THE STRUCTURE OF BEING IN ARISTOTLE'S METAPHYSICS

The New Synthese Historical LibraryTexts and Studies in the History of Philosophy

VOLUME 52

Managing Editor:

SIMO KNuUTTILA, University of Helsinki

Associate Editors:

DANIEL ELLIOT GARBER, University of Chicago

RICHARD SORABJI, University ofLondon

Editorial Consultants:

JAN A. AERTSEN, Thorrws-Institut, Universitiit zu Koln, Germany

ROGER ARIEW, Virginia Polytechnic Institute

E. JENNIFER ASHWORTH, University ofWaterloo

MICHAEL AVERS, Wadham College, Oxford

GAIL FINE, Cornell University

R. 1. HANKINSON, University of Texas

JAAKKO HINTIKKA, Boston University, Finnish Academy

PAUL HOFFMAN, University of California, Riverside

DAVID KONSTAN, Brown University

RICHARD H. KRAUT, Northwestern University, Evanston

ALAIN DE LIBERA, Ecole Pratique des Hautes Etudes, Sorbonne

JOHN E. MURDOCH, Harvard University

DAVID FATE NORTON, McGill University

LUCA OBERTELLO, Universita degli Studi di Genova

ELEONORE STUMP, St. Louis University

ALLEN WOOD, Stanford University

The titles published in this series are listed at the end of this volume.

THESTRUCTURE OF BEING IN ARIS TOTLE , S

METAPHYSICS

By

JlYUANYU

State University of New York at Buffalo, U.s.A.

SPRINGER-SCIENCE+BUSINESS :MEDIA, B.V.

Library of Congress Cataloging-in-Publication Data

ISBN 978-94-010-3991-8 ISBN 978-94-010-0055-0 (eBook) DOI 10.1007/978-94-010-0055-0

Printed an acid-free paper

AII Rights Reserved © 2003 Springer Science+Business Media Dordrecht Originally published by Kluwer Academic Publishers in 2003 Softcover reprint ofthe hardcover Ist edition 2003

No part of this work may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, microfilming, recording

or otherwise, without written permission from the Publisher, with the exception of any material supplied specifically for the purpose of being entered

and executed on a computer system, for exclusive use by the purchaser of the work.

To Yajie

CONTENTS

Acknowledgements ixAbbreviations xi

Introduction xiii

Chapter 1 Per se Being and Potential/actual Being 1i. Per se being 1ii. PotentiaVactual being 9iii. The scope of the focal structure of being 18

Chapter 2 From Being to Substance 21i. The focal connection of per se beings 22ii. PotentiaVactual being and substance 36iii. The science of being 42

Chapter 3 Hylomorphism and Its Two Approaches 49i. The composition of substance 51ii. Two hylomorphic approaches 57iii. The new start ofMetaphysics vii. 17 65iv. Re-grouping the central books 76

Chapter 4 Subject, Form~ and Essence 83i. Substance and subject 85ii. Form and subject 92iii. Substance and essence 96iv. The identity of form and essence 105

Chapter 5 Form: Tode ti and Toionde 113i. Primary reality and knowledge 115ii. Separation, tode ti, and toionde 118iii. Tode ti, particularity, and individuality 122iv. Separation 127v. Tode ti and definition 132vi. Form as toionde 137vii. The universal and toionde 145viii. The ending ofMetaphysics vii 150

Chapter 6 Substantial Potentiality and Actuality 155I. Kinesis and energeia 156

vii

viii

ii.iii.iv.

Chapter 7i.ii.iii.

Chapter 8

Bibliography

CONTENTS

Substantial generationSubstantial activitySubstantial unity

Actuality and the Prime MoverFrom actual form to the Prime MoverThe Prime Mover and eternityThe Prime Mover and the world order

The Unity of Aristotle's Metaphysics

162169176

181181188197

201

211

Index of PassagesName IndexSubject Index

223231235

Acknowledgements

The central idea of this work is that the distinction that Aristotle drawsbetween per se being and potentiaVactual being should be taken seriously. Iargue that a correct application of this distinction to the central books of theMetaphysics leads to a distinct interpretation of these texts and to a betterunderstanding of Aristotle's science of being.

