Tropes, Universals and the Philosophy of Mind 9783110327113, 9783110326703

Citation preview

Simone Gozzano, Francesco Orilia (Eds.) Tropes, Universals and the Philosophy of Mind Essays at the Boundary of Ontology and Philosophical Psychology

Philosophische Analyse Philosophical Analysis Herausgegeben von / Edited by Herbert Hochberg • Rafael Hüntelmann • Christian Kanzian Richard Schantz • Erwin Tegtmeier Band 24 / Volume 24

Simone Gozzano, Francesco Orilia (Eds.)

Tropes, Universals and the Philosophy of Mind Essays at the Boundary of Ontology and Philosophical Psychology

ontos verlag Frankfurt I Paris I Ebikon I Lancaster I New Brunswick

Bibliographic information published by the Deutsche Nationalbibliothek The Deutsche Nationalbibliothek lists this publication in the Deutsche Nationalbibliografie; detailed bibliographic data are available in the Internet at http://dnb.d-nb.de.

North and South America by Transaction Books Rutgers University Piscataway, NJ 08854-8042 [email protected] United Kingdom, Ireland, Iceland, Turkey, Malta, Portugal by Gazelle Books Services Limited White Cross Mills Hightown LANCASTER, LA1 4XS [email protected]

Livraison pour la France et la Belgique: Librairie Philosophique J.Vrin 6, place de la Sorbonne; F-75005 PARIS Tel. +33 (0)1 43 54 03 47; Fax +33 (0)1 43 54 48 18 www.vrin.fr

2008 ontos verlag P.O. Box 15 41, D-63133 Heusenstamm www.ontosverlag.com ISBN 978-3-938793-83-1 2008 No part of this book may be reproduced, stored in retrieval systems 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 of the purchaser of the work Printed on acid-free paper FSC-certified (Forest Stewardship Council) This hardcover binding meets the International Library standard Printed in Germany by buch bücher dd ag

Contents Tropes, Universals, and the Philosophy of Mind: Editors’ Introduction FRANCESCO ORILIA and SIMONE GOZZANO

7

Modes and Mind JOHN HEIL

13

Does Ontology Matter? ANNA-SOFIA MAURIN

31

Basic Ontology, Multiple Realizability and Mental Causation FRANCESCO ORILIA

57

The “Supervenience Argument”: Kim’s Challenge to Nonreductive Physicalism AUSONIO MARRAS and JUHANI YLI-VAKKURI

101

Tropes’ Simplicity and Mental Causation SIMONE GOZZANO

133

Zombies from Below DAVID ROBB

155

Tropes and Perception E. JONATHAN LOWE

175

About the Authors

195

Tropes, universals and the philosophy of mind: editors’ introduction In the current resurgence of interest in metaphysics that has followed the demise of logical positivism, the debate about the nature of properties has gained centre stage. There appear to be two main theoretical options, universalism and tropism, as we may say in short. According to the former, properties are viewed as universals, abstract entities that can be “shared” at various locations and moments, thereby characterizing different objects, while also accounting for their similarity. According to the latter, properties are tropes, abstract particulars that can characterize just one object at a time and can account for the similarity of distinct objects by their forming classes of “natural resemblance.” What does this have to do with the philosophy of mind? The past few years have seen the publication of works suggesting that our theoretical decisions at this foundational level may have farreaching consequences for issues in this discipline, including the contrast between reductive and nonreductive physicalism, multiple realizability and mental causation. To wit, there are recent papers, such as “The Properties of Mental Causation” by D. Robb (Philosophical Quarterly, 47, 1997, 17894; PMC hereafter),1 that show the existence of an ongoing controversy on whether one is better off resorting to tropism or rather to universalism in order to back up nonreductive physicalism; more specifically, in order to support the claim that nonreductive physicalism can supersede reductive physicalism in an effort to provide a materialist world-view that accommodates multiple realizability without presenting mental properties as causally irrelevant and mental events as causally inefficacious (thereby succumbing to epiphenomenalism). Moreover, since the 1970’s, the ontological controversy between Kim’s and Davidson’s conceptions of 1

See also, e.g., “Mental Causation, Determinables and Property Instances” by D. Ehring (Noûs, 30, 1996, 461-480), “The Ontological Turn”, by C. B. Martin and J. Heil (Midwest Studies in Philosophy, 23, 1999, 34-60), “Mental Properties” by J. Heil and D. Robb (American Philosophical Quarterly, 40, 2003, 175-196), “The Metaphysics of Mental Causation” by C. and G. MacDonald (The Journal of Philosophy, 103, 2006, 539-576).

8 events has always worked more or less explicitly behind the scenes of the debate on the nature of mental phenomena. In particular, Kim’s apparent inclinations over the years toward forms of reductive physicalism is at least in part explainable by his adherence to his own conception of events. Similarly, Davidson’s anomalous monism—a version of nonreductive physicalism—and the popularity of nonreductive physicalism in current philosophy of mind are hardly accountable without a Davidsonian view of events in the background. Although this is not often recognized, this dispute over events is tightly linked to the one between tropism and universalism, for, on the one hand, Kim’s conceptions of events presupposes an ontology with properties viewed as universals, and, on the other hand, Davidsonian events may be viewed as tropes (as explained in Orilia’s contribution to this volume). The idea of a book that explores connections between basic ontological issues and topics in the philosophy of mind along the lines outlined above came to these editors at the Seventh Meeting of the Italian Society for Analytic Philosophy, when one of us discussed in his presentation possible applications of tropism to the philosophy of mind2 and the other informed him that he had been working lately on similar topics. We thought that it could have been fun as well as rewarding to delve deeper into these matters together, by joining our efforts to those of other philosophers interested in exploring the repercussions in philosophy of mind of theoretical decisions at the level of basic ontology, where one discusses the nature of properties and events. The goal we set forth for ourselves was to provide a collection of essays that as a whole contributed to a deeper understanding of the extent to which theoretical decisions at the level of basic ontology—particularly about the nature of properties, whether they are tropes or universals, and the closely related issue of the nature of events—may have an impact on topics of interest to the philosopher of mind. We thus contacted some experts who have written very specifically on these issues and others whose works in either ontology or philosophy of mind indicated a potential interest in our project. We were happy to see that most of them warmly accepted our invitation to join this project. The result is the present collection of essays, each of which is briefly introduced below. 2

F. Orilia, “Universals, Tropes and Philosophy of Mind”, presented at Cervelli, persone e Società, VII Congresso Nazionale, Società Italiana di Filosofia Analitica, 28-30 September 2006, Università Vita-Salute S. Raffaele, Cesano Maderno (Milan).

9 John Heil’s “Modes and Mind” discusses extensively the way in which Robb relies on tropism in his PMC, in order to accommodate both a non-epiphenomenalist account of the mental and multiple realizability within a version of nonreductive physicalism. Heil admits that Robb indeed presents a solution to the conundrum of the apparent incompatibility of reductive physicalism and the causal relevance and efficacy of the mental—one that nicely makes room for multiple realizability. Nevertheless, Heil argues that this solution is not essentially tropist, as we may put it, for its recourse to tropes may be “simulated” in a parallel approach based on a form of universalism that is willing to recognize that a (mental) predicate may correspond, rather than to a single universal, to a family of different but similar universals. Anna-Sofia Maurin’s “Does Ontology Matter?” compares and contrasts Robb’s approach to the causal relevance and efficacy of the mental in PMC with another one based on tropism, namely the one defended by Martin and Heil in their “The Ontological Turn”, cit. She argues, contrary, as we have just seen, to Heil, that the former approach indeed depends essentially on its appeal to tropes for its account of mental causation. She contends, however, that the latter does not, for in it the real argumentative work is done by the rejection of the view that “language pictures reality”, the Picture Theory. According to Maurin, once the Picture Theory is set aside, it is immaterial whether one resorts to tropes or universals in trying to defuse the arguments that attempt to demonstrate that nonreductive physicalism leads to epiphenomenalism. Maurin then argues that the picture theory must indeed be replaced, but that this replacement cannot be of a radical variety. In order to preserve the possibility of doing ontology in a justified and not too speculative way, she concludes, ontological theorising must take place in a framework that preserves the possibility of drawing ontological conclusions from language. She ends her paper with a sketch of a “new Picture Theory”. Francesco Orilia’s “Basic Ontology, Multiple Realizability and Mental Causation” can be subdivided into two parts. The first one provides an outline of the ontological contrasts between universals and tropes on the one hand and Kim’s and Davidson’s conceptions of events on the other hand and explains how these contrasts are connected. Moreover, it discusses the opposition between reductive and nonreductive physicalism, how multiple realizability militates in favour of the latter and how epiphenomenalism constitutes a threat for it. This first part could thus be read as an introduction to the themes in ontology and the philosophy of

10 mind that play a crucial role in the more specific arguments in the rest of the paper and in the other papers. This is especially so if one adds to it section 2 of Gozzano’s contribution, which provides additional details on the nature of tropes. A reader who is not particularly expert in these topics may want to consult the first part of Orilia’s paper and section 2 of Gozzano’s paper before dealing with the rest of the book. The second part of Orilia’s essay compares and contrasts three versions of nonreductive physicalism: one centred on tropism along the lines suggested by Robb in his PMC, one based on the version of universalism considered by Heil in his contribution to this volume (“Heil universalism”) and one that relies on a version of universalism that acknowledges “higher-order” universals and revises Kim’s identity condition for events (“monist universalism”). It is admitted that—pretty much as Heil urges—the choice between tropism and Heil universalism is immaterial for philosophy of mind in that both doctrines accommodate multiple realizability and the causal efficacy and relevance of the mental in perfectly parallel ways. Nevertheless, it is argued that choosing universals rather than tropes may still matter for philosophy of mind, since monist universalism—as opposed to tropism and Heil universalism—makes room for multiple realizability and the efficacy and relevance of the mental in a palpably different way, a way consistent with the contention that creatures with significant physical differences can still be in identical mental states. Moreover, monist universalism appears to invite a positive answer when it comes to the question of whether one has a special acquaintance with oneself in the way admitted by philosophers such as Chisholm or Russell (at some point of his career). Ausonio Marras’ and Juhani Yli-Vakkuri’s paper, “The ‘Supervenience Argument’: Kim’s Challenge to Nonreductive Physicalism”, addresses Jaegwon Kim’s most recent attempt—in his Physicalism or Something Near Enough (Princeton: Princeton University Press, 2005)—to show that nonreductive physicalism is committed to epiphenomenalism. The authors point out that Kim’s argument relies on an unstated assumption about the identity condition of events as “property instances”, and argue that this assumption could be seen as by itself implying that nonreductive physicalism—or at least a classic, “tokenidentity” version of it—is false. This paper thus provides a further illustration of how Kim’s hostility towards nonreductive physicalism presupposes certain commitments at the level of basic ontology. Marras and Yli-Vakkuri also consider how Kim could modify these

