντων, ἀψύχων δὲ πάμπαν, διανοούμεθα· δεῖ δ’ ὡς μετεχόντων ὑπολαμβάνειν πράξεως καὶ ζωῆς· οὕτω γὰρ οὐθὲν δόξει παράλογον εἶναι τὸ συμβαῖνον.” (ROTA) As soon as one attributes life and action to the stars, our understanding of them is considerably improved. This passage provides a striking example for the teleological explanation in the second sense. Aristotle famously claims that nature does nothing in vain (De Cael. I 4, 271a33, Pol. I 2, 1253a9). This formula basically amounts to the thesis that natural things must be described as purposive or goal-directed. The order of the world is in its best state, Aristotle claims in several passages (e.g. De gen. et corr. II 10, 336b27–8). All natural entities serve a certain function, and they are equipped to do so to the best possible extent. Also in this respect,

282

Christoph Horn

Aristotle follows Plato, namely his ‘function theory’ as we find it in Republic I, 352d–354b. Each entity is characterized as fulfilling a certain function either solely or optimally (better than all the others). The precise way in which Aristotle spells out this idea is that many things in the universe ‘have a nature’, meaning by this nature (φύσις) an internal principle of goal-directedness (especially in Phys. II 1). In Lambda 10, 1075a23 we find the fact that each entity which has such a φύσις possesses it as its principle. A minimalist reading of this version of teleology might maintain that Aristotle confines his considerations to this small point, namely the question of how natural entities can be interpreted as being in their best states. It need not be the case, one might say according to this reading, that serving a function does eo ipso imply acting for an external goal or in the interest of a beneficiary. It would thus be a mistaken sort of question to ask ‘What is the good for the sake of which the universe and its entities are in an optimal condition?’ or ‘Who benefits from the fact that all things are acting for the best?’ The first problem would find a sufficient answer only in some providential history; but this is completely absent in Aristotle since the universe exists eternally, without beginning and end, and can thus have no goal in time. But on the second question, there exists some controversy among the interpreters since D. Sedley (1991 and 2000) advanced an anthropocentric interpretation of Aristotle’s teleology. According to Sedley’s interpretation, human beings are the ultimate end for which the universe is teleologically arranged. This approach possesses the advantage of not pluralizing the beneficiaries for the sake of whom nature operates. As Sedley points out, this does not imply that man is the highest being in the universe; but as e.g. Politics I 8 (1256b10–22) claims, man is the beneficiary of the food chain: Plants exist for the sake of animals, and animals exist for the sake of human beings. With regard to Physics II 8, Sedley argues that the regularity of winter rain provides an indication of its teleological character; regular winter rain makes the crops grow and hence serves the nutrition of men. Sedley adds the argument that human art or craft, according to Aristotle, both imitates nature and perfects it. The link between the two, he contends, lies in the purposiveness of both of them. In order to make this plausible, Sedley assumes that the nature under consideration is not a specific one (e.g. that of water falling down from the sky or that of an animal nourishing human beings), but universal nature (“the nature of the entire eco-system, so to speak”, 2000: 192). I will come back to this in a moment. There are two passages (but only these two) which can be adduced in favour of Sedley’s interpretation. Besides Politics I 8 on the food chain there is Physics II 8. There, Aristotle discusses the relevance of final causality with regard to the popular example of ‘Zeus lets the rain fall’ (ὕει ὁ Ζεὺς). Could

The Unity of the World-order According to Metaphysics Λ 10

283

it be that the rainfall has to be explained by mere necessity (ἐξ ἀνάγκης), not since it is better (ὅτι βέλτιον) that it rains? Aristotle adds in favour of a naturalistic view that nobody would assume that the rain which spoils someone’s crop situated on the threshing-floor has fallen purposively in order to spoil the crop. It is merely a coincident. Isn’t it possible to explain the growth of teeth etc. as an effect of blind necessity? Aristotle rejects this sort of antiteleological naturalism by the following argument (Phys. II 8, 198b34–a8): “Yet it is impossible that this should be the true view. For teeth and all other natural things either invariably or for the most part come about in a given way; but of not one of the results of chance or spontaneity is this true. We do not ascribe to chance or mere coincidence the frequency of rain in winter, but frequent rain in summer we do; nor heat in summer but only if we have it in winter. If then, it is agreed that things are either the result of coincidence or for the sake of something, and these cannot be the result of coincidence or spontaneity, it follows that they must be for the sake of something; and that such things are all due to nature even the champions of the theory which is before us would agree. Therefore action for an end is present in things which come to be and are by nature.” “ἀδύνατον δὲ τοῦτον ἔχειν τὸν τρόπον. ταῦτα μὲν γὰρ καὶ πάντα τὰ φύσει ἢ αἰεὶ οὕτω γίγνεται ἢ ὡς ἐπὶ τὸ πολύ, τῶν δ’ ἀπὸ τύχης καὶ τοῦ αὐτομάτου οὐδέν. οὐ γὰρ ἀπὸ τύχης οὐδ’ ἀπὸ συμπτώματος δοκεῖ ὕειν πολλάκις τοῦ χειμῶνος, ἀλλ’ ἐὰν ὑπὸ κύνα· οὐδὲ καύματα ὑπὸ κύνα, ἀλλ’ ἂν χειμῶνος. εἰ οὖν ἢ ἀπὸ συμπτώματος δοκεῖ ἢ ἕνεκά του εἶναι, εἰ μὴ οἷόν τε ταῦτ’ εἶναι, μήτε ἀπὸ συμπτώματος μήτ’ ἀπὸ ταὐτομάτου, ἕνεκά του ἂν εἴη. ἀλλὰ μὴν φύσει γ’ ἑστὶ τὰ τοιαῦτα πάντα, ὡς κἂν αὐτοὶ φαῖεν οἱ ταῦτα λέγοντες. ἔστιν ἄρα τὸ ἕνεκά του ἐν τοῖς φύσει γιγνομένοις καὶ οὖσιν.” (ROTA) If something happens regularly, it must be arranged by nature, and hence it must be for a purpose. This is the case for rainfall in winter, not in summer. Rainfall in summer is characterized as accidental and contingent, whereas that in winter is purposive. If rainfall is purposive only during the winter season, not everything in the world is arranged in a teleological sense; many facts can also be traced back to matter which is seen by Aristotle as the principle of accidental features and contingency (e.g. in Met. E 2, 1027a8– 14). But even if not all facts in the world require a teleological explanation, a considerable number of them do.12 12

There are highly diverging lines of interpretation of Aristotelian teleology, e.g. in Owens, J. 1968: The Teleology of Nature, in: Monist, 52, 159–173; reprinted J. R. Catan (ed.), Aristotle: The Collected Papers of J. Owens, Albany 1981, 136–147; Nussbaum, M. 1978:

284

Christoph Horn

Nevertheless, Aristotle’s teleology cannot be meant in an anthropocentric sense: Human beings are not that important to him, and the multitude of arrangements in his universe are certainly not entirely purposive for the life of human beings. Furthermore Sedley himself mentions the problem that an animal’s striving for self-fulfillment and its tendency for self-transcendence by reproduction (thus gaining some sort of immortality) are not easy to reconcile with being part of a comprehensive purpose. In Aristotle’s teleology, there exists no principle that generally determines lower entities to be at the service of higher ones; if there in fact are such constellations of unilateral or mutual benefit, this is contingent (see on this Scharle 2008a and 2008b). Hence the ‘business’ of teleology is rather to improve any given entity or species, not to improve the living conditions of one species alone. In my view, the decisive passage in Physics II 8 has a quite different focus: it mirrors not an anthropocentric, but a perfectionist account of teleology. According to a perfectionist reading, one can defend a teleological theory without assuming a beneficiary of the cosmic arrangement. A perfectionist teleology is simply based on observations about ideal arrangements, that e.g. the front teeth of many animals are sharp whereas their molars are broad (as we read it in Phys. II 8). This fact is a non-coincidental feature of their natural endowment. Since we notice that animals are in all cases (or at least mostly) endowed with an appropriate set of teeth, we can conclude that, in such a case, this is due to their respective natures, which equip living creatures with all sorts of instruments necessary or useful for their survival (as already spelled out in the myth of Plato’s Protagoras). Hence, according to the perfectionist model of teleology, the teleological character of nature lies in its tendency to distribute advantages to all beings; nature tends to enable them to have a constant and successful existence. Even if the individuals of a given species may lead a suboptimal life due to their concrete defective equipment (or the insufficient realization of their endowment) and are killed or destroyed, the species themselves remain constantly the same, on the basis Aristotle on Teleological Explanation, in: M. Nussbaum, Aristotle’s De motu animalium, Princeton 59–99; Balme, D., 1987, “Teleology and Necessity,” in A. Gotthelf and J. G. Lennox (eds.), Philosophical Issues in Aristotle Biology, Cambridge, 275–286; Sauvé Meyer, S. 1992: Aristotle, Teleology, and Reduction, in: Philosophical Review, 101, 791–825; reprinted in T. Irwin (ed.), Classical Philosophy. Collected Papers, New York–London, 1995, 81–116; Furley, D. J. 1999: What Kind of Cause is Aristotle’ Final Cause? In: M. Frede/ G. Striker (eds.), Rationality in Greek Thought, Oxford, 59–79 and Johnson, M.R. 2005: Aristotle on Teleology, Oxford. I am mainly following the reading advanced by Scharle, M. 2008a: Elemental Teleology in Aristotle’s Physics II 8, in: Oxford Studies in Ancient Philosophy, 34, 147–184 and 2008b: The Role of Material and Efficient Causes in Aristotle’s Natural Teleology, in: Apeiron 41, 27–46 and Leunissen, M. 2010: Explanation and Teleology in Aristotle’s Science of Nature, Cambridge.

The Unity of the World-order According to Metaphysics Λ 10

285

of sexual reproduction and the self-protection conferred on them by being optimally adapted to their habitats. What sort of features then are supposed, in Lambda 10, to make the universe good and even the best? According to our text, I think we have three possible candidates: The first is the regularity which is found in the universe: namely both the regular behaviour of its higher parts which are similar to the free persons in the household, and the weaker regularity of the lower parts which consists in a regular process of coming-into-being and passing away. The second is its hierarchical order, especially the fact that the sublunary world is governed by the motions of the celestial bodies in a similar way as slaves and animals are ruled by free persons. And the third is the fact that it shows a structure of coordination and of interrelatedness. It is this third aspect which is, so far, completely unclear. If we can take it for granted that the Aristotelian universe is teleological in the perfectionist, not in the anthropological sense, we must still give an explanation of precisely how Aristotle’s comparison of the cosmos with an army works. And Lambda 10 is not isolated; the idea of a something which has brought „the universe into this order“ (εἰς ταύτην τὴν τάξιν τὸ πᾶν: Phys. II 4, 196a24–35), can also be found elsewhere. What seems to be lacking so far in our reading is an explanation of the order in the universe, the coordination problem, as one might call it. If, in an apartment, every detail is in an optimal condition, this does not yet imply that everything is optimally coordinated. Here, Sedley’s interpretation comes back with a second important aspect: The idea of a universal nature, as it is introduced at the beginning of Lambda 10 in 1075a11 and then reappears in 1075a22–3. I already mentioned the fact that the second text can be read in two different ways: either in the sense of one single nature of the entire universe (Sedley’s interpretation) or in the sense of one nature working in each of the natural entities separately. As I. Bodnár (2005) has shown, both readings are grammatically possible, but a certain linguistic advantage lies on the side of the second view.13 But on the other hand, the Aristotelian version of teleology, then, would be unable to account for the coordination problem since, even if, to take again the apartment example, all items and processes are working regularly at the will of the resident, there can be a serious lack of coordination. Bodnár instead proposes to return to the traditional reading according to which all entities of the universe strive for the Prime Mover by the regularity of their motions and the everlastingness of their reproduction. As Bodnár

13

As Bodnár (Bodnár, I. 2005: Teleology across natures, in: Rhizai 2, 9–29), 20 points out, it would seem somewhat strange that, on Sedley’s reading, nature has the good instead of being good.