Some parts of the book have appeared as journal articles, though in allcases they have been substantially altered or recast. Sections i and iii ofChapter 2 are from Apeiron 34 (2001). Part of Chapter 3 has appeared inOxford Studies in Ancient Philosophy 15 (1997). Sections iii and iv of Chapter4 can be found in the Southern Journal of Philosophy 39 (2001). A remoteancestor of Chapter 5 was published in the Philosophical Inquiry 16 (1994).Permission to re-use these materials is gratefully acknowledged.

I am deeply indebted to Jorge J. E. Gracia, who spent a whole weekreading the penultimate version of the manuscript and who provided detailedcomments. I am truly fortunate to have a supportive and passionate colleaguelike him. I am also grateful to Kenneth Dorter, who greatly helped when Iworked on the first version of this project. Many other colleagues and friendshave read chapters of this book at various stages of its development orversions of papers which have become parts of this book. For their help andencouragement, I thank David Bostock, Lesley Brown, Nicholas Bunnin,Roger Crisp, Michael Frede, John Keams, Gareth Matthews, ChristopherTaylor, and Wang Zisong. Ingrid Krabbenbos of Kluwer Academic Publishersdeserves special thanks for her patience and encouragement.

Jonathan Sanford, Marcus Marenda, and Mark Niswonger, as graduateassistants at different stages of this work, provided valuable assistance. Whenthe manuscript was accepted, the Julian Park Fund at the College of Arts andSciences, State University of New York at Buffalo, generously provided me agrant to cover the costs of the proofreading and the preparation of the indices.I am grateful to Gerol Petruzella for his careful and critical review of the finalversion and his skillful compilation of the indices.

ix

Abbreviations of Aristotle's Works

APo.APr.De.Cat.DA.D/.EE.EN.GA.Ge.HA.MA.Meta.Meteor.PA.Ph.PN.Pol.Pn.Rh.SE.Top.

Analytica PosterioraAnalytica PrioraDe CaeloCategoriaeDe AnimaDe InterpretationeEthica EudemiaEthica NicomacheaDe Generatione AnimaliumDe Generatione et CorruptioneHistoria AnimaliumDe Motu AnimaliumMetaphysicaMeteorologicaDe Panibus AnimaliumPhysicaParva NaturaliaPoliticaProtrepticusRhetoricaSophistici ElenchiTopica

Aristotle's works are cited in standard form: abbreviated title, book. chapter,Bekker page, column, and line. Except where it is otherwise stated, quotations ofAristotle are from The Revised Oxford Translations of The Completed Works' ofAristotle (abbreviated as "ROT"). When a technical term is the subject of thediscussion but is open to various translations, I leave it untranslated both inquotations and in discussions. All necessary Greek is transliterated andexplained.

xi

Introduction

In his Metaphysics, Aristotle claims that he is seeking to establish ascience of being. Being, at the most general level, is divided by Aristotle intothe following four types:1. Accidental being2. Being as truth3. Potential/actual being4. Per se being l

Per se (kath hauto) being can also be translated as "being in its own right" or"intrinsic being". This type of being has been referred to by Aristotle indifferent ways. The list of per se beings includes substance, quantity, quality,place, time, etc., and this is also the list of categories. At Meta. ix.l, 1045b28Aristotle calls this list the "categories of being" (hai kategoriai tou ontos). AtMeta. vi.2, 1026a36 and ix.1O, 1051a33-b2 per se being is called "being withreference to the figures of predication" (ta schemata tes kategorias, or "figuresof categories,,).2

Of these four types of being, accidental being is briefly treated in Meta.vi.2-3 and there Aristotle claims that the study of accidental being can bedismissed on the grounds that accidental being is indeterminate and cannot bethe object of knowledge.3 He also does not pay much attention to being as truthand treats it briefly in two short texts: Meta. viA and ix.1O. His reason is thatbeing as truth depends on combination and separation in thought but not inthings, and hence ''that which is in this sense is a different sort of being fromthe things that are in the full sense (for the thought attaches or removes eitherthe 'what' or quality or quantity or one of the other categories), and that whichis accidentally and that which is in the sense of being true must be dismissed."(viA, 1027b30-34) The science of being, then, is mainly concerned with per sebeing (or categorial being) and potential/actual being.