11 presuppositions by adopting a different identity condition for events, which the reader could usefully compare and contrast with the one underlying the “monist universalism” discussed in Orilia’s paper. Simone Gozzano’s “Tropes’ Simplicity and Mental Causation” provides a detailed account of how tropes are viewed by their supporters, an account which highlights in particular that tropes are taken to be essentially simple entities. Gozzano then argues that tropes, because of their simplicity, can hardly vindicate the causal efficacy of mental properties in the way implicitly suggested in a number of papers by Davidson’s anomalous monism and explicitly voiced by Robb in his PMC. The conclusion is that an ontology based on tropes is of no help in solving the problem of mental causation. The overall recommendation is to stay with a universalist conception of properties when facing the problem of the causal efficacy of the mental. David Robb’s “Zombies from Below” explores the possibility of zombies—creatures pretty much like us, except that they fully lack consciousness—from the point of view of the tropist ontology defended by him in previous works and much discussed, as can be seen from the above, in other papers of this collection. Robb however dwells on a feature of his approach that is neglected in such papers (as they focus on other matters), namely the idea that tropes can be at the same time qualitative and dispositional. In the light of this, after having distinguished different kinds of zombie, Robb argues that only some of them are possible, whereas others should be ruled out. As is well-known, whether zombies are possible or not is a hot issue nowadays in the philosophy of mind, as many arguments pro or against physicalism depend on it. Robb’s paper suggests that tropism (or at least a certain form of it) invites its own specific answers with respect to this issue. It remains to be investigated whether a supporter of universals may come up with similar results or not. Finally, E. Jonathan Lowe’s “Tropes and Perception” presents an Aristotelian four-category ontology wherein both tropes and universals find their place. Lowe defends his approach against recent attacks by Jerrold Levinson to the very idea of tropes and in particular responds to Levinson’s criticism to an argument from perception for the existence of tropes previously proposed by Lowe himself. Perception is of course a crucial topic for the philosophy of mind and Lowe’s paper argues that it is best accounted for by appealing to tropes. In sum, if we consider this collection as a whole, we see arguments in favour of the idea that certain stands taken at the level of basic ontology

12 make a difference for philosophy of mind, and arguments that militate against this idea, or at least that are meant to minimize its import. We leave to the reader the task of evaluating them. Whatever the judgement may be, we hope to have contributed to strengthen the conviction that, as Heil puts it in his essay for this volume, “we must pursue philosophy of mind in a metaphysically self-conscious way”.3

3

We wish to thank J. Heil, A. Marras and A.-S. Maurin for their suggestions regarding this introduction.

Modes and Mind JOHN HEIL Monash University and Washington University in St Louis [email protected] ABSTRACT. The traditional mind-body problem lives on as the problem of causal relevance: could a mental cause, qua mental, have a physical effect? The problem would resist formulation without appeal to properties—mental and physical properties. The difficulty is to understand how mental properties could make a difference in the physical world. It might seem that before we could solve this problem, we would need to ascertain the nature of properties. Consider two possibilities: (1) properties are what have traditionally been called modes or accidents and are nowadays called tropes; (2) properties are universals. Although there are good reasons to think that properties are modes, not universals, the distinction is largely irrelevant to recently proposed solutions to the problem of causal relevance.

1.

Metaphysics and philosophy of mind

The philosophy of mind is applied metaphysics. True, many philosophers of mind are concerned with empirical dimensions of mind: the implications of neuroscience and psychology for issues once thought to be exclusively the province of philosophers. If we see metaphysics as continuous with science, this is what you would expect. We can acknowledge continuity, however, and still recognize core metaphysical questions for what they are: metaphysical questions. Take the problem of mental causation or the related problem of causal relevance. Neither of these problems is to be solved by examining the practices of psychologists and neuroscientists and noting that investigators routinely invoke causal relations in explanations of mental phenomena. We are unlikely to find an interesting account of qualia or have a clear sense of the relation of the mental to the physical until we have a sustainable conception of properties and their place in the causal structure of the world. Is the philosophy of mind, then, hostage to metaphysics? Must we obtain agreement on central metaphysical theses before moving on to applications in the philosophy of mind? That would threaten to make philosophy of mind a hopeless enterprise. Problems in the philosophy of mind are difficult, but at least they are local. Broader metaphysical problems are all-encompassing. Perhaps we should cope first with the local

14 issues, then move on, or encourage others to move on, to the more general questions. An old adage has it that all politics is local. Maybe this is how it is with philosophy: local problems are solved, not from the top down, but by deploying local resources, advancing gradually toward more encompassing results: bottom-up, not top-down. This has a pleasing ring to it, but it distorts the philosophical predicament. If it is a mistake to expect progress in the philosophy of mind to depend on the deduction of truths about the mind from unassailable metaphysical axioms, it is no less a mistake to imagine that metaphysical truths about the mind are discoverable in isolation from broader metaphysical concerns. It need not be thought that we could provide insights into problems in the philosophy of mind only after we have a completed metaphysics. What we need is a satisfactory appreciation of the options and their respective costs and benefits. By focusing exclusively on the trees, we lose sight of the forest, the overall picture to which our theories commit us. Too often in the philosophy of mind we pick and choose metaphysical theses to suit parochial interests. And too often the theses we pick combine to yield an untidy package. In metaphysics, one thing leads to another. So long as we remain narrowly focused, it will be hard to appreciate the extent to which local theses can be at odds with one another or with broader theses with attractions—and liabilities—of their own. My suggestion is not that we replace philosophy of mind with unadulterated metaphysics. Rather, we must pursue philosophy of mind in a metaphysically self-conscious way. In discussing mental causation, for instance, we need to be sensitive to the metaphysics of causation generally. Does our account of mental causation mesh with what we want to say about mental properties? Can we see our account as fitting into a broader metaphysical scheme with plausible implications elsewhere? One danger of addressing local issues in isolation is that, in so doing, it is easy to take on board substantive metaphysical doctrines as unexamined “background” assumptions. Indeed, this strikes me as characteristic of much work in the philosophy of mind. Ask yourself, for instance, what assumptions concerning the nature of properties, causation, and modality proponents of the zombie possibility make and what work unstated premises do in their arguments. I do not say that, in making metaphysical assumptions explicit, problems in the philosophy of mind will go away. At the very least, however, the logical space of possible solutions to those problems becomes

15 better calibrated. Once we admit that in doing the philosophy of mind we are doing metaphysics, we place up for grabs theses that are otherwise likely to exert undue, because unappreciated, influence over our thinking. Our aim should not be patchwork solutions to the latest crisis, but the production of accounts exhibiting metaphysical clarity and power. 2.

Mental and physical predicates and properties

Many arguments in present-day philosophy of mind turn on the relation mental properties bear to physical properties, and the role of properties in causal transactions. The waters have been muddied by a tendency to introduce talk of properties in a casual way, and moving from this to heavy-duty ontological conclusions. Ontologically detached talk of properties enjoys a venerable history. The Episcopal Prayer Book, for instance, speaks of “God whose property is always to be merciful”. In so speaking are we ascribing a property to God, the property of always being merciful? Perhaps not. Perhaps this way of describing God is just a mildly poetic way of saying that God is always merciful. Some readers will wonder what distinction I have in mind. If it is true that God is always merciful, then it must be true that God has the property of always being merciful. But this is my point. A predicate can be truly applied to an object in virtue of that object’s properties, or these together with relations it bears to other objects, without its thereby being the case that the predicate names or designates a property. This last claim will strike Episcopalians about properties as senseless. We have a predicate applying truly and literally to an object, so we must have a property. But this is precisely what is at issue. If you are going to be ontologically serious about properties, if you are going to address the role of properties in causal transactions, for instance, you will want to distinguish (1) A predicate, P, holds of an object, o, in virtue of o’s properties. (2) A predicate, P, holds of an object, o, in virtue of o’s possessing a property possessed by every object to which P applies.

16 (1) requires only that predicates—or those predicates concerning which we are “realists”—apply to objects in virtue of ways those objects are. This is much weaker than (2), which says, in effect, that whenever a predicate applies truly to an object it does so in virtue of that object’s possession of a single property shared by every object to which the predicate applies. According to (2), predicates concerning which we are realists name properties. Clearly, (2) is appreciably stronger than (1). Although (1) could be satisfied by objects’ possession of any of a family of similar properties, (2) requires that objects satisfying a predicate possess the very same property. My concern for the moment is not that this might be taken to commit us to universals. You might understand (2) to imply that objects answering to the same predicate possess either (a) numerically the same property (a universal) or (b) exactly similar properties (tropes or modes). “The same” can be used in either sense. What concerns me here, is the idea that objects answering to a single predicate must be exactly similar in some respect. Respect talk is property talk. If you accept universals, you will reduce exact similarity to identity, but neither (1) nor (2) requires universals. It seems to me obvious that most of the predicates we employ in ordinary life and in the sciences fail to satisfy (2). If you take properties seriously, then objects will possess the same property (or share a property) only if they are exactly similar in some respect. But in most cases predicates correspond, not to a single property (or collection of exactly similar properties) but to families of similar properties. Consider the predicate, “is red”. This predicate applies truly to objects in virtue of those objects’ possessing any of a somewhat unruly family of similar properties. Some would disagree. “Red” is a determinable. Admittedly, when an object is red, it is red in virtue of being some determinate shade of red. But (it will be said) this just means that there must be two properties, being red and being this determinate shade of red. This is a nice example of the linguistic tail wagging the ontological dog. We have two predicates that we regard as applying truly to objects, so we must have two properties. A single predicate, “is red”, applies, so the objects must have a property in common. When we fail to find a unique physical property possessed by all and only red objects, we conclude that the property of being red must be a “higher-level” property, distinct from, but somehow dependent on, assorted “lower-level” realizing properties. Being red “supervenes on” or is “realized by” determinate shades of red.