286

Christoph Horn

admits, there would remain a certain gap then in Aristotle’s teleology which can be filled by a Platonic theory as well as by a Stoic one, but not on an Aristotelian account. This, however, would be a very weak reading of Lambda 10. It does not really do justice to Aristotle’s examples of the army and the household. For the decisive advantage of the organizational structure of an army (and a household) does not simply lie in the fact that each member of them behaves on a regular basis. Additionally, we should find an explanation for their coordination.

III. Does the imitation of the Prime Mover, practiced by each entity in the universe, suffice as an answer to the coordination problem, as Bodnár claims? This seems implausible since in Lambda 10, as we saw, Aristotle emphasizes the interrelatedness of all things in the universe. A world consisting of isolated beings individually (or collectively within a species) striving for the Prime Mover would not bring about the crucial feature of cooperation. An additional objection arises from the observation that, in the second half of Lambda 10, Speusippus is attacked precisely for his departmentalization of reality, resulting from his claim that mathematical numbers are principles. In the more detailed refutation of Speusippus’ standpoint in Metaphysics N 2–3 it becomes clear that Aristotle criticizes him for having disconnected the reality into different layers of being – namely numbers, spatial magnitude, soul, sensible things – which are supposed to exist independently from one another. In the same vein, Aristotle’s criticism of his other adversaries – Plato, Empedocles, and Anaxagoras – regards their mistaken dualism; we can infer from this that Aristotle’s own position must be meant to imply some sort of cosmic monism. The coherence contained in Aristotle’s worldview must be a much more unified one. But in what sense can this be meant? In order to deal with the cooperation problem, we should first remind ourselves of some fundamental facts about Aristotle’s theology. It is crucial to see what kind of imitation his philosophical theology permits and what kind it excludes. As is widely (if not generally) accepted, we find in Aristotle no creationism, no doctrine of individual immortality for humans, no salvation or redemption; furthermore, the God does not take care for the sublunary world because he is immutable and self-directed (Lambda 9, 1074b15– 35). Divine providence, if it exists, is at least not a decisive phenomenon of universal history since contingency, according to Aristotle, is real in the sublunary world and cannot be overcome. In his discussion of good fortune as brought about by the God in E. E. VIII, Aristotle does not completely disregard it, but gives us also no clear affirmation of its existence. The Aristotelian

The Unity of the World-order According to Metaphysics Λ 10

287

God, however, possesses certain personal features such as thinking and leading the most pleasurable life (Met. Lamba 7, 1072b14–30). According to Aristotle, there can be friendship between God and human beings despite the strong asymmetry between these friends (E. E. VII 3, 1238b18–30 and VII 12, 1245b14–9). Furthermore, we learn that God loves most the philosopher, apparently because of his intellectual activity (E. N. X 8, 1179a22–32; on this see Broadie 2003; cf. E. E. VII 14, 1247a27–1248b6). Here, we find the idea of a human imitation of God. It is plausible to say that Aristotle shares the Platonic advice that humans should ‘become like God’ and so participate in immortality (ἀθανατίζειν: Sedley 1999). Doubtlessly, the traditional reading is correct: Aristotle’s dominant way of characterizing the relationship between God and the world is that of divine perfection and the universe imitating it. To imitate the Prime Mover can be realized (a) by perfect circular motion, as in the case of the fixed stars (Phys. VIII 9, 265a28–b9; Met. XII 7, 1072b1–2), (b) by sexual reproduction, as in the case of animals and plants (De an. II 4, 415a25–b7), (c) by contemplation, as in the case of human beings (E. N. X 7, 1177b26–1178a8, X 8, 1178b7–32, E. E. I 7, 1217a26–9; E. E. VIII 3, 1249b15–614), (d) by things happening on a regular basis (as in the case of winterrain), and (e) by an uninterrupted activity, as in the case of the elements; Aristotle declares in Metaphysics Theta 8, 1050b28–30: “Imperishable things are imitated by those that are involved in change, e.g. earth and fire. For these also are ever active; for they have their movement of themselves and in themselves.” “μιμεῖται δὲ τὰ ἄφθαρτα καὶ τὰ ἐν μεταβολῇ ὄντα, οἷον γῆ καὶ πῦρ. καὶ γὰρ ταῦτα ἀεὶ ἐνεργεῖ· καθ’ αὑτὰ γὰρ καὶ ἐν αὑτοῖς ἔχει τὴν κίνησιν.” (ROTA) The Aristotelian God is the final good of universal strive and desire; the Prime Mover is to be understood as „a kind of metaphysical magnet“, as Charles Kahn has put it (1985, 184). Aristotle speaks of strive, love, desire, and imitation. In addition to the language of appetitive behavior, we can also identify the language of efficient causality. God is explicitly described as active or operative (ποιητικός: Lambda 6, 1071b12–13). Given these basic facts about Aristotelian theology, what might be an adequate answer to the coordination problem? Coordination can receive its value either from the fact that coordinated entities are producing an (external or internal) good or from the perfection of the structure under consideration. As far as I see, one can hence distinguish between four possible solutions, 14

See also E. E. 1245b14–9 and Pol. VII 13, 1331b25

288

Christoph Horn

namely the assumption of (I) a common external good, (II) a common good of mutual benefit, (III) a common internal good, and (IV) a common good of an overarching perfection. (I) What Aristotle clearly expresses, in Lambda 10, is that there exists a common good to which the different, but coordinated entities in the universe contribute in various ways, as the members of an army or a household do. This seems to favour a reading according to which the universe strives for some external good, while the entities within the cosmos contribute to it following a certain division of labour. Can this be the correct reading? Certainly not. The Prime Mover has no end or goal for the sake of which he could act at all, given that he is mere ἐνέργεια without potentiality and that he already leads a perfect, eternal and happy life. In the same vein, the idea that the universe is an instrument for an external end is obviously absurd; there is nothing outside the universe. (II) The common good of the universe might be seen in some sort of benefit resulting from the interaction between the different entities in the world, as in the case of an eco-system or habitat, constituted by the elements, plants, animals, and humans living in it. But Aristotle develops no clear idea of interaction and complex living conditions. So when he says e.g. that men possess such valuable tools as hands because they are the most intelligent creatures and can hence make the best use of this item (De part. an. IV 10, 687a8–18), all he wants to express is that hands are exactly suited to the purposes of their human owners. He does not mean that hands are a divine gift to men, let alone that the God arranged this with regard to the possession of other endowments of other biological species. When Aristotle discusses the advantage of the fact that the front teeth of many species are sharp whereas their molars are broad in order to optimally fit the different purposes for which the teeth are needed in Physics II 8, he does not bring in the idea of a divine coordination. It is an idea found in Plato’s Protagoras, Phaedo and Timaeus to describe the characteristics of a species in the sense of a divine distribution. For Aristotle, it would indeed be strange to interpret the natural endowment of animals as God’s will, since then, one could not understand why God privileged a certain species up to this or that degree, but no more and no less. (III) Can the common good be an internal one? The benefit which is brought about by the members of a household if they are acting jointly can be e.g. wealth, i.e. something which is of use for the members themselves. Likewise, we could assume that the universe is good and valuable since it is made for the benefit or, say, for the happiness of its inhabitants. This, however, would not be a κοινὸν ἀγαθόν, except in the sense of some collective advantage in which all items of the universe partake distributively. But benefit or happiness are irreducibly individual. So the assumption that Aristotle

The Unity of the World-order According to Metaphysics Λ 10

289

might mean an internal good also seems implausible. Given additionally that Aristotle ascribes happiness to a very small number of people – he is even reluctant to attribute the capability to become happy – it is unlikely that the world can have happiness as its internal end, especially if we can exclude the anthropocentric reading of Lambda 10. (IV) There is one final option for understanding the idea of a common good of the universe, generated by the cooperation of all entities within it. The best way to make sense of this is, to my mind, to point out that the cosmos, as seen from an imaginary external standpoint, shows some sort of perfection. This perfection might be spelled out in terms of a principle of plenitude, i.e. in the sense that the universe contains all valuable entities which are ontologically possible; but Aristotle does not clearly subscribe to such a principle. Its perfection is rather due to the fact that it is ruled and thus coherently organized. And this is exactly what Aristotle expresses in Lambda 10, especially by the concluding quotation of Homer. There can be no doubt that Aristotle conceives of the ‘ruler’ of the universe in terms of an organizer. In Politics I 5 we learn that the very fact that something is organized implies the idea that there is a ruler of the system (1254a28–33): “For in all things which form a composite whole and which are made up of parts, whether continuous or discrete, a distinction between the ruling and the subject element comes to fight. Such a duality exists in living creatures, but not in them only; it originates in the constitution of the universe; even in things which have no life there is a ruling principle, as in a musical mode.” “ὅσα γὰρ ἐκ πλειόνων συνέστηκε καὶ γίνεται ἕν τι κοινόν, εἴτε ἐκ συνεχῶν εἴτε ἐκ διῃρημένων, ἐν ἅπασιν ἐμφαίνεται τὸ ἄρχον καὶ τὸ ἀρχόμενον, καὶ τοῦτο ἐκ τῆς ἁπάσης φύσεως ἐνυπάρχει τοῖς ἐμψύχοις· καὶ γὰρ ἐν τοῖς μὴ μετέχουσι ζωῆς ἔστι τις ἀρχή, οἷον ἁρμονίας.” Aristotle emphasizes the naturalness and ubiquity of order and traces it back to the order of the universe. Order presupposes a ruling principle. In Lambda 10, the figure that is explained through comparison to an army general and a king is the Prime Mover. But then, the God must be in a sense the active organizer and designer of the universe. Are we entitled to consider Aristotle’s Prime Mover as such a figure? Of course, Aristotle does not accept, by contrast with Plato, the idea of a universal δημιουργός. But the comparison of the Prime Mover to an army general clearly implies that the general is not due to the order; rather, the order is due to the general (οὐ γὰρ οὗτος διὰ τὴν τάξιν ἀλλ᾽ ἐκείνη διὰ τοῦτόν ἐστιν: 1075a15). This inference is supported by the remark at the end of ch. 10 that the beings do not want to be ruled in a bad way. Even if the Prime Mover is not conceived of as an intentionally and deliberatively planning subject, he can bring about a coordinative effect