1 Meta. v.7. cr. alsoMeta. vi.2, 1026a34-b2, ix.10, 1051a33-b2.2 The relation between being and category will be discussed in Chapter I, section I. I shall use theterm "categorial being" interchangeably with "per se being".3Meta. vi.2,1026b4, vi.2, 1027310, viA, 1027b17.

xiii

xiv THE STRUCTURE OF BEING IN ARISTOTLE'S METAPHYSICS

According to the structure of the Metaphysics as we have it, the discussionof these two types of beings occupies the central books of the Metaphysics:Meta. vii, viii, and ix.4 Aristotle himself indicates explicitly that these twokinds of beings are the subject-matter of these central books:

We have treated of that which is primarily and to which all the other categories ofbeing are referred - i.e., of substance.... Since being is in one way divided into thisthing, quality, and quantity, and is in another way distinguished in respect ofpotentiality and fulfillment, and of function, let us discuss potentiality andfulfillment. (Meta. ix.I, 1045b28-35)

This passage singles out per se being and potential/actual being from the four­fold division of being, and indicates that these two types of being belong toseparate investigations.

The central books of the Metaphysics have been regarded as the mostmature and fundamental texts, but they are notoriously difficult. Indeed, almosteverything said in these books is open to various interpretations and debates,and the scholarship on these books is marked by numerous controversies anddebates.5

However, although these books have been the subjects of intensivediscussions, the distinction between per se being and potential/actual being hasbeen commonly treated as something verbal. F. Brentano representativelyexpresses this widely-shared indifference toward this distinction: 'They [per sebeing and potential/actual being] have in common that the science of being,metaphysics, is concerned in the same way with one as with the other.'06Commentators usually say that Meta. vii and viii deal with per se being, and ixwith potential/actual being.7 Yet, to divide the texts in this way is a matter ofconvenience in exposition rather than a division of any philosophicalsignificance. Even this expository division can be seen as an indication of howtrivially the distinction has been treated. For it has been noted that the mostevident difference between Meta. vii and viii is that potentiality and actuality

4 Some commentators prefer to use Z, H, and e to refer to these three books respectively. For thesake of consistency, I modify the references in relevant quotations.5 J. Barnes summarizes the situation in this way: "The argument is tortuous in the extreme, and it isfar from clear what Aristotle's final views on the subject are-if indeed he bad any final views."(1995,67) M. Loux in the preface to his Primary Ousia remarks: "Having had at least ten differentopinions on virtually every sentence I am not naive enough to suppose that the views I express herewill be my final ideas of vii and viii." (1990, xi)6F. Brentano, 1975,27.7 The following comments of W. D. Ross on the treatment of the four types of beings in theMetaphysics are routinely followed among the commentators: "Of the four senses of 'being'mentioned there [Meta. v.7], ta kala sumbebikos [accidental being] is briefly discussed in vi.2 and3, to hOs alithes [being as truth] in vi.4; to kota ta schemata tis kotigorias [being with respect tothe figures of predications] or rather substance, the first category, is discussed in vii and viii, and tokota dunamin koi entelecheian [potentiality and actuality] in ix." (1924, i, 358)

INTRODUCTION xv

are the main subject-matter of viii, but not of vii.8 If the distinction between perse being and potential/actual being is taken as a serious one, one might wonderwhether it could still be maintained thatMeta. vii-viii are, as a unity, concernedwith per se being, whileMeta. ix is about potential/actual being.Consequently, so far as I am aware, it has never been taken to be the case

that in these central books there are two separate and distinct discussions: onebelonging to the study of per se being, and the other to the study ofpotential/actual being. Rather, the assumption that is often made is that in thesecentral books, the issues raised and discussed in Meta. vii are further explainedand developed in Meta. viii and ix. Furthermore, books viii and ix areconsidered to be supplementary to or some development of the argument ofvii.9 Accordingly, these central books have been thought to constitute acomplex but single argument, and the doctrines in all three books are taken toaddress the same question.What I try to do in this book is to explore the significance of the

distinction between per se being and potentiaVactual being in Aristotle'smetaphysics. I will show that Aristotle is utterly serious in setting apart thesetwo types of being and in claiming that they need separate discussions. Hismetaphysics of being, in its core, contains the following two studies:

1. the investigation of per se being which is about the basic constituents ofreality and is related to predication, categories, and definition, and2. the study of potential/actual being which deals with the motion, process, andfunction of the world, as motion is defined by Aristotle as ''the actualization ofthe potentiality as such".