17 Here I follow Descartes and my other Enlightenment heroes. Properties of spatio-temporal objects are “maximally determinate”. Being red is a matter of being some determinate shade of red, a matter of being red in some determinate way. It is true that something is red when it is some determinate shade of red. What goes for red goes for triangularity. Something is triangular in virtue of possessing determinate sides and angles. There is no property of triangularity, no property of trilaterality. An object is triangular in virtue of having three determinate angles; it is trilateral in virtue of having three sides of determinate length.1 Some will disagree. Sophie the pigeon pecks at a button because a red light has gone on, not because a light of a particular shade of red has gone on. The shelf support does its job because it is triangular, not because its sides are precisely this long. This seems to imbue the determinables, redness and triangularity, with causal powers distinct from the causal powers of the determinates that realize them (or on which they “supervene”). And causal powers are thought to be importantly related to propertyhood. Let me pause here to remind you of difficulties besetting the causal influence of “higher-level”, multiply realized properties. Considerations that lead us to view higher-level properties as distinct from their realizers are hard to reconcile with considerations inclining us to regard such properties as having roles in causal transactions. Distinctness and causal relevance pull in opposite directions. Many readers will be aware of attempts to reconcile causal relevance and distinctness, but I think I am safe in saying that none of these attempts has been regarded as especially successful.2 Rather than discuss any in detail, I would prefer to apply my diagnosis of the problem to show why the need for such accounts is grounded in the move from an Episcopalian conception of properties directly to an ontologically serious conception. Suppose I am right. Suppose that objects are red or triangular, not in virtue of possessing the property of being red or the property of being triangular, but in virtue of being some definite shade of red or being 1

Note that, if this is right, it is misleading to imagine that the properties, being triangular and being trilateral, covary of necessity. An object is triangular in virtue of possessing three determinate angles; it is trilateral in virtue of possessing three determinate sides. You can change the lengths of a triangle’s sides without changing its angles. 2 See, for instance Pereboom and Kornblith 1991, Yablo 1992, Pereboom 2002, Shoemaker 2003.

18 triangular in a definite way. The first thing to note is that this in no way implies an anti-realism about redness or triangularity. Objects really are red, really are triangular. They are red or triangular, not in virtue of possessing the property of being red or the property of being triangular, but in virtue of being some determinate shade of red or being triangular in some definite way. To think that this implies anti-realism is to conflate the Episcopalian conception of properties with an ontologically serious conception. If you start with the idea that predicates truly applicable to objects name properties, and combine this with the idea that there is no single property answering to “is red” or “is triangular”, the result is anti-realism. But if you reject property Episcopalianism, if you accept that many different, but similar properties answer to these predicates, you will not be tempted to see the denial that “is red” and “is triangular” name properties as implying anti-realism. But wait! What of the causal contribution of redness and triangularity? Surely this speaks in favour of the idea that these are genuine higher-level properties! Sophie the pigeon responds indifferently to lights of different shades provided these are similar enough to count as being red. To say that Sophie pecks because the light is red, and not because it is crimson, is not to say that Sophie’s pecking is caused by some colour property other than crimson. The light is crimson, and this is what she responds to. But it is also true of Sophie that, were the light to be scarlet or some other shade of red, she would peck. The same is true of our triangular shelf bracket. The bracket has sides of determinate length. But another bracket with sides of a different length would work as well. The suggestion on the table is that properties commonly taken to be higher-level, supervenient or multiply realized properties are Episcopalian. Predicates that seem to designate such properties, instead encompass families of similar properties. Once we dispense with property levels, we bypass the need for tortured accounts of the realizing relation and vexed questions about the causal relevance of higher-level properties. The truths remain untouched. What we give up is an ill-conceived philosophical thesis. One way to put this is to note that different truths can have a single truthmaker. On a particular occasion, the truthmaker for “the light is red” and “the light is crimson” can be the light’s being crimson. You can have a “sparse” conception of properties without abandoning any of the predicates we deploy in describing the world both in

19 scientific and everyday discourse. The only true substances might turn out to be the corpuscles, or the electrons and quarks, or the quantum field; and the only true properties might be properties of these. This does not mean, however, that there are no trees, planets, tables, or mountains. It does not mean that nothing is round or red, or in pain. We keep the predicates, and we allow that the predicates can apply truly. What we give up are shadows cast by those predicates when they figure in philosophical theories that accept the idea that each significant predicate must correspond to a property. None of these comments bears on the question whether properties are universals or tropes (or modes; see below). All of the above could be accepted by David Armstrong and by Locke or D. C. Williams. Nothing I have said thus far implies that ontological seriousness about properties requires a commitment to or a rejection of universals. The point is worth making explicit because philosophers often appear to regard a commitment to properties as tantamount to a commitment to universals: “property” is synonymous with “universal”. Scepticism about universals is seen as scepticism about properties. Sceptical arguments can then be deployed as reductios for contradictory positions: (i) “So, you don’t believe in universals? So you think nothing is red or square?” (ii) “Ah, you believe in properties? But then you must believe in universals, and that would be crazy!” But why assume that, if there are properties, they must be universals? One unhappy side-effect of this way of thinking is a tolerance for weirdness in ontology. The thought might be this. If you need properties, you need universals. But universals are weird. They are either abstracta or multiply-locatable concreta. As long as we have weirdness in our ontology anyway, why draw the line with properties? Why should the weirdness of a posit be regarded as a liability? The door is opened to every sort of ontological extravagance. 3.

Tropes and modes

The idea that properties are particulars is one on which Enlightenment philosophers, despite broad disagreement on other matters, generally agreed. Nowadays this same idea is regarded with deep suspicion. In part this is due to general misgivings about properties. Do we really need them? Isn’t an ontology that avoids mention of properties preferable to one that

20 does? Ockham’s Razor, the vaunted Principle of Parsimony, tells us to avoid unnecessary posits. What exactly is the role of parsimony in ontology? All things considered, a theory is to be preferred if it commits us to the existence of fewer kinds of entity than its competitors. The mistake is to invoke parsimony early in the game as a constraint on theorizing. Thus deployed, parsimony is no virtue but a manifestation a species of small-mindedness inimical to serious ontology. Parsimony has application in the endgame, not preemptively. I believe that there are good reasons to include properties in our ontology. But what are properties? Today many philosophers take for granted that properties, if they exist, must be universals. It was not always so. Descartes, Spinoza, Leibniz, Locke, Priestley, Berkeley, Hume, even Kant all included properties in their ontologies, but denied that properties are universals. They regarded properties as modes (from the Latin modus, way or manner of being), ways substances are.3 Nowadays, D. C. Williams’s term “trope” has replaced “mode” in the philosophical lexicon, but I shall continue to honour tradition and speak of modes in what follows.4 I prefer “mode” to “trope” because it honours philosophical tradition and because self-described “trope theorists” have tended to embrace onecategory ontologies. Objects, they contend, are bundles of tropes. I prefer the two-category ontology common in the Enlightenment. Modes are ways objects—substances—are. Modes are dependent; the identity of a mode is bound up with the identity of the substance of which it is a mode. Socrates’s whiteness is Socrates’s whiteness. Modes, on this reading, are “non-transferable”. Socrates’s whiteness cannot be detached from Socrates. I thus reject “trope transfer” accounts of causation. When one billiard ball strikes another, the second acquires momentum from the first. But this is not because a momentum mode or trope is exchanged. Bundle theorists regard modes as akin to parts—substantial parts—of objects. This strikes me as ill-conceived, but I admit I have no conclusive

3

Jerrold Levinson has informed me (in correspondence) that he characterized properties as “ways of being” in Levinson 1978 (see also Levinson 1980). My first encounter with the idea of properties as ways came in a lecture on Spinoza’s “vacuum argument” delivered by Jonathan Bennett at the University of Virginia at about the same time. See also Seargent 1985, ch. 4; Armstrong 1997, § 3.6. 4 E. J. Lowe is also partial to “mode”. See his 2006.

21 argument favouring my substance-mode ontology. Nothing I have to say here depends on my two-category picture, however. 4.

Modes and universals

My antipathy toward universals is grounded neither in an a priori argument as to their impossibility, nor a taste for desert landscapes. I admit that I don’t “get” universals. Once we abandon the idea that universals are molds that God uses to stamp out kinds of particular, I don’t understand what role Platonic, transcendent, universals could possibly have in the spatio-temporal world, what their existence could possibly explain. Nor do I understand how Armstrong-style, immanent universals could be wholly present in each of their spatio-temporally distinct instances. You could chalk this up to a failure of imagination. My more immediate worry concerns the role of universals in philosophical theorizing: I cannot see what advantage universals enjoy over collections of similar modes (see Heil 2005 for discussion). Proponents of universals tell us that objects behave, or would behave, similarly in virtue of possessing one and the same property. How is this an advance over the idea that objects behave, or would behave, similarly in virtue of possessing similar properties? In the case of the fundamental things, the similarity is precise. If similarity is understood, as I understand it, to include both dispositional and qualitative elements, then it is hard to see why we should need to posit a category of being that differs dramatically from particularity (which we need in any case).5 Is this a preemptive application of Ockham’s Razor of the sort I advised against earlier? I hope not. One of the morals of this discussion is that there is a kind of ontological parity between modes and immanent universals. For every move in an ontology friendly to universals is paralleled by a move in a mode-friendly ontology. If that is so, then it would seem that we have some reason to prefer modes to universals. Of course, all of this is controversial. Proponents of universals will insist that universals can explain similarity relations and provide a grounding for scientific practice unavailable to purveyors of modes.6 I side 5

Having expressed doubts about universals, I should point out that most of what I say here (and in Heil 2003), is perfectly compatible with the thesis that properties are universals. 6 See, for instance, Armstrong 1997, Ellis 2001, and Lowe 2006.

22 with my Enlightenment heroes in finding the charms of universals elusive (see Heil 2003, chaps. 12-14; Heil 2005). It is worth noting that these same heroes were major players in the Scientific Revolution. The thought that adequate accounts of the lawful structure of nature required universals was not one that would have occurred to them. I do not expect these assertions to win over lovers of universals. I will rest content with the observation that, if the postulation of universals provides no clear advantages, we are entitled to resist it. Meanwhile, I hope that I have said enough to cast doubt on the thought that, of course properties, if there are any, must be universals: any ontology that includes properties thereby includes universals. This thought makes a commitment to universals the default view; universals are the heavyweight camps, secure until decisively defeated. In philosophy there are no default views, no heavyweight champs, no positions we are obliged to accept in the absence of conclusive countervailing reasons. 5.

Fundamental ontology

Enlightenment philosophers did not argue against universals so much as get along without them. Something else they got along without was the idea that fundamental ontological facts about concrete objects could be extracted from linguistic facts. Yes, we regard trees, persons, and planets as substances, but we could be—and probably are—wrong in this. A tree for Descartes, Locke, or Spinoza, for instance, was not a substance but a mode: a way the corpuscles are organized or a distinctive kind of local “thickening” of space. The idea that language is a telescope through which we view the world was not on the table.7 The thought that our best, or in some cases only, route to objects and their properties is via language is a product of the twentieth century linguisticization of philosophy. Yes, we 7

The telescope metaphor is invoked by Timothy Williamson. “Philosophers who refuse to bother about semantics, on the grounds that they want to study the nonlinguistic world, not our talk about that world, resemble astronomers who refuse to bother about the theory of telescopes, on the grounds that they want to study the stars, not our observation of them. Such an attitude may be good enough for amateurs; applied to more advanced inquiries, it produces crude errors. Those metaphysicians who ignore language in order not to project it on to the world are the very ones most likely to fall into just that fallacy, because the validity of their reasoning depends on unexamined assumptions about the structure of the language in which they reason.” (Williamson 2006, 182.) Heather Dyke called my attention to this quotation.