290

Christoph Horn

among the beings in the universe. The Prime Mover is in pure actuality and, as we saw, he is the source of actuality of all things which pass from potentiality to actuality. The idea of actualization is sufficient for bringing about a coherent order of the universe, as we learn from the De generatione et corruptione II 10 (336b25–34): “Coming-to-be and passing-away will, as we have said, always be continuous, and will never fail owing to the cause we stated. And this continuity arises with a good sense. For in all things, as we affirm, nature always strives after the better. Now being […] is better than not-being; but not all things can possess being, since they are too far removed from the principle. The God therefore adopted the remaining alternative and fulfilled the perfection of the universe by making coming-to-be uninterrupted; for the greatest possible coherence would thus be secured to existence because that coming-to-be should itself come-to-be perpetually is the closest approximation to eternal being.” “ἀεὶ δ’, ὥσπερ εἴρηται, συνεχὴς ἔσται ἡ γένεσις καὶ ἡ φθορά (καὶ οὐδέποτε ὑπολείψει δι’ ἣν εἴπομεν αἰτίαν), τοῦτο δ’ εὐλόγως συμβέβηκεν. ἐπεὶ γὰρ ἐν ἅπασιν ἀεὶ τοῦ βελτίονος ὀρέγεσθαί φαμεν τὴν φύσιν, βέλτιον δὲ τὸ εἶναι ἢ τὸ μὴ εἶναι (τὸ δ’ εἶναι ποσαχῶς λέγομεν, ἐν ἄλλοις εἴρηται), τοῦτο δ’ ἐν ἅπασιν ἀδύνατον ὑπάρχειν διὰ τὸ πόρρω τῆς ἀρχῆς ἀφίστασθαι, τῷ λειπομένῳ τρόπῳ συνεπλήρωσε τὸ ὅλον ὁ θεός, ἐνδελεχῆ ποιήσας τὴν γένεσιν· οὕτω γὰρ ἂν μάλιστα συνείροιτο τὸ εἶναι διὰ τὸ ἐγγύτατα εἶναι τῆς οὐσίας τὸ γίνεσθαι ἀεὶ καὶ τὴν γένεσιν.” (ROTA) The divine contribution to the perfection of the universe is precisely its permanent and everlasting actualization. Given what we said in section I, the God is the principle behind the causality exerted by the different beings. At this point, we should return to Sedley’s second proposal: that of a unified nature of the universe. I think that he is right on this, and we can now see why. Nature can be described by Aristotle as the ordering effect of the divine rule. When in the De Caelo Aristotle uses the formula ‘Nature does nothing in vain’, he does so by saying: Ὁ δὲ θεὸς καὶ ἡ φύσις οὐδὲν μάτην ποιοῦσιν (De Cael. I 4, 271a33). The parallelism between God and Nature need not be epexegetic, having the meaning ‘God, i.e. Nature’. But would also be implausible that the phrase is meant simply in the sense of ‘Paul and Susan do nothing in vain’, presupposing that these two persons act independently from one another. We should read it in the sense that all instantiations of order that we find in the universe can ultimately be traced back to the divine constitution of the universe. Order and ‘the best’ can be realized, as we know it from Lambda 10, in layers or degrees, as Aristotle spells out in the De Caelo II 12 (292b13–19):

The Unity of the World-order According to Metaphysics Λ 10

291

“Thus, taking health as an end, there will be one thing that always possesses health, others that attain it, one by reducing flesh, another by running and thus reducing flesh, another by taking steps to enable himself to run, thus further increasing the number of movements, while another cannot attain health itself, but only running or reduction of flesh, so that one or other of these is for such a being the end. For while it is clearly best for any being to attain the real end, yet, if that cannot be, the nearer it is to the best the better will be its state.” “οἷον εἰ ὑγίεια τέλος, τὸ μὲν δὴ ἀεὶ ὑγιαίνει, τὸ δ’ ἰσχνανθέν, τὸ δὲ δραμὸν καὶ ἰσχνανθέν, τὸ δὲ καὶ ἄλλο τι πρᾶξαν τοῦ δραμεῖν ἕνεκα, ὥστε πλείους αἱ κινήσεις· ἕτερον δ’ ἀδυνατεῖ πρὸς τὸ ὑγιᾶναι ἐλθεῖν, ἀλλὰ πρὸς τὸ δραμεῖν μόνον ἢ ἰσχνανθῆναι, καὶ τούτων θάτερον τέλος αὐτοῖς. Μάλιστα μὲν γὰρ ἐκείνου τυχεῖν ἄριστον πᾶσι τοῦ τέλους· εἰ δὲ μή, ἀεὶ ἄμεινόν ἐστιν ὅσῳ ἂν ἐγγύτερον ᾖ τοῦ ἀρίστου.” (ROTA) This text gives us an impression of how Aristotle conceives of a multi-layered universe which is teleologically organized and is oriented towards one single center. There are various degrees of proximity to the center, as the example of health illustrates. I think that the solution to the cooperation problem should be formulated as follows: The universe is a gradually ordered totality and thus it is good; to its goodness in being ordered, the inferior things contribute only a little. It is good by its perfection, not by the goods or advantages it brings about externally or internally. This reading is attractive since it explains why regular behaviour is seen as beneficial for everything; it is simply the idea of a perfect order which makes the universe valuable. This line of thought is developed in favour of a perfectionist reading in the sense of a divine designargument. The common good which is under consideration is that of perfection, spelled out as regularity, hierarchy, and everlastingness.

Bibliography Balme, D.: Teleology and Necessity, in: A. Gotthelf and J. G. Lennox (eds.), Philosophical Issues in Aristotle Biology, Cambridge 1987, 275–286. Bodnár, I.: Teleology Across Natures, in: Rhizai 2 (2005), 9–29. Bolton, R.: The Material Cause: Matter and Explanation in Aristotle’s Natural Science, in: W. Kullmann and S. Föllinger (eds.), Aristotelische Biologie, Stuttgart 1997, 97–126. Byrne, C.: Aristotle on Physical Necessity and the Limits of Teleological Explanation, in: Apeiron, 35 2002, 20–46.

292

Christoph Horn

Cameron, R.: The Ontology of Aristotle’s Final Cause, in: Apeiron, 35 2002, 153–179. Charles, D.: Teleological Causation in the Physics, in: L. Judson (ed.), Aristotle’s Physics: A Collection of Essays, Oxford: Oxford University Press 1991, pp. 101–128. Charles, D.: Aristotle on Matter and Change: a study of Lambda 2, in M. Frede/D. Charles: Aristotle’s Metaphysics Lambda: Symposium Aristotelicum, Oxford 2000, 89–110. Code, A.: The Priority of Final Causes over Efficient Causes in Aristotle’s PA, in: W. Kullmann/S. Föllinger (eds.), Aristotelische Biologie. Intentionen, Methoden, Ergebnisse, Stuttgart 1997, 127–143. Cooper, J. M.: Aristotle on Natural Teleology, in: M. Schofield and M. Nussbaum (eds.), Language and Logos, Cambridge,197–222; reprinted in J. M. Cooper, Knowledge, Nature and the Good: Essays on Ancient Philosophy, Princeton 2004, 107–129. Cooper, J. M.: Hypothetical Necessity and Natural Teleology, in: A. Gotthelf and J. G. Lennox (eds.), Philosophical Issues in Aristotle Biology, Cambridge 1987, 243–274. Falcon, A.: Aristotle and the Science of Nature: Unity Without Uniformity, Cambridge 2005. Furley, D. J.: What Kind of Cause is Aristotle’ Final Cause? In: M. Frede/G. Striker (eds.), Rationality in Greek Thought, Oxford 1999, 59–79. Gotthelf, A.: The Place of the Good in Aristotle’s Teleology, in: J. J. Cleary/ D. C. Shartin (eds.), Proceedings of the Boston Colloquium in Ancient Philosophy, 4 1988, 113–39. Gotthelf, A.: Understanding Aristotle’s Teleology, in: R. Hassing (ed.), Final Causality in Nature and Human Affairs, Washington DC 1997, 71–82. Johansen, T.: From Plato’s Timaeus to Aristotle’s De Caelo: The Case of the Missing World Soul, in: C. Wildberg/A. C. Bowen (eds.): New Perspectives on Aristotle’s De Caelo, Brill 2009, 9–28. Johnson, M. R.: Aristotle on Teleology, Oxford 2005. Judson, L.: Aristotelian Teleology, in: Oxford Studies in Ancient Philosophy, 29 2005, 341–365. Kahn, C. H.: The Place of the Prime Mover in Aristotle’s Teleology, in: A. Gotthelf (ed.), Aristotle on Nature and Living Things, Bristol–Pittsburgh 1985, 183–205. Lewis, F.: Teleology and Material / Efficient Causes in Aristotle, in: Pacific Philosophical Quarterly, 69 1988, 54–98. Leunissen, M.: Explanation and Teleology in Aristotle’s Science of Nature, Cambridge 2010. Leunissen, M./Gotthelf, A.: What’s Teleology Got to Do with It? A Reinterpretation of Aristotle’s Generation of Animals, Phronesis, 55 2010, 325– 356.

The Unity of the World-order According to Metaphysics Λ 10

293

Nussbaum, M.: Aristotle on Teleological Explanation, in: M. Nussbaum, Aristotle’s De motu animalium, Princeton 1978, 59–99. Owens, J.: The Teleology of Nature, in: Monist, 52, 159–173; reprinted J. R. Catan (ed.), Aristotle: The Collected Papers of J. Owens, Albany 1981, 136–147. Pellegrin, P.: Les ruses de la nature et l’eternité du mouvement. Encore quelques remarques sur la finalité chez Aristote, in: M. Canto-Sperber/P. Pellegrin (eds.), Le Style de la pensée. Recueil des textes en hommage à Jacques Brunschwig, Paris 2002, 296–323. Ross, A.: Dios, eternidad y movimiento, Pamplona 2007. Sauvé Meyer, S.: Aristotle, Teleology, and Reduction, in: Philosophical Review, 101, 791–825; reprinted in T. Irwin (ed.), Classical Philosophy. Collected Papers, New York–London 1992, 1995, 81–116. Scharle, M.: Elemental Teleology in Aristotle’s Physics II 8, in: Oxford Studies in Ancient Philosophy, 34 2008a, 147–184. Scharle, M.: The Role of Material and Efficient Causes in Aristotle’s Natural Teleology, in: Apeiron 41 2008b, 27–46. Sedley, D.: Is Aristotle’s Teleology Anthropocentric? In: Phronesis 36 1991, 179–96. Sedley, D.: Metaphysics Λ 10, in: M. Frede/D. Charles (eds.), Aristotle’s Metaphysics Lambda. Symposium Aristotelicum, Oxford 2000, 327– 350.