These two studies are of course connected, yet each has its own perspective andagenda. It is not accidental that he never puts potentiality and actuality on thelist of categories. As I shall argue, the distinction between per se being andpotential/actual being suggests that Aristotle views the world in two ways:statically and dynamically. Indeed, in his discussion of the soul, Aristotle

8As W. D. Ross says: "The expressions potentiality and actuality, almost entirely absent in vii, playa considerable part in viii." (1924, i. cxxiv) The same observation is also made in the Notes On Etaand Theta: "while a concern with substance as actuality is the mark of viii, it is conspicuously not amark of vii." (M. Burnyeat et al., eds., 1984,3)9 The dominance of this position can be seen from W. Jaeger's remark made earlier in the 20thcentury that the discussions of substance "begin with vii; then came viii; viii was presumablyfollowed by ix, as now" [1948/1923, 201-2], and from T.H. Irwin's comment that viii and ix"explain some of the major claims in vii" (1987, 207). M. L. Gill holds a position that there are twostages of development in these central books and that much said in book vii is denied later.Nevertheless, she insists that "Aristotle's works can be approached as an intelligible whole and thathis systematic of cross-references is our most reliable guide in deciding how the treatments relate toone another." (1989, ]]4)

xvi THE STRUCTURE OF BEING IN ARISTOTLE'S METAPHYSICS

himself expresses clearly that he has these two ways of approaching the studyof the soul:

First, no doubt, it is necessary to determine in which of the summa genera soul lies,what it is, is it 'a this-somewhat', a substance, or is it a quale or a quantum, or someother of the remaining kinds of predicates which we have distinguished? Further,does soul belong to the class of potential existents, or it is not rather an actuality?Our answer to this question is of the greatest importance. 10

By exploring the implications of this distinction, I seek to develop analternative understanding of the program that Aristotle pursues in theMetaphysics.

Given that these two types of being are discussed in the central books ofthe Metaphysics, if the distinction of these two types of being is serious, we areled to think that these books contain two, rather than one, doctrines of being.Since substance is primary being and the question of what being is isinvestigated in terms of what substance is in these central books, we would liketo think that there are two theories of substance in them: one takes substance asa per se being, and the other takes it as a potentiaVactual being. Furthermore,since in these central books substance is the composite of form and matter andis discussed in terms of the matter/form relation, it should be the case that thematter/form relation is treated in two different ways:

1. One belongs to the study of per se being in which the matter/form relation isnot associated with the potentiality/actuality relation.2. The other belongs to the study of potentiaVactual being in which thematter/form relation and the potentiality/actuality relation are interwoven.

The bulk of this book seeks to establish and justify that the distinction betweenthese two kinds of being runs through Aristotle's theory of substance in thesecentral books.

In my reading, a demarcation of these two studies of being in the centralbooks is Meta. vii.17, a text which starts with an announcement that ''weshould say what, and what sort of thing, substance is, taking another starting­point." (l041a6-7) As most commentators do not take seriously the distinctionbetween per se being and potentiaVactual being, they also dismiss vii. ITsclaim of a new start and view it as a mere figure of speech. However, there arestrong reasons to believe that this claim is seriously and should be takenaccordingly. I shall argue that the approach initiated in vii.17, which takes formas a formal cause, is continued in books viii and ix, but is contrasted with the

10 DA. i.l, 402313-7.

INTRODUCTION xvii

main argument in vii.3-I6. Accordingly, I submit, the three central booksshould be divided into two groups:

1. vii.3-I6, which belong to the study ofper se being, and2. vii. 17, viii and ix, which belong to the study of potentiaVactual being.