23 talk about and describe objects. Yes, we use words to negotiate extralinguistic reality. But we also interact with objects in endless nonlinguistic ways. Take trees. We plant and cultivate them, stand in their shade, climb them, crash automobiles into them, chop them down, build furniture and dwellings from them, turn them into paper. The idea that working out the ontology of trees must be based on a close analysis of the tree concept appears misguided. I said that many of my Enlightenment heroes were prepared to deny that tables, trees, planets, and the like were substances. That does not mean that they were anti-realists about tables, trees, and planets. Such things exist all right, claims about them are often true—literally true. But the truthmakers turn out to be modes, ways the corpuscles are arranged or local disturbances in space. This outlook is hard for us to appreciate. We imagine that realism requires that the world line up with our ways of talking about the world. If you are a realist about the F’s, then the Fpredicate must designate a property shared by everything to which it applies and in virtue of which it applies. This condition is only rarely satisfied, however. As noted already, it appears that the only predicates that name properties are those pertaining to the fundamental things. “Mass of an electron”, for instance, appears to designate a property possessed by everything to which it applies and in virtue of which it applies. But consider an everyday predicate, “is red”. This predicate applies truly to many things, but not in virtue of designating a property shared by every object to which it truly applies. Rather “is red” is satisfied by any of a family of different, but similar properties. This is enough for realism about red. If you agree, you should consider that the same holds for “is in pain”. “Nonreductive physicalists” have wanted to insist that this predicate singles out a “higher-level” property multiply realized by a family of lower-level realizing properties. But why not think that what goes for red goes for pain? When you have a headache, you are in pain in virtue of possessing some property, P1. It is true that you are in pain. When you bark your shin it is true that you are in pain, although what makes it true is not your possessing P1, but your possessing some similar property, P2. The pains are different, just as shades of red are different, but they are similar enough to count as pains. A Martian’s pain, or an octopus’s might be more different still, but still similar enough to satisfy the pain predicate. This suffices for realism about pain. We can take the functionalist’s lead here

24 and identify truthmakers for applications of the pain predicate with what functionalists would regard as pain realizers. “Is in pain” is made true by any of a family of similar properties. Functionalists focus on causal similarity: the properties are similar with respect to their dispositionalities or “causal powers”. Others might focus on qualitative similarity. We have the sense that these two dimensions of similarity are at best contingently related: qualities and dispositions could vary independently. This is the inspiration for zombies. You and your zombie twin are dispositionally indiscernible, indiscernible with respect to your respective causal powers, but qualitatively discernible. I have argued at length against this picture of the relation of the qualitative and the dispositional (Heil 2003, 2007). My inspiration, again, is traceable to the Enlightenment. Locke, I contend, regards properties as powerful qualities. The identity of a property is bound up with its qualitativity and its dispositionality. These are not distinct higher-order properties, but the selfsame property, differently considered. I see a analogous picture in Spinoza. Spinoza and Locke differ on the question, how many substances there are. Locke says, “many”, Spinoza says, “one”. For our purposes this distinction is not a deep one. Where Locke sees arrangements of corpuscles answering to “is a tree”, Spinoza sees a local congealing of the One. In either case, trees are modes, not substances. According to Spinoza, the One has an infinite number of attributes. You could think of two of these attributes as qualitativity and dispositionality. A mode is a determinate way of having an attribute. Descartes tells us that extended objects must be extended in determinate ways. Being one-meter square, is one way of being extended, so is being Prussian blue.8 Going along with these qualities are determinate powers. How are the qualities and powers related? I have said that, for Locke, properties are powerful qualities. If we think of dispositionality and qualitativity as “aspects”, then for Spinoza dispositionality and qualitativity are strictly identical. I believe that there are excellent reasons to accept the identification of the qualitative and the dispositional (Heil 2007). My aim here, however, is not to convince you of this thesis, but merely to illustrate my contention that what goes on in fundamental ontology has broad implications for issues in the philosophy of mind. It is a bad idea to finesse deeper ontological questions when competing answers to these questions yield 8

Colours are “secondary” qualities, arrangements of primaries.

25 very different environments for pursuing answers to questions philosophers of mind like to ask. Similarly, it is no good taking substantive ontological doctrines as though these fell from the sky. It is just possible that many of the vexed questions about mental causation and the like stem, not from special features of the mind, but from unexamined conceptions of properties generally or the nature of causation. 6.

Robb on mental causation

I have thus far skirted the question whether appeal to modes—tropes— promises resolution to problems arising in the philosophy of mind. Do modes afford resources not available to those who regard properties as universals? I doubt it. I suspect that, for every theoretical move made by proponents of modes, there is a corresponding move open to devotees of immanent universals. The point can be illustrated by considering David Robb’s elegant trope-based solution to the problem of mental causation (Robb 1997). Robb considers types to be collections of similar tropes. These collections would correspond to Armstrong-style immanent universals, the difference being that, where Armstrong sees identity, Robb sees similarity. A given mental type is made up of collections of distinct physical types. Thus, any token of a mental type is a token of some physical type. The idea is that mental predicates are satisfied by broadly similar tropes that can belong to distinct physical types. Physical predicates encompass narrower similarities. Think of trees and think of oak trees, elm trees, pine trees. Every tree is going to be an oak, or an, elm, or a pine, …. Robb’s idea is illustrated in the diagram below. Tropes are represented by dots. M1 is a mental type, a collection of similar tropes. P1, P2, P3, …, Pn, are physical types, sub-collections of similar tropes. Think of physical categories as typing tropes more tightly than the mental. At the limit, physical types will consist of collections of exactly similar tropes.

26

Mental and Physical Types

Mental and Physical Types

Now, reading “type” as “property”, suppose you have a given mental property: you are in pain. Could your having this property be “causally relevant” to what you do? Well, you have the mental property in virtue of, having some physical property, some trope typed physically. Take away the physical property and you take away this instance of the mental property. Compare: this tree is a tree in virtue of being an oak; remove the oak, and you’ve removed the tree. Now supposed that the physical trope— the trope typed physically—figures in the production of some physical effect. Did the effect ensue in virtue of the trope’s being mental or in virtue of its being physical? This is the notorious qua problem. Bear in mind that types are collections of tropes, not collections of objects. A red ball might roll in virtue of its sphericity, not in virtue of its being red. But the ball is an object. It is unwise to transfer object-talk to talk of properties or tropes. The ball rolls or would roll in virtue of its sphericity. Its sphericity doesn’t contribute to the rolling in virtue of some further property. This is so in spades for Robb’s case. Types are collections of similar tropes. One and the same trope can be a member of many different collections. (Compare: a given object can be a plant, a tree, an oak tree, a white oak tree.) It makes little sense to ask whether a given trope is efficacious in virtue of being typed one way rather than another. 7.

Is the trope solution a trope solution?

Robb’s account of mental causation affords a tidy solution to the problem of causal relevance, and does so in a way that makes sense of the kinds of phenomena often brought forward in defence of the “multiple realizability” of mental properties. We needn’t imagine that mental properties are “higher-level” properties, dependent on, but distinct from, their realizers.

27 Although mental types are not identical with physical types, every mental cause is a physical cause. The question is, to what extent does an account of this kind require a commitment to tropes? Robb’s types are collections of tropes. Collections of tropes are hard to distinguish from Armstrong’s universals. This is not quite right. Robb’s mental types correspond to collections of similar universals. Neither Armstrong nor Robb would accept the idea that there is a one-one predicate-property correspondence. Some predicates do designate properties possessed by every object to which they apply and in virtue of which they apply, but most do not. Most predicates that apply truly to objects do so in virtue of those objects’ possession of any of a (possibly open-ended) family of properties. Such a predicate holds true of objects not in virtue of those objects’ possession of a unique property (if properties are universals) or exactly similar properties (if properties are tropes), but in virtue of their possession of similar properties. For Armstrong, not every predicate designates a universal; for Robb, not every predicate designates a collection of exactly similar tropes. For Armstrong, P1, P2, P3, …, Pn in Figure 1 represent universals (dots are instances), and M1 represents a collection of similar universals. We can, then, replace tropes with universals without affecting Robb’s solution to the problem of mental causation. Physical types, on this modified view, will be physical universals; mental types will be collections of similar physical universals. The really interesting move is not the move from universals to modes or tropes, but the recognition that mental predicates can be satisfied by members of families of distinct physical properties (whether tropes or universals). If you thought of physical properties as properties answering to physical predicates and mental properties as properties satisfying mental predicates, then you would have something close to anomalous monism—properly construed. You do not need universals to do ontologically serious philosophy of mind, but neither do you need modes or tropes to provide an ontologically serious solution to the mind-body problem. What is important is ontological seriousness about properties. You could think of Armstrong and Robb as agreeing on every substantive point, then diverging. What Robb calls collections of exactly similar tropes, Armstrong will call instances of a single universal. If you are like me and have trouble understanding universals, understanding how something could be wholly present at different places at the same time, you will wonder whether this might not be all there is to Armstrong-style universals. In that case, the

28 difference between Robb and Armstrong, at least on this point, is merely verbal. 8

Concluding remark

You should not imagine that I consider the question whether properties are modes or universals to be purely notational. Rather, I am illustrating my earlier contention that philosophers of mind must address fundamental metaphysical issues. This does not mean that, before engaging in the philosophy of mind, you must have at your disposal a fully articulated metaphysical system from which theorems in the philosophy of mind could be derived. You do need a strong sense of the territory and how apparently unrelated claims knit together at a deeper metaphysical level. It turns out that the question whether properties are modes are universals is one the answer to which has less bearing than you might think on questions that have bedevilled philosophers of mind. Still, it is useful to be aware that there is more than one way to understand properties, even if only to recognize that a commitment to properties need not bring with it a commitment to universals. As for me, I go with Locke: “All things that exist are only particulars” (Locke 1690, III, iii, 6). Universality is not something in the world but a feature of our way of representing the world. You can have a thought of dogs generally, but I follow Locke in taking this to be a matter of your deploying a symbol in a way that encompasses any dog at all. The symbol might be the image of a particular dog, or an auditory image of an utterance of “dog”. Generality is a product of our use of particulars. The idea that, corresponding to general terms there must be general entities, comes from a failure to appreciate the dynamics of use. I mention this last point only by way of suggesting that there are topics of interest to philosophers of mind concerning which the question whether properties are modes or universals finds purchase. An impulsive commitment to universals hamper progress in our understanding of intentionality. I could be wrong. But the mistake would be to foreclose promising options. REFERENCES Armstrong, D. M. (1997). A World of States of Affairs. Cambridge:

29 Cambridge University Press. Ellis, B. (2001). Scientific Essentialism. Cambridge: Cambridge University Press. Greenough, P and Lynch, M. P., eds. (2006). Truth and Realism. Oxford: Clarendon Press. Heil, J. (2003). From an Ontological Point of View. Oxford: Clarendon Press. Heil, J. (2005). Kinds and Essences. Ratio, 18, 405-19. Reprinted in Metaphysics in Science, ed. A. Drewery, Oxford: Blackwell, 2006, 3346. Heil, J. (2007). Précis of From an Ontological Point of View. SWIF Philosophy of Mind Review, URL: . Levinson, J. (1978). Properties and Related Entities. Philosophy and Phenomenological Research, 39, 1-22. Levinson, J. (1980). The Particularization of Attributes. Australasian Journal of Philosophy, 58, 102-15. Locke, J. (1690⁄1978). An Essay Concerning Human Understanding, ed. P. H. Nidditch. Oxford: Clarendon Press. Lowe, E. J. (2006). The Four-Category Ontology: A Metaphysical Foundation for Natural Science. Oxford: Clarendon Press. Pereboom, D. (2002). Robust Nonreductive Materialism. Journal of Philosophy, 99, 499-531. Pereboom, D. and H. Kornblith. (1991). The Metaphysics of Irreducibility. Philosophical Studies, 63, 125-45. Robb, D. M. (1997). The Properties of Mental Causation. Philosophical Quarterly, 47, 178-94. Ryle, G. (1949). The Concept of Mind. London: Hutchinson. Seargent, D. A. J. (1985). Plurality and Continuity. Dordrecht: Martinus Nijhoff Publishers. Shoemaker, S. (2003). Realization, Microrealization, and Coincidence. Philosophy and Phenomenological Research, 67, 1-23. Yablo, S. (1992). Mental Causation. Philosophical Review, 101, 245-80.

30 Williamson, T. (2006). Must Do Better. Greenough and Lynch 2006, 17778.

Does Ontology Matter? ANNA-SOFIA MAURIN Lund University [email protected] ABSTRACT. D. Robb, on the one hand, and C. B. Martin and J. Heil on the other, have argued that the problem of the causal relevance of mental properties (dis)solves when properties are treated as tropes rather than as universals. Though similar in appearance, their positions differ importantly. Robb relies on the notion that properties are tropes, but close scrutiny of their position suggests that Martin and Heil do not. Martin and Heil reject the Picture Theory (the view that language pictures reality). I argue that this rejection is sufficient in itself to (dis)solve the problem of mental properties’ causal relevance. The Picture Theory should indeed be rejected; but if it is, not only can the argumentative role that tropes are supposed to play in this context no longer be offered as a reason for accepting an ontology of tropes rather than universals; many other reasons often cited in our attempts to justify one ontology over another must be given up as well. If ontological conclusions are to be justified, the old Picture Theory must be replaced by a new and better one. The paper ends with a sketch of such a theory.

1.

Introduction1

In his seminal paper “On the Elements of Being” D. C. Williams argues for the existence of tropes partly on the basis of what he takes to be their exceptional explanatory power. With recourse to tropes, he exclaims, ageold problems pertaining to mind, knowledge and morality will be solved or dissolved, because (1953, 16)2

1

I would like to thank the participants at the 2007 Metaphysics of Science workshop in Lund (Johannes Persson in particular), my audience at the 2007 National Conference in Philosophy, and this book's editors (especially Francesco Orilia) for helping me to improve on this paper, both in form and content. A very special thanks to Johan Brännmark for his many insigthful comments on and objections to earlier drafts of this paper. 2 Very roughly (a rough characterisation will suffice for the purposes of this paper) a trope is an abstract particular. Contemporary trope theorists include D. C. Williams (1953), C. B. Martin (1980), K. Campbell (1990), P. Simons (1994), J. Bacon (1995), J. Heil (2003) and myself (Maurin 2002).

32 the trope … provides the one rubric which is hospitable to a hundred sorts of entity which neither philosophy, science, nor common sense can forego.

Contemporary literature on tropes is littered with similar exclamations, yet they are rarely accompanied by detailed explanation of exactly how tropes are supposed to contribute to the (dis)solution of this or that problem. In this regard the treatment of tropes in relation to some issues in the philosophy of mind represents an exception, and in this paper I take a closer look at how tropes have been employed to solve one such issue: that of the causal relevance of mental properties.3 If properties are tropes, it has been repeatedly maintained, this problem (at least) will go away. More precisely, I propose to investigate what I take to be two interestingly different suggestions to this effect; one formulated by David Robb (1997), the other by C. B. Martin and John Heil (1999). I am not at this point concerned with establishing that these suggestions manage to successfully (dis)solve the problem at hand; I simply assume that they do, and that they do it well. My central question concerns, instead, what difference tropes make to this success. I will argue that in order to be successful, Robb’s position essentially depends on properties being tropes, whereas, on the approach taken by Martin and Heil, appearances perhaps to the contrary, tropes do no real argumentative work. Whether our particular ontological choices make a difference to what conclusions can be drawn in the philosophy of mind (and elsewhere) depends, I shall maintain, on one’s stand on broader (meta)metaphysical issues.

3

The treatment of tropes in relation to issues in the philosophy of mind has more or less co-evolved with contemporary trope theories generally. Important contributions have been formulated by C. B. Martin, John Heil and David Robb (cf. Martin and Heil 1999, Robb 1997, Heil 2003 and Heil and Robb 2003). Interesting and important metaphysical (although not necessarily trope-theoretical) discussions of the problem can also be found in Bennett 2003, Ehring 1997, Gibb 2004, Kim 1999, Lowe 2003, Macdonald and Macdonald 1986, 1995 and S. Yablo 1992. This list could be extended.

33 2.

The problem4

Suppose that Lin is in pain, and that as a result she quickly removes her hand from the hot stove. It seems like there are at least two ways in which what happens to Lin can be described: Lin’s being in pain caused her to remove her hand from the hot stove. Lin’s C-fibres firing caused her to remove her hand from the hot stove. Ignoring the possibility that these are descriptions of two entirely distinct causal processes, we ought to claim that these seemingly competing descriptions differ only in the sense that each involves a different property of the cause. This claim is certainly less contentious than a substantially dualist one but is by no means completely innocuous. Just as Lin’s fate can be described in two ways, it can be (causally) explained in two ways. Two equally plausible and informative answers can be given to the question: why did Lin remove her hand from the hot stove? Because she felt pain. Because the C-fibres in her brain fired. The fact that each of these explanations refers to a different property of the cause seems to entail that both the mental and the physical property of the cause are relevant to the production of the effect.5 It is hard to spell out exactly what it means to say of a property that it is “causally relevant” (a fact that may be responsible for some confusion in the debate). For the purposes of this paper, however, a rudimentary understanding of the notion will suffice. Braun offers just what we need (1995, 447): 4

There is more than one “problem of the causal relevance of mental properties”. The problem concerning us here is sometimes referred to as the “exclusion problem” to set it apart from its cousins. For an overview of the problems, see Bennett 2007 and Robb and Heil 2003. 5 Discussions of the causal relevance of mental properties have their roots in criticisms of Donald Davidson’s anomalous monism (see his 1980). They rest on the following assumption: events do not simply cause their effects; they cause them qua mental or qua physical. Horgan (1987) says that they quause them.

34 A causally relevant property is a certain kind of property of causes. Every cause of an effect has many different properties. But many properties of a cause appear to have no role in causing its effect. Those properties of the cause are causally irrelevant to the effect. The properties of a cause that do have some role in causing its effect are causally relevant to the effect.6

A plausible, and intuitively appealing, conclusion seems to follow from the preceding observations: Relevance: Mental properties are (sometimes) causally relevant. Unfortunately, however, when it is combined with two further claims that are, similarly, independently plausible, Relevance lands us in a troublesome contradiction. These additional claims are: Distinction: Mental properties ≠ Physical properties. Closure: Every physical event has a sufficient physical cause. Distinction is plausible because mental properties are multiply realisable. The proposition is true, when it is, because the predicatephrase “is in pain” picks out some property of a neural event in the brain of the individual named “Lin”. Since the same predicate can also be applied truly to creatures with a physical constitution that differs from Lin’s, being in pain cannot be identified with being in a particular physical state. Mental properties, it is argued, must therefore be higher level properties, and so Distinction must be true. Closure, next, is justified by the success

6

Braun gives an example: a soprano sings the word “shatter” at a high pitch, thereby causing a nearby glass to shatter. It is in virtue of its pitch (the causally relevant property) but not in virtue of its meaning (the causally irrelevant property) that the sung word causes the glass to break. Braun does not rest content with the “explanation” of causal relevance cited above; he goes on to spell out his own, much more substantial, understanding of the notion (one that I will not examine here, however).

35 and practice of science and needs, if the argument is to go through, to be understood as barring systematic overdetermination.7 Pretty clearly, these three claims are inconsistent. If mental properties are distinct from physical properties, and if physical properties are what characterise the sufficient cause of every physical event, then, barring systematic overdetermination, mental properties have no real job to do in causation.8 Relevance must be given up. The trope strategy consists in accepting all three claims, yet denying that their combination is inconsistent. If mental properties are tropes, it is maintained, they can be causally relevant even if physically characterised events are always sufficient for their effects. 3.

Robb’s solution: “property” is ambiguous

Robb argues that in order to appreciate that Relevance, Distinction and Closure do not necessarily form an inconsistent set we need only note that the term “property” is ambiguous (Robb 1997). It is ambiguous between: Property1 = that which imparts on an individual thing its particular nature Property2 = that which makes distinct things the same. The observation that our notion of a property is apt to play, not one but two, distinct roles forms an important part not only of Robb’s argument for the causal relevance of mental properties, but also of an argument designed to justify trope theory generally. So Keith Campbell distinguishes between what he labels the A- and the B-question. “We can take one single red object,” he tells us, “and ask of it: what is it in virtue of which it is red? … [W]e can ask of any two red things: what is it about these two things in virtue of which they are both red (share the same nature)?” (1990, 29). The 7

In fact, it can do its argumentative work only if combined with a background assumption along the following lines (Macdonald and Macdonald 2006, 544): “If a property, P, of a cause, c, is causally sufficient for an effect, e, then no other property, Q, distinct from and independent of P, is causally relevant for e”. Cf. Lowe 2000, 26f. 8 In Bennett’s colourful terms (2007): “There are two hurdles to becoming gainfully employed. First, one must have the requisite skills; second, there must be employment opportunities available. It’s one thing to be fit for work, and quite another to actually find a job!”