Indices 1. Index locorum Alexander of Aphrodisias Commentary on Aristotle’s Metaphysics (following Hayduck) 136,13‒15 31 (81) 137,12‒15 31 (84) 138,2‒3 31 (82) 138,6‒9 31 (83) 167,14‒20 29 (74) 169,21‒26 31 (83) 169,26‒170,4 31 (85) 246,13–24 34 (92) 352,10–16 29 (77) De anima 88.24 f. 188 (16) De providentia 163.23–165.3 Zonta 157.14–18 185 (9)

185 (9)

On the principles of the Universe (following Genequand) § 64 201 Quaest. I.1 188, 201 I.1 passim 188 I.1, 3.20 188 I.1, 3.21 188 I.1, 3.23 188 I.1, 4.10 188 I.1, 4.11 188 I.1, 4.14 188 I.1, 4.21 188 I.25 8, 188, 190, 201 I.25, 39.9–12 186 I.25, 39.10 f. 187 (13) I.25, 39.11 188, 190, 191 I.25, 39.14 187 I.25, 39.18 187 I.25, 39.20 188 (17), 191 I.25, 39.30 187

XVIII 210 (15) XIX 210 (15) Aristotle An. Post. I 2, 71b33–72a5 162 (29) I 31, 88a2–8 167 (48) II 12 132 II 12 133 (25) II 19 167 (48) Cat. 1a6–12 123 (11) 1a25 72 2a11–13 72 2b6 140 3b10 140 3b39–4a2 177 4a10–11 141 4a21–b18 141 5 140 5, 3b10–14 151 5, 3b15 150 7 239 De an. I 1, 403a8–10 167 I 3, 406a7 167 (45) I 3, 406a11 167 (45) I 3, 406b15 167 (45) I 4, 408b29 169 (52) II 1 147, 186 (11) II 1, 412a6–9 127 (15) II 1, 412a19–22 185 (10) II 1, 413a3–7 134 (27) II 2, 413a22 162 II 4 216, 218, 223, 224 II 4, 415a20–22 174 (72) II 4, 415a25–b7 287 II 4, 415a26–b7 161 (22), 216 (49), 237 (22) II 4, 415a28–b7 255 (25)

296

Indices

II 4, 415a29 f. 161 (21) II 4, 415b12–13 224 (77) II 4, 415b15–21 262 (39) II 4, 415b18–19 224 (78) II 4, 415b21–22 224 (79) II 4, 415b23–27 224 (80) II 5, 417a25 173 (70) II 5, 417a26–28 170 II 5, 417a28 172 (64) II 5, 417a32 f. 173 (70) II 5, 417b5 173 (70) II 5, 417b6 f. 171 II 5, 417b23 f. 170 II 6 167 (44) III 3, 427b14–16 167 (46), 171 (61) III 3, 427b25 171 (61) III 4 168, 170 III 4, 429b5–10 166 (42), 171 III 4, 429a15 f. 173 III 4, 429a15–18 168 III 4, 429a21–24 170 III 4, 429a21–25 173 III 4, 429b30–430a2 170 III 4, 430a3 f. 170, 241 III 4, 430a6 f. 174 (72) III 5 7, 188, 190, 191, 192, 193 III 5, 430a14 f. 170 (56) III 5, 430a14–25 191 III 5, 430a18 173 (71), 190, 191, 192, 193, 194 III 5, 430a15–17 174 (72) III 7, 431a14–17 167 (46) III 7, 431a16 f. 167 III 7, 431b2–19 167 (46) III 7, 431b16 f. 170 (57) III 8, 431b20–432a3 170 (55) III 8, 432a3–5 174 (72) III 8, 432a3–6 167 (47) III 8, 432a7–9 167 (49) III 9, 432b5–7 160 (13) III 10, 433a24–30 160 (13) De cael. I 4, 271a33 281, 290 I 9, 278b23 214 (40) I 9, 279a24 214 (40) I 9, 279a32 169 (52) II 1, 284a7 214 (40) II 1, 284a27–35 262 (38) II 2, 285a29 ff. 210 (14) II 8 265

II II II II II II II II II II II II

8, 290a18–20 266 (47) 12 9, 256, 259, 261, 264, 281 12, 292a12–14 266 (47) 12, 292a14–22 281 12, 292a20 ff. 210 (14) 12, 292b1 ff. 210 (14) 12, 292b4–7 257 (28) 12, 292b10 236 (20) 12, 292b13–25 290 12, 292b19–25 258 (30) 12, 293a4–8 260 (32) 12, 293a6 f. 266 (47)

De gen. an. II 1, 731b24–732a11 161 (22), 237 (22) IV 10, 777b16–22 277 De gen. et corr. I 2, 316a11 74 I 5, 320b17–21 277 I 7, 324b13–14 211 (23) II 9 135 II 9, 335b7–24 136 (31) II 10 279 II 10, 336a31–32 279 II 10, 336a31–b15 278 II 10, 336b15–19 279 II 10, 336b25–34 290 II 10, 336b25–337a15 237 (22) II 10, 336b27–28 281 II 10, 336b27–337a7 217 (51) II 10, 337a1–7 161 (22) II 11 218 II 11, 337b7–29 264 (43) De mem. I 449b30–450a9

167 (49)

De part. an. I–II 81 I 5, 644b22–645a6 235 I 5, 644b25 f. 169 (52) IV 10, 687a8–18 288 E. E. I 7, 1217a26–29 287 I 8, 1217b21 74 VII 3, 1238b18–30 287 VII 12, 1245a7–10 241 (28) VII 12, 1245b14–19 287 VII 12, 1245b16–19 164 (34), 178

1. Index locorum VII 12, 1245b17–19 175 VII 14, 1247a27–1248b6 287 VIII 286 VIII 3, 1248b18–20 160 (17) VIII 3, 1249b14 8 VIII 3, 1249b14 ff. 4 VIII 3, 1249b15–16 160 (19), 287 E. N. I 2 275 I 4, 1096b26–29 177 I 12, 1101b23 f. 176 (78) VI 2, 1139a10 f. 176 (82) VI 2, 1139a27 f. 176 (81) VII 15, 1154b25 f. 165 VIII 2, 1155b13–15 177 VIII 2, 1155b14 f. 164 VIII 5, 1157a12–14 177 (86) VIII 8, 1158b4 f. 177 (86) IX 4, 1166a26 172 IX 9 241 (28) X 4, 1174b21–23 168 X 4, 1174b31–33 165 (38) X 4, 1175a12 158 (4), 162 X 5, 1175a29 f. 165 (38) X 7–9 7 X 7, 1177a14 f. 168 X 7, 1177a18 163 (31) X 7, 1177a19–21 168 X 7, 1177a26 f. 171 X 7, 1177a27 f. 172 (64) X 7, 1177b26–1178a8 287 X 7, 1177b33 f. 161 (22) X 8 162, 164, 176, 177, 178 X 8, 1178b7 f. 176 X 8, 1178b7–32 163 (31), 176, 287 X 8, 1178b9 176 X 8, 1178b18 f. 176 X 8, 1178b22 162, 176, 177 X 8, 1178b22–31 7 X 8, 1178b26 f. 176 X 8, 1178b27 164, 177 X 8, 1178b28–30 178 X 8, 1178b29 f. 177 X 8, 1178b32 177 (84) X 8, 1179a22–32 287 Met. Α 30, 31, 32, 33, 38 (100), 39, 43, 45, 52, 53, 61 Α–Δ 61

297

Α 2, 982a223 ff. 234 Α 2, 982b6 f. 160 (18) Α 2, 982b28–30 172 (64) Α 2, 983a8 f. 272 (3) Α 3–5 60 Α 3–7 30, 34, 60 Α 3, 983b9–10 71 Α 3, 984b8–22 280 (11) Α 4, 985a18–22 280 (11) Α 5, 987a6 30 Α 7 60 Α 7, 988b2–5 189 Α 8, 989b24–25 69 (11) Α 9 135 Α 9, 991b3–9 136 (31) Α 9, 991b5–7 130 (19) Α 9, 991b6–7 129 (18) Α 9, 992a18–993a10 82 Α 10, 993b20–24 68 α 5, 31, 32, 33, 38 (100), 39 (104), 45, 49, 53, 61, 68, 69, 85 α‒Κ 54 α–Ν 18 α 1 83 α 1, 993a30 68 α 1, 993b7–11 162 α 1, 993b20–21 83 α 2 83, 113 α 3, 995a3‒4 29 (75) α 3, 995a5‒6 29 (74) α 3, 995a17–19 31 (83) α 3, 995a18 83 α 6, 987b31–32 74 Β 31, 38 (100), 43, 47, 49, 115, 116, 123 Β 1, 995b10–13 35 (92) Β 1, 995b13–18 161 (25) Β 1, 995b18–20 35 (92) Β 1, 995b31–36 119 (2) Β 2, 997a15–25 35 (92) Β 2, 997a25–34 35 (92) Β 2, 997a34–998a19 161 (25) Β 4 127 Β 4, 999a24–b24 119 (2) Β 4, 999b5–16 123 Β 4, 999b6–14 94 (11) Β 4, 999b18–20 130 (19) Γ 35, 38 (100), 43, 47, 68, 81, 83, 144 Γ 1 79 Γ 2, 1003b10 72 Γ 2, 1003b12–14 145 Δ 40 (108), 43

298

Indices

Δ 2, 1013b28 ff. 222 (70) Δ 8, 1017b25 150 Δ 9 240 Δ 9, 1018a7 f. 240 Δ 15 240 Δ 26 69 Ε 16, 37, 43, 68, 80, 81, 83 Ε–Κ 15 Ε–Ν 11, 12, 13, 15, 16 Ε 1 79, 80, 81, 83, 84 Ε 1, 1025b25 176 (81) Ε 1, 1026a20–29 169 (52) Ε 2, 1027a8–14 283 Ε 2, 1028b4 72 Ε 2, 1042b10 f. 185 (10) Ζ 5, 37, 39, 40, 42, 45, 50, 56, 61, 68, 72, 73, 81, 83, 111, 116, 119, 148, 153, 183, 193 Ζ–Η 2, 6, 7, 40 (110), 48, 116, 139, 143, 144, 145, 147, 148, 149, 150, 152, 153, 154, 155 Ζ–Ν 12, 16 Ζ 1 69, 72 Ζ 1, 1028a25 f. 177 (87) Ζ 2 154 Ζ 2, 1028b27–32 161 (25) Ζ 3 127, 151 (9) Ζ 3, 1029a11–33 234 Ζ 3, 1029a29 151 Ζ 3, 1029b4–12 162 (29) Ζ 4–6 116 Ζ 7 40 (106) Ζ 7–8 119 (2) Ζ 7–9 6, 39 (106), 40 (109), 94, 114, 116, 119, 121, 148 (6) Ζ 7, 1032a24 124 (12), 274 Ζ 7, 1032a29–b5 40 (106) Ζ 7, 1032b1–14 130 (20) Ζ 7, 1032b11–14 96 (13) Ζ 7, 1032b14–20 40 (106) Ζ 7, 1032b29–1033a1 40 (106) Ζ 8, 1033b1–16 122 (7) Ζ 8, 1033b29 134 (28) Ζ 8, 1034a8 153 Ζ 9 119 (2) Ζ 10–11 116 Ζ 11, 1037a11–12 154 Ζ 11, 1037a29 167 (47) Ζ 12 116 Ζ 13 39 (106), 143 (3), 152 Ζ 13–16 116