Each has its own distinct set of issues to discuss. Vii.3-I6 is concerned withwhich element of substance-form, matter, or the composite-is primarysubstance and with the universality and particularity of primary substance. Incontrast, vii. 17, viii, and ix deal with the issue of how a sensible substance canbe generated and unified and how it functions. The theory of substance in thisgroup of texts is neither supplementary to nor a different stage of the argumentof the same theory that is investigated in vii.3-I6. The common practice oflumping them together is probably one major reason why Meta. vii-ix are socontroversial. To separate them in the way that I recommend will help us to seethe internal agenda of each group of texts, and also to set many disputed issuesin different perspectives, which I will try to present when we move on.To treat seriously Aristotle's distinction between per se being and

potentiaVactual being also sheds new light on the relation between Metaphysicsvii and the Categories. In the Cat. the concrete particulars are said to beprimary substances, whereas in Meta. vii form is the claimant to the title ofprimary substance. How to explain this inconsistency has been one of thecentral issues in understanding Aristotle's metaphysics. My view is that thesetwo texts belong to two levels of discussion in the study of per se being. Theyare complementary and do not conflict. The Cat. deals with the relationbetween substance and other categories, while Meta. vii mainly deals with therelation between form, matter, and the composite--a relation that is within thecategory of substance itself. However, the approach to reality in these two textsis similar. Both seek to distinguish different elements and to determine whichone really exists in virtue of itself (per se). The Cat. divides beings intodifferent categories of beings and argues that substance is the mostfundamental, while Meta. vii divides substance into form, matter, and thecomposite and takes pains to determine that form is primary substance (being)among these. Further, as the Cat. distinguishes between universality andparticularity within each categories, and in particular within the category ofsubstance, Meta. vii also tries to determine whether form is particular oruniversal. It is true that in Meta. vii substance is divided into form, matter, andthe composite, a division which was missing in the Cat. But this does not entailthat the theory in the Cat. is denied or replaced in Meta. vii: Meta. vii.I doesnot mention the division of form and matter either, and its discussion of therelation between substance and the other categories is continuous with Cat. 5.Hence, Meta. vii should be seen as presupposing the essential truth of the

xviii THE STRUCTURE OF BEING IN ARISTOTLE'S METAPHYSICS

theory of the relation between substance and the other categories in the Cat. Itproceeds to discuss the per se being of substance itself in terms of the relationbetween form and matter.A serious treatment of the distinction between per se being and

potential/actual being also opens a new avenue to understanding the relationbetween the Metaphysics and the Physics. Since the concepts of potentialityand actuality are related to motion and change, the study of potential/actualbeing discussed inMeta. viii and ix are continuous, not with Meta. vii, but withthe Ph. In Ph. iii. I Aristotle defines motion in terms of potentiality andactuality, and in Meta. viii.l-2, when Aristotle starts to associate thematter/form relation with the potentiality/actuality relation, he immediatelyrefers back to the analysis of change in Ph. i. The concept of nature as the innersource of motion, which is the theme of Ph. ii., is explicitly said to be a kind ofpotentiality in Meta. ix.8:

And I mean by potentiality not only that definite kind which is said to be aprinciple of change in another thing or in the thing itself regarded as other, but ingeneral every principle ofmovement or of rest. For nature also is in the same genusas potentiality; for it is a principle of movement-not, however, in something elsebut in the thing itselfqua itself. (l049b5-1O)

Thus, the Ph. can be viewed as a part of Aristotle's discussion ofpotential/actual being.Accordingly, the distinction between per se being and potential/actual

being appears to inform the basic framework of Aristotle's metaphysicalthinking. No matter what the biographical sequence of his treatises is, themetaphysical doctrines in different treatises of Aristotle can be grouped undertwo headings: the study of per se being and the study of potential/actual being.Aristotle's doctrines within each heading are more sophisticated and completein some treatises than in others, and in this sense we can say that there is aprocess of development in his thinking. Nevertheless, his main concerns andprograms show no fundamental change. Hence, the issues of chronology are ofrelatively little significance for my discussion. It seems to me that the Meta. isthe most important and foundational metaphysical work of Aristotle preciselybecause in this work the substantial core is the basic framework of hismetaphysical thinking, that is, the distinction between per se being andpotential/actual being.