36 observation is important because it points to a crucial difference between trope theory and alternative theories of properties. Whether you defend the existence of universals (in or ante rem) or instead require some suitably structured set of concrete objects to play the part of property, you will have this much in common; you will expect your one posit to play both roles and to answer both questions. Trope theory is different from every other theory of properties because it is the view that, not only are there these two roles, but there are two distinct role-fillers as well. Tropes are what impart on an object its particular nature. But the observation that this object is red because it is partly constituted by a red-trope will not automatically explain how come it is the same, colour-wise, as that red object. If you believe that properties are tropes, the “what makes them the same?” question will, as a consequence of the particularity of tropes, reproduce itself at the level of properties. Once again there are two individuals (the tropes) about which we can ask: in virtue of what are they the same? Distinct objects are the same, trope theorists now tell us, if at least some of the tropes that characterise them belong to the same similarity class.9 Tropes are what impart, on particular objects, their particular nature; similarity classes of tropes are what ground the sameness of distinct objects. Two roles; two role-fillers. Once “property” is disambiguated, we see that Relevance, Distinction and Closure are no longer inconsistent. In Relevance and Closure “property” should be understood as that which imparts on a particular object its nature. According to trope theory this is the role played by particular tropes. This is not a role played by the properties referred to in Distinction, however. Mental and physical properties must be distinct, because objects that are mentally the same might be physically different. This means that Distinction is plausible only if “property” is taken to pick out a type, not a token. To treat a property as type is to treat it as that which “makes” distinct things the same; and as we have seen this role is played, not by individual tropes, but similarity-classes of them. On Robb’s assessment the three claims should therefore be understood as follows: 9

As the topic here is not tropes per se, I will simply assume in what follows that it is acceptable and unproblematic to answer the “what makes them the same?” question by referring to similarity classes of tropes. Objections have been made to this suggestion, some referring to the threat of an infinite, vicious resemblance-regress. I have argued elsewhere (2002, 2007) that these can be met and will not repeat myself here.

37 RelevanceR: Mental tropes are (sometimes) causally relevant. DistinctionR: Classes of mentally similar tropes ≠ Classes of physically similar tropes. ClosureR: Every physical event has a sufficient cause characterised by a physical trope. These claims are not inconsistent. For, as Robb notes, once properties as tropes are distinguished from properties as similarity-classes of tropes we are free to identify mental properties with physical ones. When Lin feels pain there is an event in her brain, and this event is the way it is because of the tropes that characterise it. It is in virtue of at least one of these tropes that Lin’s pain causes her to remove her hand from the hot stove. The trope that characterises Lin’s pain-event is, as a result, causally relevant to the behavioural outcome. This is a mental trope, which is to say that it belongs to a similarity class of tropes that are similar in a sense relevant to our calling them “mental”.10 But Lin’s pain-event is also physical: that is, the trope that characterises her pain-event, and which is thereby relevantly involved in the production of the behavioural outcome, belongs to not just one, but two similarity classes. One and the same trope belongs to a class of functionally similar tropes and a class of neurophysiologically similar tropes, and these classes remain distinct.11 The situation is represented in the figure below.

10

I assume in what follows that this means that they are functionally similar—nothing substantial depends on this assumption, however. 11 Robb (1997) does not discuss the nature of tropes capable of belonging to such radically different similarity classes. In a later paper he has more to say on the matter (Heil and Robb 2003, 47): “What we propose is a surprising identity: the dispositional and the qualitative are identical with one another and with the unitary intrinsic property itself.” Tropes, it seems, can simultaneously belong to mental as well as physical resemblance classes because a trope is both a quality and a disposition. See also Robb and Heil 2003 and Heil 2003.

38 = mental trope = physical trope = mental = physical trope

C

E

Noordhof has complained that this approach fails to respect what he calls the “bulge in the carpet constraint” (1998, 223):12 No candidate solution to a philosophical problem should raise another problem which appears just as intractable and which requires the resolution of an issue similar to that which made the original problem so intractable.

Robb’s suggestion moves, but does not remove, the bulge in the carpet, because the same type of question as was ambiguously asked about properties can now be unambiguously asked about tropes: Is it in virtue of being mental (i.e. in virtue of belonging to a class of functionally similar tropes) or in virtue of being physical (i.e. in virtue of belonging to a class of neurophysiologically similar tropes) that the trope is causally relevant to Lin’s removing her hand from the stove? But Noordhof’s complaint is unwarranted. As Robb (1997, 190) points out: It would be odd if once the qua problem were solved for events, it arose again regarding the very thing used in the solution: the objector seems to be making a kind of category mistake. Tropes 12

In his response to Noordhof, Robb agrees that, to be successful, a suggestion must live up to precisely this constraint (although, as he also and somewhat smugly points out, in his version of the constraint, “appears” would be replaced by “is”). He then goes on to argue that his suggestion does live up to the constraint (Robb 2001).

39 are not causally relevant qua this or that, they are causally relevant (or not) period.

It is important here to see exactly why the objection does not go through. Tropes are not causally relevant in virtue of belonging to this or that similarity class, because they are not mental (or physical) in virtue of belonging to some such class. A trope is what it is, primitively. It is in virtue of being what it is that it belongs to this or that similarity class. It is in virtue of being what it is that it is causally relevant (if it is). No bulge left.13 Mental properties (tropes) are (sometimes) causally relevant to the production of physical effects because mental tropes are identical with physical tropes. Relevance is therefore safe. 4.

Martin and Heil’s solution: properties are no slaves to the predicates

Martin and Heil seek not so much to solve as to dissolve the problem of the causal relevance of mental properties. They suggest that our temptation to think that Distinction is true derives from false assumptions. Distinction, remember, was introduced in recognition of the multiple realisability of mental properties—something itself justified by the fact that the same (mental) predicate can be satisfied by physically different distinct individuals. But according to Martin and Heil this route to Distinction is misguided, because “it would be a mistake, to imagine that every predicate, even every predicate that figures in a going empirical theory, designates a property” (1999, 44). The alternative is not to eliminate mental properties; mental predicates may still be satisfied by events in virtue of the properties of these events. The alternative is, instead, to insist that from the fact that the same (mental) predicate is satisfied by distinct individuals, nothing 13

Even if it is unsuccessful, the objection is instructive in that it draws attention to the considerable ontological cost of Robb’s suggestion. The primitive and simple nature of the trope must be capable of grounding many and varied similarities and (as a result of this) many and varied truths. Trope theories (and this is most of them) on which tropes are simultaneously abstract, particular and simple have been criticised for introducing mysterious, highly problematic trinities. In view of this, one can only wonder at how simple tropes able to sustain numerous resemblances of the kind here envisaged will be received. For criticism see Hochberg 2004. I defend the mysterious trinities in my 2005.

40 follows about exactly which properties are in play. It does follow that the objects in question are in some way saliently similar. Although, as the properties in virtue of which the predicate is satisfied need not be the same in every instance, it does not follow that the individuals in question are exactly similar. At most, therefore, we should expect there to be a “family resemblance” among individuals satisfying the same (mental) predicate. Pain (once again) serves as an illustration (1999, 51): [T]he idea is that mental predicates standardly thought to designate multiply realizable properties—“is a pain,” for instance—apply to objects that are similar in certain respects: similar enough to merit application of the predicate. These similarities stem from the objects’ possession of certain properties. But the properties need not be the same in every case: objects are not exactly similar in those respects in virtue of which it is true of them that they satisfy the predicate “is in pain”.

If multiple realisability no longer gives us reason to distinguish mental from physical properties, the problem of the causal relevance of mental properties dissolves. Whatever it is in virtue of which Lin satisfies the predicate “is in pain” can now be precisely the same as that in virtue of which she satisfies the predicate “is in neurophysiological state F1”. The fact that Robot-Lin could satisfy the very same mental predicate in virtue of some physical property (or properties) other than those in virtue of which that predicate was satisfied by Lin will not change this. Our three claims, properly understood, can be reconciled: RelevanceMH: Mental and physical predicates pick out causally relevant properties. DistinctionMH: Mental predicates ≠ physical predicates. ClosureMH: Every event has a sufficient physical cause. What motivates the claim that mental properties are no slaves to the predicates is not primarily the unfortunate role played by the assumption that they are such slaves in the generation of the problem of causal relevance. “[I]t is a mistake”, Martin and Heil point out, “to pursue ontology piecemeal” (1999, 43). And again, “ontology does not lend itself

41 to one-off answers to particular difficulties” (ibid., 44). The driver here, then, is not piecemeal interest in the dissolution of a particular problem, but considerations of a much more general and programmatic variety. Philosophical theorising must take place against a backdrop of fundamental, as it were, “framework” assumptions. If the theorising is ontological, these will, among other things, dictate what kinds of existential and categorising conclusion can be justifiably drawn from what kinds of evidence. Ideally, framework assumptions should always, and explicitly, be taken into account when a particular solution to a problem is evaluated. For, unless we are aware of the meta-assumptions on which a particular conclusion depends, we are in no position to criticise either that conclusion or the framework that is supposed to supply it with its justification. The same is true of the philosophical problem itself. Whether it is or is not a problem, and, if it is a problem, just how serious a problem it is, will be decided (at least partly) by the framework. Unless sufficient attention is spent on the framework, therefore, too much time may be devoted to the formulation of justified solutions to what are really pseudoproblems. It is just these sorts of consideration that motivate Martin and Heil. Their point is, in effect, that although the theoretical framework plays such an important role in theorising, it appears to operate, surprisingly often, behind the scenes—hidden from view, and hence from critical scrutiny. This introduces a highly problematic possibility: namely, that the framework assumptions which in fact inform most of our theorising are untenable and should be rejected. According to Martin and Heil this is not just a possibility; it is a fact. The meta-assumptions on which most ontological theorising (since Kant) depends do not stand up to critical scrutiny, and ontological theorising has, as a result, been consistently steered wrongly. Specifically, philosophers have been misled for centuries by the view that language (or a suitably regimented language) pictures reality in such a way that, simply by paying close attention to our conceptualisation of reality, we can obtain substantial information about reality itself. Sometimes it is assumed merely that the study of language will provide knowledge of the form or structure of reality. At its most extreme, however, it is tacitly assumed that not only formal, but also substantial ontological conclusions, conclusions about the content of reality, can be justifiably drawn from purely linguistic premises. Heil (2003) calls this view the “Picture Theory”.