Ζ 17 116 Ζ 17, 1041a8–9 154 Ζ 17, 1041b28 153 Η 2–3 185 Η 37, 39 (105), 40, 42, 45, 50, 56, 61, 68, 81, 116, 148, 183 Η 1 147, 151 (9) Η 1, 1042a7 77 Η 2, 1042b31–1043a1 37 (94) Η 3, 1043a14–23 119 (2) Η 3, 1043b14–16 130 Η 3, 1043b19–21 130 (19) Η 3, 1043b20 131 Η 4, 1044a29–31 130 (20) Θ 39 (104), 41, 42, 45, 47, 50, 56, 61, 68, 108, 183, 194 (25) Θ 1, 1045b34‒35 47 Θ 6, 1048b6–9 162 Θ 6, 1048b18–36 174 (73) Θ 8 40 (108), 174 (72), 185 (10), 194 (25), 196 Θ 8, 1050b7 f. 190 Θ 8, 1050b22–28, 194 (25) Θ 8, 1050b28 f. 161 (22) Θ 8, 1050b28–30 237 (22), 255 (24), 287 Θ 10, 1051b31 181 (2), 185 (10) Ι 39 (104), 40 (108), 41 (112), 42, 45, 47, 50, 56, 61 Ι 1, 1052a25 ff. 40 (108) Ι 3, 1054a29‒32 47 Κ 47, 50, 51, 54, 55, 56, 61 Κ 3, 1061a29 234 Κ 4, 1061b17 ff. 234 Κ 7 80 Λ 1 1, 4, 68, 72, 74, 77, 82, 84, 87, 88, 106, 119, 141, 146, 252 (19), 272 Λ 1–5 89 (2), 91 (5) Λ 1–7 62 Λ 1, 1069a18–19 67 Λ 1, 1069a19–21 69 Λ 1, 1069a20 f. 160 (16) Λ 1, 1069a21–24 71 Λ 1, 1069a24 72, 75 Λ 1, 1069a25–30 73 Λ 1, 1069a27–28 74 Λ 1, 1069a28 82 Λ 1, 1069a30–36 75, 88 Λ 1, 1069a32 75, 76, 77 Λ 1, 1069a33 f. 161 (25) Λ 1, 1069a34 84 Λ 1, 1069a36 84

1. Index locorum Λ 1, 1069a36–b2 78 Λ 1, 1069b1 88 Λ 1, 1069b1–2 141 Λ 1, 1069b2 83 Λ 1, 1069b3 90 Λ 2 6, 87, 88, 89, 90, 91, 92, 94, 97, 102, 106, 113, 114, 120, 121, 127, 132, 272 Λ 2–3 82, 114 Λ 2–5 1, 5, 77, 84, 88, 89, 106, 110, 115, 116, 143, 146, 234 Λ 2, 1069b3–9 89, 90, 91, 92 Λ 2, 1069b6 92 Λ 2, 1069b7–8 92 Λ 2, 1069b9–13 120 (5) Λ 2, 1069b24–26 89 (2) Λ 2, 1069b25–26 77 Λ 2, 1069b28–29 133 (23) Λ 2, 1069b32–34 91 Λ 3 6, 87, 94, 95, 96, 102, 113, 114, 119, 120, 121, 124, 129, 132, 135, 147, 148 (6), 149 Λ 3, 1069b35–36 94 Λ 3, 1069b36 131 (21) Λ 3, 1069b36–1070a1 94 Λ 3, 1070a1 95, 104 Λ 3, 1070a4 95 Λ 3, 1070a4–5 95 Λ 3, 1070a7–8 95 Λ 3, 1070a8 274 Λ 3, 1070a9 132 Λ 3, 1070a9–11 133 (23) Λ 3, 1070a9–13 6, 147, 148, 149, 150 Λ 3, 1070a13 132 (27) Λ 3, 1070a13–14 129 Λ 3, 1070a13–20 150 Λ 3, 1070a14 131 Λ 3, 1070a14–15 129 Λ 3, 1070a18–20 132 Λ 3, 1070a19–20 129 Λ 3, 1070a20 131 Λ 3, 1070a21 132, 133 Λ 3, 1070a21–22 96, 132 Λ 3, 1070a21–24 132 Λ 3, 1070a21–26 133 Λ 3, 1070a24–26 132 Λ 3, 1070a27 134 Λ 3, 1070a27–28 125, 274 Λ 3, 1070a29 96, 133 Λ 4 1, 5, 87, 95, 96 (12), 97, 99, 100, 102, 103, 104, 105, 106, 107, 109, 110, 113, 114, 115, 121, 145, 150, 152, 274

299

Λ 4–5 82, 96, 113, 114, 115, 116, 142 Λ 4, 1070a31–33 97 Λ 4, 1070a32 144 (4), 152 Λ 4, 1070a33–35 98 Λ 4, 1070a33–b10 98 Λ 4, 1070a35–b4 2, 274 Λ 4, 1070a35–b10 121 Λ 4, 1070a36 98 Λ 4, 1070b4–9 99 Λ 4, 1070b8–10 98 Λ 4, 1070b9–10 100 Λ 4, 1070b10 99 Λ 4, 1070b10–11 100 Λ 4, 1070b10–12 124 (14) Λ 4, 1070b10–14 149 Λ 4, 1070b10–35 121 Λ 4, 1070b16–21 101 Λ 4, 1070b18–19 2, 274 Λ 4, 1070b22–25 97 Λ 4, 1070b23 145 Λ 4, 1070b25–26 2, 274 Λ 4, 1070b27 104 Λ 4, 1070b28 96 (13) Λ 4, 1070b31–33 87 Λ 4, 1070b33 103 Λ 4, 1070b34 145 Λ 4, 1070b34–35 82, 272, 276 Λ 4, 1070b35 145 Λ 5, 1070b36–1071a3 106 Λ 5 2, 87, 99, 104, 105, 106, 107, 108, 110, 113, 114, 120, 121, 144 (4), 146, 152, 276 Λ 5, 1071a3–11 108, 109 Λ 5, 1071a3–17 106 Λ 5, 1071a11–17 109 Λ 5, 1071a14 152 Λ 5, 1071a15–16 276, 279 Λ 5, 1071a17 1, 111, 274 Λ 5, 1071a17–24 106 Λ 5, 1071a18–19 110, 111 Λ 5, 1071a21 145 Λ 5, 1071a22 1, 274 Λ 5, 1071a24–29 106, 107, 112 Λ 5, 1071a26 142 Λ 5, 1071a27–28 112 Λ 5, 1071a27–29 152 Λ 5, 1071a29–30 106 Λ 5, 1071a29–b2 106, 112 Λ 5, 1071a31 144 (4) Λ 5, 1071a34–35 107 Λ 5, 1071a35–36 82, 110

300

Indices

Λ 5, 1071a36 146 Λ 5, 1071a37 144 (4) Λ 5, 1071b1–2 87, 106 Λ 6 77, 78, 88, 115, 116, 146, 185, 191, 196, 211, 219, 220 Λ 6–7 84, 110, 157, 174, 183, 188, 190, 196, 197, 199 Λ 6–9 88, 221 Λ 6–10 1, 2, 5, 6, 79, 80, 84, 143, 155, 208 Λ 6, 1071b3 88 Λ 6, 1071b3–11 208 (2) Λ 6, 1071b4 196 Λ 6, 1071b4 f. 76, 184 Λ 6, 1071b10 f. 251 (15) Λ 6, 1071b12 8, 197 Λ 6, 1071b12–13 287 Λ 6, 1071b12–17 208 (3), 211 (22), 219 (62 f.) Λ 6, 1071b12–22 3, 196 Λ 6, 1071b14 197 Λ 6, 1071b17 197 Λ 6, 1071b17–20 208 (4) Λ 6, 1071b18 197 Λ 6, 1071b19 f. 158 (6) Λ 6, 1071b19–20 199 Λ 6, 1071b19–22 195 Λ 6, 1071b20 3 (1), 169, 182 (6), 193, 194, 196 Λ 6, 1071b20–22 208 (5) Λ 6, 1071b21 3 (1) Λ 6, 1071b22 183, 194, 195, 196, 197 Λ 6, 1071b31 103 (15) Λ 7 7, 8, 70, 177, 188, 190, 191, 211, 212, 213, 216, 217, 220, 223, 229, 232, 233 Λ 7, 1072a5 f. 183, 197 Λ 7, 1072a19–26 217 (52) Λ 7, 1072a21–23 251 (16) Λ 7, 1072a22 196, 198 Λ 7, 1072a23 ff. 3 (1) Λ 7, 1072a23–26 159, 184 Λ 7, 1072a24 197 Λ 7, 1072a24–25 197, 203 Λ 7, 1072a24–26 182, 196, 202, 248 (2) Λ 7, 1072a24–27 208 (6) Λ 7, 1072a25 3 (1), 183, 193, 197 Λ 7, 1072a26 7, 196, 202, 238, 247 Λ 7, 1072a26–27 212 (30) Λ 7, 1072a26–b4 211 (25) Λ 7, 1072a27 166 (41) Λ 7, 1072a27–28 208 (7) Λ 7, 1072a30 234, 242

Λ 7, 1072a30–32 160 Λ 7, 1072a30–33 3 (1) Λ 7, 1072a30–b4 200 (34) Λ 7, 1072a31 3 (1), 168 Λ 7, 1072a31 f. 165 Λ 7, 1072a32 158 (6), 169 Λ 7, 1072a34–b1 160 Λ 7, 1072a35 166 (41) Λ 7, 1072b1–2 287 Λ 7, 1072b1–3 213 (32), 250 (11) Λ 7, 1072b2 f. 160 (19) Λ 7, 1072b3 4, 238, 247 Λ 7, 1072b3 f. 160 Λ 7, 1072b4 8 Λ 7, 1072b4 ff. 208 (8), 223 (72) Λ 7, 1072b5 183 Λ 7, 1072b8 3 (1), 159 (8) Λ 7, 1072b8–10 251 (17) Λ 7, 1072b10 3 (1) Λ 7, 1072b11 3 (1) Λ 7, 1072b12 191 (18) Λ 7, 1072b13 3 (1), 159 (8) Λ 7, 1072b13 f. 161, 237 (22) Λ 7, 1072b13–14 3, 82, 115, 141, 208 (9), 276 Λ 7, 1072b13–30 3 Λ 7, 1072b14 f. 163 Λ 7, 1072b14–16 158, 163, 165 Λ 7, 1072b14–19 165 Λ 7, 1072b14–26 7, 157, 164, 166, 178 Λ 7, 1072b14–30 162, 287 Λ 7, 1072b15 174 Λ 7, 1072b15–26 166 (42) Λ 7, 1072b16 165, (37) Λ 7, 1072b17 166 Λ 7, 1072b17 f. 165 Λ 7, 1072b18 f. 165, 166, 167, 168, 169, 233 Λ 7, 1072b18–24 163 Λ 7, 1072b19 f. 229 Λ 7, 1072b19–21 173, 175 (77), 191 Λ 7, 1072b19–24 169, 174, 233 Λ 7, 1072b20 169, 242 Λ 7, 1072b20 f. 170, 171 Λ 7, 1072b20–24 164 (36), 166 Λ 7, 1072b21 242 Λ 7, 1072b22 f. 191 (18), 233 Λ 7, 1072b22–24 172 Λ 7, 1072b23 169, 233 Λ 7, 1072b24 68, 234 Λ 7, 1072b24 f. 174