When I say that Aristotle's science of being contains two distinctinquiries, I am not thereby committing myself to endorsing the position that theMeta. is a disparate work. On the contrary, it is through a serious treatment ofthe distinction between per se being and potential/actual being that we are ledto an alternative understanding of the age-old debate regarding the relationbetween the science of being and theology in Aristotle. The science of being,

INTRODUCTION xix

however, contains both the study of per se being and the study ofpotential/actual being, since substance as primary being is the focal referent ofboth per se beings and potential/actual beings. The study of per se being doesnot concern motion and does not lead to the theory of the Prime Mover.However, Meta. vii.17, after announcing that it will start anew in investigatingsubstance, immediately claims that: "for perhaps from this we shall get a clearview also of that substance which exists apart from sensible substance."(l04la7-8) This implies that the new start will lead to theology. As I shallargue, Aristotle's theology belongs to the study of potential/actual being and isan intrinsic part of the science of being. The science of being contained in thecurrent Meta. is an unified enterprise because together these two studies ofbeings provide a complete picture of the world as both a static and a dynamicreality.This book is divided into eight chapters. Chapter I explains what per se

being is, what potential/actual being is, and what the distinction between thesetwo types of beings implies. It also argues that the focal structure of being,which is the subject matter of the science of being, includes not only per sebeings, as is usually thought, but also potential/actual beings.Chapter 2 shows the different ways that per se being and potential/actual

being are focally connected to substance. It reveals Aristotle's rationale thatsubstance is not only the focus of the study ofper se being, but also of the studyof potentiaVactual being. The chapter will also offer a different theory about therelation between the focal structure of being and the nature of the science ofbeing as being.Chapter 3 documents that there are two distinct approaches to the

matter/form relation in Meta. vii, viii, and ix, corresponding, respectively, toper se being and potential/actual being. It also argues that vii.17' s new start is areal demarcation between these two approaches.Chapters 4 and 5 interpret how the matter/form relation is discussed as a

study of per se being. In chapter 4, I first explain that Aristotle's criticism ofthe subject criterion in vii.3 suggests that the substance/attribute relation is of adifferent nature from the matter/form relation. I then argue that the discussionof form and matter in vii.3 and the discussion of essence in vii.4 are relatedbecause they represent two analyses of the individuals that are considered asprimary substance in the Cat. On such a basis I develop an understanding ofAristotle's rationale in identifying essence and form.

In chapter 5, I demonstrate that Meta. vii.3-l6, taken as a whole, impliesan aporematic discussion regarding whether primary substance is universal orparticular. This aporematic discussion is based on the dichotomy of tode ti ("athis") and toionde ("a such") and is presented as follows:

xx THE STRUCTURE OF BEING IN ARISTOTLE'S METAPHYSICS

(a) form is either "a this" (viL3's program), or "a such" (vii.7-9);(b) fonn cannot be "a this", for a this is not an object of defInition;(c) fonn cannot be "a such", for the universal is not a substance(viLB).Is fonn, then, universal or particular?

On my account, this aporematic structure results from Aristotle's response toPlato's Third-Man fallacy.Chapters 6 and 7 are concerned with Aristotle's study of substance as

potential/actual being. Chapter 6 demonstrates that Aristotle presents twodifferent types of substantial potentiality and actuality. One is substantialchange, in which potentiality and actuality are at different stages ofdevelopment, and the other is substantial activity, in which potentiality andactuality are different aspects of the same entity. Both types of relationship areimplied in the thesis of substantial unity in Meta. viii.6.Chapter 7 discusses how the study of potential/actual being leads to an

understanding of the Prime Mover. As the theory of potentiality and actualityexplains what motion is, the theory of the Prime Mover accounts for thecontinuity and eternity of motion; as the theory of potentiality and actualityexplains substantial unity, the theory of the Prime Mover accounts for the orderof the universe.Finally, in Chapter 8, I argue how the distinction between per se being and

potential/actual being helps clarify the relation between the science of beingand theology.