42 The Picture Theory, it is maintained, is responsible for much that has gone wrong in philosophy. It has, among other things, led philosophers on a long and hopeless search for intricate, yet in the end always unsuccessful, paraphrases for expressions which, though intuitively true, do not immediately align well with reality. Worse, at least according to Heil, the Picture Theory is responsible for ontological theories importing levels of reality (2003, 7): We discover that most of the predicates we routinely use to describe the world fail to line up with distinct basic-level physical properties or collections of these. We conclude that the predicates in question must designate higher-level properties. Now we have arrived at a hierarchical conception of the world, one founded on the inspiration that there are levels of reality. Higher levels depend on, but are not reducible to, lower levels. My contention is that the idea that there are levels of reality is an artefact spawned by blind allegiance to the Picture Theory.

It should come as no surprise that the account of the causal relevance of mental properties offered by Martin and Heil depends essentially on their explicit rejection of the Picture Theory. It is only if you assume that the Picture Theory is true, they maintain, that you are led, when confronted with a multiply realisable predicate, to conclude that its corresponding property belongs to a different—higher—level of being. It is because the Picture Theory fools you into believing that properties are slaves to predicates that you are forced, wrongly, to conclude that Distinction (as originally formulated) is true. It is not true, which means that Relevance, Distinction and Closure do not form a problematic triad. Thus as soon as the Picture Theory is rejected, the problem of the causal relevance of mental properties stops being a problem and begins its new career as a pseudo-problem—a “problem” unworthy of philosophical attention. Problem dissolved. Relevance (again) safe. 5.

Does ontology matter?

What may come as more of a surprise is that, on my interpretation, Robb’s (1997) suggestion not only does not oblige us to reject the Picture Theory; it requires us to keep it. Not everyone would agree, but this is mainly because not everyone would agree that Robb’s position differs from that of

43 Martin and Heil. The notion that the two positions are in essence the same is one that even the authors themselves acquiesce in at times. In their (1999, 59), Martin and Heil refer to Robb’s view as one that “falls within the spirit of our discussion.” In later writings, Heil and Robb (who have now joined forces) insist that the views are not only similar in spirit; they are identical.14 In this section, I argue that the proposals are substantially different; that they differ in the strength of the framework assumptions on which they rest; and that this is a difference which results, among other things, in different implications vis-à-vis the argumentative use that can be made of tropes. Why do I think that Robb’s suggestion essentially depends on the Picture Theory? Robb himself certainly does not think so. In his joint papers with Heil (Robb and Heil 2003, Heil and Robb 2003) he argues that his original suggestion requires what Heil (in his 2003) called a “corollary” of the Picture Theory to be rejected. To put the point a little more exactly, the principle on the rejection of which the success of Robb’s suggestion supposedly depends is this (Heil and Robb 2003, 176): (P) For each predicate “F”, there exists one, and only one, property F, such that, if “F” is applicable to an object a, then “F” is applicable in virtue of a’s being F. Now, there is certainly a sense in which it is more or less trivially true that Robb’s suggestion depends on the rejection of (P). Robb, remember, solves the problem of the causal relevance of mental properties by pointing out that “property” is ambiguous: it can mean either property1 (that which characterises an individual thing) or property2 (that which makes distinct things the same). In (P) this distinction is ignored; but there is reason to believe that advocates of (P) would in fact assume that whatever plays the role of characterising individual objects is also what plays the role of 14

In their (2003, section 6.4) Robb and Heil maintain that the following view—which is obviously identical with the one I have just attributed to Robb—is the same as the one that Heil and Martin defend in their 1999: “The essential idea is that ‘property’ as we’ve used the term so far is ambiguous. Sometimes it is used to refer to those entities that characterize objects (events, substances, etc.); other times it is used to refer to those entities that unify objects, entities that are each a ‘one across many’.” Cf. Heil and Robb 2003.

44 unifying distinct objects as “the same”. That this is how Heil and Robb read (P) is clear from the fact that they also take (P) to entail (U), where (U) is the claim that (see their 2003, 177): (U) Properties are universals. Understood as a principle whose truth entails the existence of universals, (P) is obviously rejected by Robb. This rejection, however, does not amount to a rejection of the Picture Theory. To see this, consider first what Heil and Robb (2003, 186) choose as a replacement for (P): (P*) If a predicate “F” applies to an object a, then there is a property f, such that “F” is applicable to a in virtue of a’s being f. Now, given the suggestion made by Robb in his 1997, (P*) cannot be the right replacement for (P).15 Or perhaps we should say that it could be—but that then, contrary to appearances, this proposal, though it manages to dissolve the problem of causal relevance, gives us no reason to believe in the existence of tropes. Let me explain. According to Robb, “property” is ambiguous. To say of “property”—the word—that it is ambiguous, is to say that what appears to be one concept is in reality two. On Robb’s suggestion, therefore, we are asked to distinguish, in language, between two linguistic functions that a particular sort of word—the predicate—can have. To put the point another way, Robb is asking us to distinguish between property1 and property2 because of the existence, in language, of a corresponding distinction. The reason why (P), on Robb’s suggestion, should be replaced, is therefore not that we want to reject the Picture 15

It could be made to work with the position advocated by Martin and Heil, but notice that not even then would embracing (P*) constitute complete rejection of the Picture Theory. According to (P*), a predicate is satisfied by an object if there is some property had by that object. This means that, although on (P*), exactly what properties there are cannot be “read off” from language, that there are properties can be. If you think that the Picture Theory is untenable, you can only replace one version of it with another (weaker) one if you can explain why whatever ailed the stronger version of the theory does not also ail its weaker cousin. A justification along these lines is, I believe, largely lacking in Heil 2003. I try to supply something of the sort in the last section of this paper.

45 Theory; it is that we want our Picture Theory to reflect a correct analysis of linguistic (and, hence, ontological) form and function. Thus understood, Robb’s suggestion makes essential use of tropes. It is because trope theory can supply distinct sorts of entities, in reality, to correspond to every distinct function revealed in linguistic analysis that the problem of the causal relevance of mental properties is solved. In a world with no tropes, but only universals, the suggestion fails. Universals, remember, both characterise individual things and make distinct things the same. One and the same universal will therefore be picked out by predicates both in their characterising and their unifying function. But if that is so, the multiple realisability of mental predicates will force us to distinguish mental from physical properties, which means that the door is closed to the type of solution proposed by Robb. On his approach, therefore, ontology does matter. It matters, however, only if one very important piece of theoretical apparatus is accepted. This is the Picture Theory. It is only if we accept the Picture Theory that there is a problem of the causal relevance of mental properties to begin with, and only then can there be a solution to this problem along the lines suggested by Robb. Martin and Heil do not accept the Picture Theory and, hence, will not accept a problem “spawned” by it. But if there is no problem, there is no solution to the problem either. And if there is no solution to the problem, there is surely no part (essential or not) for tropes to play in it. As properties are no longer slaves to the predicates, we have no reason to suppose (say) that there is a specific property such that “being in pain” is a predicate true of an object in virtue of the object’s possessing that very property; and this means that even if the properties in virtue of which it is true that Lin is in pain are universals, the problem of causal relevance cannot be formulated. Predicates are multiply realisable; whether properties are as well, is not something we can tell by looking at language. Once the Picture Theory has been rejected, there is no argumentative work left for an ontology of tropes. On the suggestion made my Martin and Heil, therefore, ontology (narrowly construed) does not matter. 6.

Ontology with or without the Picture Theory

It is my hope that you are now convinced that, at least on one, not implausible, interpretation, Martin and Heil’s position differs substantially

46 from Robb’s. I hope also to have made a case for saying that this difference depends on the fundamental framework assumptions one accepts; and that it results, among other things, in a difference in the argumentative work you may expect to get out of your particular ontological posits. In this section I want to say just a few more words about whether, and how, we should endorse the Picture Theory. In particular, I want to argue that although there is certainly a sense in which the Picture Theory should be rejected, there is also a sense in which it must be retained. In view of this, ontological theorising could do with a new and improved Picture Theory. I conclude the paper with an initial attempt to accommodate this need. The Picture Theory comes in many different guises. Some are more plausible than others, but all of those proposed to date are implausible. On the strongest and (I believe) least plausible version of the theory, every semantically distinct general term (every predicate) in a true subjectpredicate proposition picks out a distinct property.16 It is on this kind of Picture Theory that Robb trades—at least, if tropes are to play the essential part in his argument they appear to. Very few of the very few philosophers explicitly concerned with spelling out their framework assumptions would accept this version of the Picture Theory. Many of them would, however, accept a version of the theory that differs from this one only in degree. This is the view that, although not every semantically distinct predicate picks out a property, every semantically distinct predicate featuring in the language of (ideal) science does.17 Irrespective of whether you assume that every, or only a select few, of the predicates can inform you about what properties there are, you must assume that the fact that there are predicates will inform you about the possible existence of properties. That is, to be able to argue from linguistic 16

Throughout this discussion it is assumed that although, on the Picture Theory, language “pictures” reality, the two are still distinct. 17 Cf. Armstrong 1978 and Mellor 1995. Mellor puts it this way (p. 192): “Let us therefore call the Ramsey sentence of (the conjunction of) all laws “Σ”. Σ by definition asserts the existence of all properties and relations which occur in laws of nature. And by so doing, I say Σ answers the question: what causal properties are there? The causal properties (and relations, if any) which exist in our world are the universals over which the Ramsey sentence of all our world’s laws must quantify.”