301

1. Index locorum Λ 7, 1072b24 ff. 237 Λ 7, 1072b24–26 164, 178 Λ 7, 1072b25 177, 195 (26), 234 Λ 7, 1072b25–30 238 Λ 7, 1072b26 164, 192 (23), 195 Λ 7, 1072b26–28 191 Λ 7, 1072b27 162, 191 (19), 193, 194, 195, 237 (22) Λ 7, 1072b28 3 (1), 195 (27) Λ 7, 1072b28 f. 176 (79) Λ 7, 1072b28–30 195 (26) Λ 7, 1072b30 3, 159 (9) Λ 7, 1072b30 ff. 212 (27) Λ 7, 1073a3 f. 3 (1) Λ 7, 1073a4 f. 159 (8) Λ 7, 1073a5–11 212 (27) Λ 7, 1073a6 f. 3 (1), 159 (8) Λ 7, 1073a11 3 (1) Λ 8 3, 84, 187, 221, 247, 249, 261, 275 Λ 8, 1073a14 f. 249 (7) Λ 8, 1073a23 3 (1) Λ 8, 1073a23–24 82 Λ 8, 1073a23–25 248 (4), 261 Λ 8, 1073a26–34 248 (5), 261 Λ 8, 1073a31 f. 252 (19) Λ 8, 1073b3–5 169 (52) Λ 8, 1073b5–7 68 Λ 8, 1074a14–23 251 (18) Λ 8, 1074a19 3 (1) Λ 8, 1074a19–20 248 (3), 250 Λ 8, 1074a19–22 250 (8) Λ 8, 1074a19–23 261 Λ 8, 1074a22–23 250 (9) Λ 8, 1074a25–28 265 (44) Λ 8, 1074a26–31 262 (40) Λ 8, 1074a31–38 261 (36) Λ 8, 1074a35 187 Λ 8, 1074a35 f. 158 (6) Λ 8, 1074a36 196 (29) Λ 8, 1074a36 f. 159 (8) Λ 9 84, 167, 175, 188, 190, 191, 229, 231, 232, 233 Λ 9, 1074b15 f. 230 Λ 9, 1074b15–35 286 Λ 9, 1074b21 242 Λ 9, 1074b26 230 Λ 9, 1074b28–35 3 Λ 9, 1074b29 f. 168 Λ 9, 1074b29–35 209 (10) Λ 9, 1074b31–33 191 Λ 9, 1074b33 f. 230

Λ 9, 1074b34 230 Λ 9, 1074b34 f. 237 Λ 9, 1074b35 f. 171 (59), 241 Λ 9, 1074b36 175, 230 Λ 9, 1074b36–38 241 Λ 9, 1075a1–3 130 (20) Λ 9, 1075a3 ff. 230 Λ 9, 1075a5 f. 230 Λ 9, 1075a6 230 Λ 9, 1075a6 f. 159 (8) Λ 9, 1075a7 f. 230 Λ 9, 1075a10 230 Λ 9–10, 1075a11 ff. 209 (11) Λ 10 10, 70, 82, 84, 212, 221, 222, 269, 270, 275, 280, 282, 285, 286, 288, 289, 290 Λ 10, 1075a11 285 Λ 10, 1075a11 ff. 209 (11), 212 (27) Λ 10, 1075a11–13 270 Λ 10, 1075a11–15 212 (28) Λ 10, 1075a11–25 3 Λ 10, 1075a14–15 271 Λ 10, 1075a15 289 Λ 10, 1075a16 142 Λ 10, 1075a16 f. 271 Λ 10, 1075a18 f. 10, 144 (4), 269 Λ 10, 1075a18–22 271 Λ 10, 1075a22–25 271 Λ 10, 1075a23 282 Λ 10, 1075a25–1076a3 271 Λ 10, 1075b37–1076a4 82 Λ 10, 1076a3 f. 269 Λ 10, 1076a4 269 Μ 56 Μ–Ν 12, 15 Μ 3, 1078a31 f. 160 Ν 39, 56, 85, 134 Ν 1, 1088a29 f. 177 (87) Ν 2–3 286 Ν 2, 1090a9 134 (29) Ν 3, 1090a17 f. 134 (30) Ν 3, 1090a17–20 134 Ν 3, 1090a20 134 (29) Meteor. I 9, 346b20

277

Parv. Nat. 149,14‒15

16 (29)

302 Phys. I 90, 91, 93, 94, 97, 102 I–II 81 I 2 88 I 2, 184b25–7 89 I 5 91, 92, 102 I 5–7 114 I 5, 188a35–b1 92 I 5, 188b6 92 (7) I 5, 188b8 91 I 5, 188b9 92 (7) I 5, 188b30–189a2 102 I 6, 189a22–26 92 I 6 92 I 6–9 120 I 7 90 (5), 92, 93, 102 I 7, 190a15–17 94 (10) I 7, 190a17–19 92 (8) I 7, 190b17–19 90 (4) I 7, 191a5–7 93 (9) I 7, 191a7–8 102 I 7, 191a19–22 89 (3) II 1 282 II 3 222 II 3, 195a26 ff. 222 (70) II 4, 196a24–35 285 II 7 133, 223 II 7, 198a21–27 254, 262 (39) II 7, 198a24–27 133 (24) II 7, 198a25–26 223 (75) II 7, 198a26–27 223 (76) II 8 282, 284, 288 II 8, 198b34–a8 283 III 1, 201b5–7 133 (25) III 5, 204b4 74 III 5, 204b10 74 V 1 120 VII 1, 243a32–33 220 (65) VII 2 220 VIII 76, 77, 79, 207, 217, 223, 224 VIII 3, 247b11 f. 174 (73) VIII 5 159 VIII 5, 256b20–27 223 (72) VIII 6 217, 223, 249 VIII 6, 258b13–16 214 (38) VIII 6, 258b16 ff. 223 (74) VIII 6, 259b6–20 222 (71) VIII 6, 259b7–37 214 (38) VIII 6, 259b28–31 249 (6) VIII 6, 260a5 217 (53) VIII 6, 260a5–10 223 (73)

Indices VIII VIII VIII VIII VIII

6, 260a7–10 217 (53) 9, 265a28–b9 287 10 276 10, 266a12–266b20 219 (64), 276 10, 267b2–4 224 (81)

Pol. I 2, 1253a9 281 I 5, 1254a28–33 289 I 8 282 I 8, 1256b10–22 282 VII 13, 1331b25 287 (14) Rhet. I 10, 1368b36–1369a4 160 (13) I 11, 1371b18 176 (82) Soph. el. 25, 180a36–38

72

Top. I 7, 103a18 f. 176 (82) I 10, 104a19 f. 176 (82) I 14, 105b20–29 74 I 17 177 I 17, 108a16 f. 177 (85) VI 6, 145a15 f. 176 (81) VI 10, 148a26–31 162 Asclepius Commentary on Aristotle’s Metaphysics (following Hayduck) 113,16‒20 32 (86) 113,20‒23 32 (87) 114,16‒17 32 (87) Homer Iliad II 204 269 Ibn al-Nadı¯m Fihrist 251,25–252,1

19 (39), 43 (117)

Ibn Rušd Long Commentary on the Metaphysics (Tafsı¯r ma¯ ba’d at-Tabı¯’at) (following Bouyges) 543,7–9 40 (108) 1393,4 20 (44)

303

1. Index locorum 1393,4‒1394,2 18, 20 1393,4–1405,12 4, 18, 20 1393, 7–8 58 1394,3‒1395,8 18 1394,4 48 1394,10 47 (123) 1395 25 (61) 1395,4 30 (80), 43, 61 1395,9‒10 19, 21, 23 1395,9‒1397,2 18 1395,9‒1405,12 18, 20, 59 (144) 1395,11 20, 27, 59 1396,3 27 (69) 1396,5‒1397,2 27 1397,3‒4 28, 57 1397,3‒1405,12 18 1397,5 43, 61 1397,5‒13 20, 29 1397,5‒1398,7 61 1397,12 29 (74) 1397,14‒1398,7 20, 29, 60 1398,8‒1399,8 33 1399,9‒1400,8 34 1400,4 34 (91) 1400,9‒1401,1 35 1401,2‒1402,3 35 1402,4‒1403,10 36 1403–1404 42 (116) 1403,11‒15 46 1403,16‒18 46 1404,1‒8 46 1404,9‒11 47, 60 1404,12‒16 47 1405,1‒3 56 1405,4‒8 20, 56, 62 1405,9‒12 19, 20, 58, 62 1406,5 41 (110) 1407,1‒5 40 (110) 1417,14–16 39 (106) 1425,1‒4 40 (107) 1425,10–13 47 (123) 1425,11 47 (123) 1425,75 76 1430,5 41 (112) 1431,3‒6 41 (112) 1439,9‒1440,1 41 (111) 1459,3‒4 40 (109) 1463,11‒12 39 (106) 1464,4–7 39 (106) 1563,2 199 1568 201

1568,12–13 199, 200 1576,5 40 1599 201 1599,7 200 (34) Ibn Sina The Metaphysics of The Healing. Al-Shifâ: Al-Ilâhiyyât 8.6, § 3 200 (35) Michael of Ephesus Commentary on Aristotle’s Metaphysics (following Hayduck) 685.25–687.22 199 687.6 199 Parmenides DK 28B8, 34 Philoponus In Phys. 241, 15–19 Plato Leg. 898 a Parm. 130b–e

242 (29)

216 (48)

218 (55)

129 (18)

Phaed. 96e–97b 135 97b9–10 280 97c4–5 280 98b–e 135 100b 128 (17) 100e 124 (13) Rep. I, 352d–354b 282 VI 275, 276 VI 509b 162 (28), 236 (20) VI 509b9 200 VII 82 Tim. 34 a 218 (55) 49D 151 (8) 49 d–e 71 (13)

304 Plotinus Enneads VI.8 [39] VI.8 [39] VI.8 [39] VI.8 [39] VI.8 [39] VI.8 [39] VI.8 [39]

Indices Cael. 398,36–399,4 201 16 8, 200 16 f. 200 16.35 200, 201 20 200 20.14 201 20.17 f. 201

Simplicius In Aristotelis Physicorum Libros Commentaria 1360 207 (1), 220 (68)

57 (142)

Theophrastus Metaphysics 2, 4a9–12 142 5a23–25 218 (57) 5a23–b10 210 (20) 5b7–10 218 (58), 251 (14) Thomas Aquinas Sententia super Metaphysicam XII lectio 11, 2614–5 9 (2)

2. Index nominum Ancient and medieval authors (The page references are italicized when they refer to a footnote) Abu¯ Bishr Matta¯ b. Yu¯nus 44, 46, 47, 50, 56, 59, 60, 199 Abu¯ l-Walı¯d 55 Alexander of Aphrodisias 2, 4, 5, 7, 8, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 37, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 67, 69, 70, 71, 73, 74, 75, 76, 77, 78, 79, 147, 172, 173, 181, 185, 186, 187, 188, 189, 190, 191, 192, 197, 199, 201, 210 Abu¯ Yu¯suf al-Kindı¯ 39, 44, 201 Ammonius 4, 32 Anaxagoras 191, 223, 271, 280, 286 Asclepius 4, 16, 32, 33, 61 Averroes: see Ibn Rušd Avicenna: see Ibn Sina Callias 7, 126, 127, 133, 148, 150 Callippus 250 Democritus

74

Empedocles 71, 74, 271, 286 Eudoxus 249 Finzi, Moses

193

Georgios Pachymeres

16

Heraclitus 71 Homer 37, 49, 269, 271, 289 H  unayn b. Ish a¯q 43, 44, 51 Ibn al-Nadı¯m 19, 39, 43, 44, 51, 60 Ibn Manz u¯r 24 Ibn Rušd 2, 4, 5, 11, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 32, 33, 34, 37, 39, 40, 41, 45, 46, 50, 54, 55, 57, 58, 59, 60, 61, 62, 67, 69, 70, 71, 73, 76, 78, 79, 147, 181, 199, 200, 201, 219

Ibn Sina 76, 79 Ibn Tibbon, Mosheh

193

Michael of Ephesus 12, 13, 15, 16, 17, 58, 62, 199 Michael Scot 11, 14 Naz ı¯f b. Ayman 30 Nicolaus of Damascus

20, 57, 62

Parmenides 71 Philoponus 16, 215 Plato 6, 36, 39, 70, 71, 73, 74, 78, 82, 84, 124, 128, 129, 131, 132, 135, 136, 154, 159, 162, 163, 189, 193, 200, 208, 217, 221, 236, 239, 240, 271, 274, 275, 276, 280, 282, 286, 289 Plotinus 8, 181, 193, 199, 200, 201 Ptolemy 257 Šamlı¯ 44 Simplicius 2, 139, 192, 207, 208, 210, 220, 221, 236 Socrates 6, 7, 121, 124, 126, 127, 128, 133, 134, 135, 148, 150, 280 Speusippus 70, 78, 81, 82, 84, 104, 236, 271, 286 Syrianus 12, 16, 17, 44 Themistius 2, 5, 18, 19, 44, 52, 67, 69, 70, 71, 72, 73, 76, 77, 79, 192, 210 Theophrastus 37, 38, 49, 153, 210, 218, 229, 236, 251, 253 Thomas Aquinas 2, 5, 8, 67, 70, 72, 74, 78, 79, 80, 81, 210 Usta¯t

38, 39, 44, 53, 55, 60, 199

Xenocrates

70, 74, 78, 82, 84, 236

Zakarı¯ya¯ Yah ya¯ b. Adı¯

43, 44

306

Indices

Modern Authors (The page references are italicized when they refer to a footnote) Adamson, P. 11, 39, 200, 202 Afnan, M. A. 33 Alexandru, S. 202, 203 Amerini, F. 28 Anagnostopoulos, G. 116, 251 Aubenque, P. 84 Badawi, A. 193, 200 Baghdassarian, F. 210 Balme, D. M. 264, 284 Baltussen, H. 221 Barnes, J. 158, 164, 165, 168, 170, 171, 177 Beere, J. 229, 256 Berti, E. 3, 4, 5, 8, 68, 81, 82, 185, 201, 202, 210, 211, 212, 213, 214, 216, 229, 231 Bertolacci, A. 39, 43, 51, 53, 55 Bekker, I. 6, 40, 75, 140, 148, 191 Bianchi, L. 202 Bodnár, I. 9, 256, 285, 286 Boeri, M. 210 Bonelli, M. 83, 210 Bonitz, H. 12, 67, 70, 74, 77, 79, 147, 185, 202 Bordt, M. 9, 158, 159, 162, 163, 165, 172, 233, 234, 250 Botter, B. 210, 219 Bouyges, M. 4, 11, 13, 17, 18, 19, 20, 21, 22, 23, 25, 26, 27, 29, 38, 39, 41, 42, 43, 44, 47, 50, 51, 55, 59, 62, 199, 200, 201 Bowen, A. C. 280 Bradshaw, D. 211, 212, 213, 215, 218, 223 Brague, R. 67, 79, 200 Brancacci, A. 210 Brandis, C. A. 12, 15, 17, 202 Broadie, S. 3, 8, 211, 212, 213, 214, 238, 251, 252, 287 Brockelmann, C. 44, 51 Bruns, I. 181, 188, 199 Brunschwig, J. 232, 233 Buddensiek, F. 169 Burnyeat, M. F. 72, 119, 121, 170, 172, 181, 230 Bywater, I. 171 Calvo, T. 68 Cardullo, L. 82

Catan, J. R. 283 Cathala, M.-R. 67 Charles, D. 4, 68, 88, 91, 94, 98, 105, 116, 119, 143, 144, 148, 154, 158, 163, 185, 202, 209, 210, 211, 231, 232, 234, 239, 272, 273 Christ, W. 182 Code, A. 105, 107, 109 Cooper, J. M. 264 Corcilius, K. 229 Cornford, F. 231 Couloubaritsis, L. 211 Crubellier, M. 6, 78, 80, 98, 144 Dahl, N. O. 171 Damschen, G. 68 De Filippo, J. G. 238 De Goeje, M. J. 14, 17 De Koninck, T. 235 Delcomminette, S. 211 Denniston, J. D 167 Devereux, D. 80 Di Giovanni, M. 4, 28, 35, 200 Dodge, B. 43 Donini, P. 81, 82, 83 Drossaart Lulofs, H. J. 57 Düring, I. 68, 83, 160 Elders, L. 70, 74, 80, 210, 238 Endress, G. 39, 44, 56 Enskat, R. 68 Erler, M. 210 Fazzo, S. 7, 8, 68, 72, 136, 182, 184, 185, 186, 191, 195, 198, 203, 218, 220, 252 Ferro, A. 116 Flashar, H. 236 Flügel, G. 19, 43 Forbes, F. H. 218, 229 Fortenbaugh, W. 80 Fränkel, S. 11, 14, 23, 25, 26, 67, 229 Frede, M. 4, 68, 69, 70, 71, 73, 74, 77, 78, 80, 81, 82, 88, 94, 98, 105, 119, 143, 144, 148, 151, 154, 158, 163, 185, 202, 209, 210, 211, 229, 231, 232, 234, 239, 272, 273, 284

2. Index nominum Freudenthal, J. 4, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 25, 26, 27, 30, 37, 41, 42, 55, 57, 58, 59, 60, 61, 62, 67, 69, 78, 229 Furley, D. J. 284 Gabriel, M. 164, 239 Gadamer, H.-G. 238 Gaiser, K. 160 Galluzzo, G. 28 Genequand, C. 17, 18, 21, 23, 26, 29, 33, 34, 44, 55, 67, 76, 78, 200, 201, 231, 238 Geoffroy, M. 19 Gerson, L. P. 159, 171, 235 Gill, M. L. 211, 236 Golitsis, P. 16 Gómez-Lobo, A. 210 Gotthelf, A. 80, 237, 264, 284 Gourinat, J. B. 210 Guerrero, R. 18, 21, 24, 26, 29, 33, 43, 45, 46, 58 Gutas, D. 23, 44, 51 Guthrie, W. 231, 237 Hadot, I. 221 Hadot, P. 201 Halfwassen, J. 239 Halper, Y. 13 Hasse, D. N. 11 Hayduck, M. 11, 12, 13, 15, 31, 32, 34, 173, 192, 199, 229 Heinze, R. 74, 78 Henrich, D. 231 Henry, P. 201 Herzberg, S. 7, 157, 167 Heubeck, A. 38, 41 Hicks, R. D. 171, 192, 193, 194 Hoffmann, R. 35 Horn, C. 158, 159, 160, 229, 235, 239 Huidobro, J. 210, 218

307

Kahn, C. 7, 80, 171, 237, 287 Kobusch, T. 210 Kosman, A. 163, 164, 173, 174, 211, 214, 236 Krämer, H.-J. 141, 160 Kraut, R. 176 Kretzmann, N. 232 Kroll, G. 12 Laks, A. 158, 159, 163, 165, 202, 209, 210, 218, 220, 229, 231, 238, 239 Landauer, S. 193 Lane, E. W. 23, 24 Lang, F. 78 Lefebvre, D. 83 Lennox, J. G. 211, 236, 264, 284 Lesher, J. 143 Leunissen, M. 260, 284 Lewis, G. 201 Liatsi, M. 8, 9 Luna, C. 11, 15, 16, 17 Madigan, A. 34 Manison, A. 83 Manuwald, B. 239 Marmura, M. E. 200 Marqués, A. 210, 218 Martin, A. 17, 18, 21, 26, 29, 33, 34, 40, 41, 42, 55, 58 Menn, S. 94, 104, 116, 159, 161, 162, 210 Meyer, S. 284 Miller, J. 171 Moraux, P. 16, 17, 32, 57, 81 Most, G. W. 229 Movia, G. 67 Murshel, A. 257 Natali, C. 210 Nord, I. 231 Norman, R. 166, 175 Nussbaum, M. C. 171, 283

Irwin, T. 284 Jaeger, W. 2, 5, 6, 32, 67, 68, 69, 71, 75, 84, 85, 147, 148, 172, 173, 186, 202, 203, 236, 273 Jeffery, L. H. 38, 41 Johansen, T. 280 Johnson, M. R. 284 Judson, L. 3, 6, 94, 119, 148, 151, 211, 214

Oehler, K. 162, 164, 166, 168, 171, 172, 175, 178, 240 Oksenberg Rorty, A. 171 Omar, I. A. 200 Owens, J. 170, 283 Patzig, G. 151 Peck, A. L. 235 Perler, D. 170

308 Peters, F. E. 43 Pfeiffer, L. 116 Praechter, K. 15, 16 Primavesi, O. 4, 37, 49, 229 Rapp, C. 5, 177 Rashed, M. 11, 210, 218, 278 Ravaisson, F. 12, 15, 16 Reale, G. 68, 70, 77, 79, 81, 210 Reeve, C. D. C. 173 Richardson Lear, G. 160, 161 Roediger, E. 43 Rose, V. 11, 12, 13, 19, 21, 59, 162 Rosemann, P. W. 229 Ross, A. 209, 275 Ross, W. D. 6, 8, 37, 67, 69, 70, 71, 72, 74, 75, 77, 79, 83, 122, 124, 136, 144, 147, 148, 150, 151, 163, 166, 168, 171, 172, 173, 185, 186, 191, 192, 194, 195, 202, 203, 210, 218, 229, 231, 273 Rudolph, U. 39, 44, 51 Salis, R. 68, 76, 211, 212 Sandbach, F. H. 231 Scharle, M. 284 Scheibe, E. 240 Schüssler, W. 231 Schwegler, A. 231 Schwyzer, H.-R. 201 Sedley, D. 161, 210, 239, 273, 282, 284, 285, 287, 290 Sharples, R. 80, 210

Indices Sinaceur, M. A. 35 Spiazzi, R. M. 67, 210 Spiegel, Y. 231 Smith, J. A. 192 Steel, C. 229 Steinschneider, M. 14, 44, 51, 55 Stevens, A. 211, 212, 214 Striker, G. 284 Taylor, R. C. 200, 201 Tarán, L. 16 Torstrik, A. 191, 193 Tricot, J. 68, 70, 72, 74, 77, 79 Usener, H.

12

Viano, C. 80 Vigo, A. 68, 218 Vlastos, G. 3 Vuillemin-Diem, G. 32 Wandrey, I. 11 Wedin, M. V. 164, 167, 175, 177 Weltecke, M. 157 Wendland, P. 12 Wieland, W. 231 Wiesner, J. 16, 17, 32, 57, 81 Wildberg, C. 280 Zeller, E. 231, 239 Zingano, M. 6, 7, 229 Zonta, M. 13, 55, 185, 192, 202

3. Index rerum absence/presence (of form) 93 Academy 74, 78, 81–5, 129, 236 accidents 70, 72–3, 130, 139, 153 activity 183–5, 189–92, 213, 215, 250–2 257; 168, 213 (highest –) actuality/potentiality 106–11, 130, 158–60, 165, 169–71, 173, 185, 189–90, 193–4, 198, 234, 278, 290; 162–3, 168, 172, 175, 199 (pure –) actus purus theory 158, 184, 197–201 analogy 101–2, 108–9, 114, 121, 142–3, 168–9, 177–8, 221–2 animal 69, 75, 127, 140, 216, 222–3, 235, 273, 277, 284–5 aporia 121, 257 arbitrariness: see randomness army (in analogy to the universe) 209, 212, 221, 269–71 art 95, 125, 248; 96, 103, 130, 150 (– of building) artefacts 129, 135–6, 150, 194 astronomy 68, 281 autonomy, divine 230

change 90–4, 120, 123, 142, 145, 149; 149 (– from cold to hot), 272 (four types of –) cho¯riston 72, 75, 146 colour 71–2, 92, 101 (– white/non-white), 92, 101 (black/non-black) compound 142, 150 consciousness 171, 175, 241–2 contemplation 163–4, 170–8; 172–8, 213, 215 (human/divine –), 287 (– as imitation of the Prime Mover) contraries 92–3, 140–2 corruption 130, 147, 150 cosmos 161, 213, 269 creationism 274

beauty 135, 168, 213 being 99, 106–11, 144, 169; 27–8 (– qua being) begetting 1, 95, 103, 106, 123, 125, 133–5, 145, 216, 255, 274 biology 131 body 73, 125, 134, 223, 248; 2, 76–7, 83, 253–4, 257–9, 281 (celestial/heavenly –)

earth (planet) 258, 275 elements 1–2, 71, 149, 274, 287; 99 (material –), 99 (intelligible –) energeia 146, 148, 279; 181–97 (in act/act) essentialism 143, 161 eternity 183–4, 208, 214, 217 ethics 83 excellence 258–61; 270–1 (cosmic –) extension/intension 140, 153

categories 2, 70–1, 98–9, 140–2, 144–7 causality 2, 207–9, 211–2, 274–5; 2 (analogous –), 1–2, 274, 276–8 (homoeidetic –), 275(– of the Prime Mover), 1–2 (synonymous –) cause 1–2, 99, 132, 143, 254, 34; 79, 82, 95–7,123–4, 133, 135, 145, 211, 224 (efficient/moving –), 79, 85, 133, 159–61, 207, 211, 224, 282–3 (final –), 79, 84–5, 133, 135, 211, 224 (formal –), 84–5, 135, 124, 133 (material –), 32 (physical –), 95– 6 (pre-existing/simultaneously existing –) chance 125–6, 283; 271 (acting by –)

definiens 146 definition 146 desire 209–10, 212, 218, 220, 287 dialectics 82 differentiae specificae 70, 168 dignity 168 divine design-argument 269–70 dunamis 146, 148, 183

father 1, 96, 109, 125, 145, 274 feminine 84 First Mover: see ‘Prime Mover’ first philosophy 31, 67, 78–9, 89, 234, 236, 238 form – (eidos) 91, 93–8, 100–16, 122–3, 127 , 131, 142, 148, 150, 152, 185, 187, 234; 2, 128, 132–3, 219, 274 (platonic –) form-division 146–50 fortune 286–7 friendship between God and humans 287 function theory 282

310

Indices

generation 93, 95, 97–8, 113, 120, 122–31, 149–50, 277 generation and corruption 216, 277 genus 69, 73–4, 100–1, 112, 139 geometry 68–70 goal 250–63 (– as a beneficiary / as not a beneficiary), 281–2 (– directedness) God 157, 162–4, 166, 169, 174–8, 269, 286–7; 194–5, 287 (– as final good) good 84, 159–60, 200, 212–3, 215, 270, 291; 275 (– as cause of all other beings), 288 (external –), 288 (– of mutual benefit), 288–9 (internal –), 288–9 (– of an overarching perfection) growth/diminution 278 happiness (eudaimonia) 176–8, 237, 288–9 health/illness 95–6, 103, 123, 126, 129, 135–6, 254, 291 heart 69 heaven 145, 195–6; 158–9, 210, 217 (first –) homoeides 1, 274, 276–7 homoeomery 131 homonymy 1–2, 70, 72 hope 165 household 221, 271, 285, 288 humans 273; 282, 284 (– as end for which the universe is teleologically arranged) hylomorphism 186 idea 134–5; 154 (Platonic –) identity 239–42 idion 140–1 imagination 167 imitation (mime¯sis) 216–7, 252, 255–6; 286–7 (– of the Prime Mover) immortality 134, 284, 286–7 indefinite dyad 82, 85 individuals 139–43, 147–54 intellect (nous) 134, 169–75, 229–30; 187– 8, 190–3, 233 (– as potentiality / as act) intelligibility 187 knowledge

72, 162, 167, 230–7, 239–42

life 158, 162, 195, 237; 163–6 (best form of –), 216, 279–80 (cycle of –) logos 133, 135, 146 love 160, 209–10, 212–3, 220–1, 247–9, 252–3 luck 6, 95, 125–6

masculine 84 mathematics 68, 70, 75, 82; 119–20 (objects of –) matter (hyle¯) 91, 93–4, 103, 107–14, 122, 127, 131, 142, 147–54, 158, 187; 272 (four types of –) medicine 129, 134–5 memory 165 metaphysics 35, 79, 81, 83, 234 moon 210, 256, 258, 273, 277, 280 motion: see ‘movement’ movement 2, 71, 80, 88–9, 95, 139, 143, 153–4, 208, 212–14, 223, 247–54, 256– 59, 263–66; 217–18, 247, 251, 275 (circular –), 219–20, 236–7 (eternal –), 272, 275–6 (four types of –), 196, 209–10 (heavenly –), 276 (infinite –) mover(s) 142; 234, 248, 275 (– of the celestial spheres) natural laws 237 natural objects 129–30, 150 nature 95, 120, 123, 125, 130, 133, 145, 151, 161 Neoplatonists 73, 183, 201 non-sensible principle 189 notion 72 numbers 70, 134–5, 223, 230, 286 omniscience 232 one 82, 85, 99 ontological thinness/thickness 143, 145–6, 149–54 ontology 141, 143, 147, 150, 152, 198, 238 opposites 92 order 213, 221–2, 230; 269, 271, 273, 275–6, 289 (cosmic –) perception 162, 165, 167–8, 241 perfection 161, 261, 273 periekhon 214, 222–3 persistence 92–3, 142, 149, 151 person 237–8 philosopher 287 philosophy 132 (– logical), 76, 119, 132 (natural –), 280–1 (– of nature), 83 (practical –) physics 31, 76, 78–81, 83–5, 89 place 149 planets 210, 256, 273

311

3. Index rerum plant 69, 75, 273, 277; 143 (deciduous trees) pleasure 165, 166, 168, 174, 235 Presocratics 68, 70–1, 73–4 Prime Mover (First Mover) 1–4, 95, 104–5, 166, 207–24, 229–237, 242, 272–76; 273– 4 (activity of the –), 275 (causality of the –) principle 2, 73–4, 76–85, 88–116, 142; 157, 247 (first –) Prinzipienlehre 83–5 privation 71, 91, 93–4, 102, 142, 149–50 privation-division 146, 148–50 providence 286 (– divine) psychology 167, 192 purity 200–1 quality 71–2, 140, 149, 177 quantity 149 rain 282–3 randomness 222, 269 realism 143 reality 70–1, 123, 130, 168–9, 286 reason (nous) 162, 167 (– activity of), 280 (– as cause of everything), 168 (faculty of –), 280 (ordering –) receptivity 172 reflexivity 239 relation 143 ruler, divine 270–1, 289 sameness 240; 97, 99–100, 105–7 (– of principles) scala naturae 259 science 83, 153, 234 (– theoretical), 83 (practical –), 83 (poietical –), 145 (– of being) seasons 277 self-reference 175, 233 self-sufficiency 172 senses 73 sexual reproduction 287 (– as imitation of the Prime Mover) slaves/freemen 221, 271, 285 soul 134, 168, 209, 214, 222–4, 248, 254; 255 (vegetative –) species 7, 36, 69, 100, 102, 112, 133, 139, 143, 152, 255, 284; 255–6 (vivipara/ ovipara) spheres 146, 209, 212–4, 247–9, 251–3, 259–66, 275

spontaneity 95 stars 262–3, 281; 256, 265, 273, 275, 287 (fixed –) substance (ousia) 70–84, 99–100, 106–7, 127, 131, 139–46, 153–5; 88–93 (changeable –), 76–7 (corruptible/incorrubtible –), 76 (eternal –), 190 (first –), 189–90 (heavenly –), 75–8, 80–2, 87–91, 93, 96, 104, 106–7, 113–6, 139, 141–3, 145–6, 153–5, 189 (sensible/non-sensible –), 1, 160, 162, 272 (terrestrial/celestrial/ immaterial –), 78–9, 82, 84 (unmovable –) sun 2, 80, 109–10, 153, 210, 256, 258, 273, 276–7, 280; 278–9 (cycle of the –) sunanairesis principle 142, 144, 146 synonymy 124; 5–6, 95–6, 119, 121, 135 (principle of –) techne¯ 1, 135 teleology 131, 160, 209, 280, 282, 284–6; 247, 263 (celestrial –), 284–5 (perfectionist –) telos 236–8 theology 2–3, 67, 79–81, 153, 193, 197–8, 234, 286–7; 234, 239 (christian –) theory (theo¯ria) 67–9, 163–4, 172–8, 234–7 thinking (noe¯sis) 157, 191, 218, 230; 163– 75 (human/divine –) this (tode) 127–9, 133, 140, 146–51 time 72, 132 transcendence 162–3, 174, 284 unio mystica 241 unity 102, 107–8, 147, 151, 164–6, 175, 177 universality 143 universals 73–4, 113, 152–3 universe (pan) 75, 81, 146, 154, 216, 221, 237, 269–70, 282, 287–91; 270 (nature of the –) unmoved mover(s) 80–2, 154, 181–3, 185– 92, 194, 198, 248–53, 261, 273; 160 (– as unchangeable and self sufficient telos), 249 (– moved/not moved incidental) virtue

168

whole 69–70, 124, 146 world 152; 235–6 (immanent –), 1, 109, 196, 221–2, 273 (sublunary –)