47 premises to conclusions about the content of reality, you must assume already that you can argue from linguistic to ontological form. The opposite is not true, however. You could very well maintain that conclusions about the form, but never about the content, of reality could be justifiably drawn from evidence obtained through a study of our conceptualisation of it. According to the most sophisticated version of this latter view, moreover, only a logical analysis of language can uncover the ontologically relevant distinctions.18 This is how Wittgenstein puts it in the Tractatus: Propositions show the logical form of reality.19 The propositions of logic describe the scaffolding of the world, or rather they represent it.20

On every version of the Picture Theory, that which, in language, informs you about reality, will so inform you in a particularly strong and straightforward way; the form and/or content of reality, it is supposed, corresponds exactly with the form and/or content of language. Language pictures reality. It is precisely this assumption which makes the Picture Theory, in whatever guise, unattractive.21 Language, first of all, obviously serves many purposes, of which the uncovering of the ultimate constituents of reality, if it is a purpose, is but one. The way we think and talk about reality depends, in other words, to a large extent not on how reality fundamentally is, but on such things as how we are: one’s interests, moods, religion, physical constitution, and so on. A theory that disregards this state of affairs is implausible and should be rejected. A Picture Theory according to which a logical analysis of language will reveal the form (but never the content) of reality escapes this sort of criticism but faces another. Why, one may ask, should we suppose that the form of reality just happens to correspond with the logical form of language as captured in predicate 18

Cf. the logical atomism associated with Russell (1956) and Wittgenstein (1922). Wittgenstein 1922, § 4.121 20 Wittgenstein 1922, § 6.124. 21 Armstrong refers to the thesis that language pictures reality as “the bane of correspondence theory” (2004, 16). Smith has called it “a dark force” which “haunts what is most admirable in philosophy of the last hundred years” (2005, 153). 19

48 logic?22 There appears to be no way in which this assumption could be justified. The Picture Theory should therefore be rejected, but must, I will next argue, in some guise be retained. The Picture Theory fulfils two distinct, albeit related, functions. On the one hand, it describes the relationship that holds between language and reality. On the other, it, thereby, regulates ontological theorising. It informs us, in other words, about what in reality as it appears to us can, and what cannot, justify conclusions about reality as it is. If the Picture Theory is accepted, ontological theorising can be justified. If it is rejected, and if we want to continue doing ontology, it must be replaced. Why, if not for reasons provided by the Picture Theory, should I hold that the world is a world of tropes? It is not easy to answer this question. To borrow some useful terminology first introduced by Strawson in his 1959, as soon the Picture Theory is rejected, the ontological enterprise ceases to be descriptive, and becomes revisionist. If reality does not mirror the structure and/or content of language, it follows that the structure and content of reality may be different, indeed very different, from that of our conceptualisation of it. But then, on what grounds can justified conclusions about reality be formulated? One would hope that some such grounds exist. We want our ontology to be more than a consistent set of propositions. We want it to be relevantly concerned with reality. But what could this mean, other than that we want ontology to be relevantly concerned with reality as it appears to us because we believe that reality as it appears to us is relevantly concerned with reality as it is? Something must replace the Picture Theory, and the only thing that can, it seems, is more of the same. The Picture Theory must be replaced by another more plausible account of what, in our conceptualisation of reality, can justifiably inform us about reality as it is, independently of how it seems. The Picture Theory must be replaced by a new Picture Theory. As we have seen, the old Picture Theory assumed an implausibly close relationship between language and reality. A new Picture Theory should therefore be one according to which, although our representation of reality 22

Specifically, first-order predicate logic has been the logic of choice for proponents of this view. But the objection does not go away if we treat some other logic as that in which the form of reality is reflected. Whatever our choice of logic, we can ask: why should we suppose that the form of reality just happens to correspond to that?

49 can justify ontological conclusions, the form and content of the latter cannot be directly “read off” from the form and content of the former. The new Picture Theory must, in other words, regulate ontological theorising, not too much, but just enough. In his 2003 (chs. 6-7) Heil seems to suggest that we should replace the Picture Theory with Truthmaker Theory. Roughly speaking, this is the view that propositions are made true by portions of reality. More precisely, it is the view that truth conforms to principle (T):23 (T)

is true if(f) there exists at least one truthmaker T for p.

Truthmaker Theory justifies the search for clues in language as to the fundamental constitution of reality by telling us that there is a relevant (and strong) connection between the two: true propositions are relevantly concerned with reality because they are made true by it. However, Truthmaker Theory does not tell us exactly what we should look for in language when formulating an ontological theory. To be able to regulate, and hence justify, ontological theorising, Truthmaker Theory, as it is characterised in (T), is therefore not enough. It is not enough, because, as it stands, it cannot help us to discriminate between different ontological theses. All it tells us is that truth depends on reality, not how it does so. We need to know what, as regards the true propositions, is capable of informing us about the fundamental constitution of reality. We do not want to say that their form (logical or otherwise) mirrors or pictures the form of reality. We certainly do not want to say that for every semantically distinct linguistic entity featuring in a true proposition there corresponds an object, a property, a state of affairs, or whatever. We want to be able to say something, however, or else it seems that the enterprise of ontological theorising will be grind to a halt. Let me end this paper by sketching one way in which (T) might perhaps be supplemented.24 23

“

” = whatever carries a truth-value. Here I will talk of propositions as the bearers of truth, but I shall regard the expression as a place-holder for whatever detailed investigation reveals as the most appropriate truth-bearer. A similar approach is taken by Simons 1992, Rodriguez-Pereyra 2005 and Armstrong 2004. Much more could be (and has been) said about (T). For some further discussion, see Beebee and Dodd 2005. 24 Cf. Keinänen 2005, Maurin forthcoming.

50 7.

A sketch of a new Picture Theory

I suggest that what we should look for in true propositions is this: the roles that whatever in reality makes them true must be able to play. The framework assumption that I believe should complement Truthmaker Theory is therefore this: the “roles” that appear to be roles that whatever exists must be able to play for propositions to be true, are roles that whatever exists must be able to play. Neither form nor content is mirrored; truthmaking roles are. Consider a simple example. , let us suppose, is a true proposition. It appears to require for its truth something which plays the roles we associate with entities like Lin. What roles are those? Further study may reveal that the roles that whatever makes true propositions like should be able to play include the types of role normally associated with what is sometimes called a concrete individual. We should not immediately conclude that the world must, therefore, include entities that belong to this category; at least, we need not conclude that the world is fundamentally made up from entities of this kind. We should continue to investigate. What roles are associated with concrete individuals? Suppose that the answer to this question will provide a list of roles including, for example, the power to exist through time, the power to change, the power to monopolise one’s position in space-time, and so on. Suppose that the list is finalised. Then we may conclude that whatever there is must be such that it (or complexes built up somehow from it) can play these roles, because whatever there is must, minimally, be able to fulfil its truthmaking function. Likewise, instead of saying, as Robb does, that “property” is ambiguous, we should say that whatever makes true a proposition such as must minimally be able to play the role of characterising an individual object but need not be able to play the role of unifying distinct things. By contrast, it would appear that whatever makes true a proposition such as , must be able to play both of these roles. Surely this is precisely the sort of analysis that philosophers have been doing for centuries? Yes, but not exactly. The type of investigation required by proper identification of truthmaking roles, I imagine, involves more than classical conceptual analysis. A better description of what it covers is in fact provided by Strawson when he describes descriptive metaphysics as follows (Strawson 1959, 9-10):

51 How should [descriptive metaphysics] differ from what is called philosophical, or logical, or conceptual analysis? It does not differ in kind of intention, but only in scope and generality. Aiming to lay bare the most general features of our conceptual structure, it can take far less for granted than a more limited and partial conceptual inquiry. Hence, also, a certain difference in method. Up to a point, the reliance upon a close examination of the actual use of words is the best, and indeed the only sure, way in philosophy. But the discriminations we make, and the connexions we can establish in this way, are not general enough and not farreaching enough to meet the full metaphysical demand for understanding.

A Picture Theory developed along these lines will, I believe, regulate ontological theorising just enough. But in adopting it we must give up many well-entrenched patterns of reasoning in ontology. On the new Picture Theory, appearances will at most justify conclusions about truthmaking roles—not (or at least, not directly) conclusions about the sort of entity that plays them. Once the truthmaking roles have been identified, therefore, only purely philosophical arguments can decide what the fundamental constitution of reality actually is (or can be). And philosophical arguments can only take us that far. If rival ontological theories posit entities equally apt to fulfil their truthmaking function, not much can be said in favour of one theory over the other. We have already made use of the evidence provided by representation. We are left to make our discriminations purely on such grounds as the ordinary theoretical virtues. Suppose it turns out, for instance, that reality must contain entities able both to characterise individual things and to make distinct things the same. Then, as we have seen, these are roles for which both a universal realist and a trope theorist can supply suitable actors. How, then, can we choose between them? Perhaps trope theory should be preferred, because it is the ontologically simpler theory (assuming, of course, that trope theory can be developed as a one-category theory, and that universal realism cannot). But simplicity is a double-edged sword. Whether or not trope theory is the simpler theory will depend, among other things, on whether ontological simplicity trumps theoretical simplicity. According to Armstrong, it does not: universal realism should be preferred precisely

52 because it is the theoretically simpler theory:25 and the debate continues. In view of unsettled issues of this kind I am inclined to take the following view: I certainly agree with Martin and Heil when they emphasise the importance of taking ontology seriously, but I am far from confident that its role in dis(solving) particular issues in the philosophy of mind (or elsewhere) will turn out to be as important as is sometimes believed. 8.

Conclusion

If the only thing pictured in language is the truthmaking roles that whatever there is must be able to play, the problem of mental properties’ causal relevance (as set up here) dissolves. It dissolves, moreover, thanks simply to our rejection of the (old) Picture Theory, not as a result of specific ontological choices. The reason for this and the reason why the problem similarly dissolves in Martin and Heil’s treatment are identical. Even if the study of our conceptualisation of reality can reveal to us what truthmaking roles the world’s inhabitants must be able to play, it simply does not follow that we can conclude, from the presence of a particular predicate such as “is in pain”, that there exists one and only one entity to which it corresponds. As this leaves no argumentative work for tropes to do, we cannot, as Robb does, hold that the successful solution of the problem of the causal relevance of mental properties is in itself a reason to accept the existence of tropes. The old Picture Theory should be rejected; Martin and Heil were right. What they perhaps failed to fully appreciate were the ramifications of such a rejection. Many of what we thought were genuine problems are now revealed as pseudo-problems. But pseudoproblems (like their pseudo-solutions) provide us with no reason to accept the existence or non-existence of anything. So, does ontology matter? Yes, in general. Ultimately, truths are made true by the most fundamental constituents of reality; ontology makes a difference to truth. What about our specific ontological choices? Do they matter? Not to the extent that some have thought they do. To establish that tropes can fulfil their 25

According to Armstrong, universal realism is more attractive than trope theory, because a trope theorist must treat the axioms of resemblance as primitive, whereas a universal realist can analyse them in terms of identity (Armstrong 1997, 23). Heil 2003 criticises this view.

53 truthmaking function is not to establish trope theory. We can still ask: Is the world a world of tropes? REFERENCES Armstrong, D. M. (1978). Universals and Scientific Realism. Cambridge: Cambridge University Press. Armstrong, D. M. (1997). A World of States of Affairs. Cambridge: Cambridge University Press. Armstrong, D. M. (2004). Truth and Truthmakers. Cambridge: Cambridge University Press. Bacon, J. (1995). Universals and Property Instances: The Alphabet of Being. Oxford: Blackwell. Beebee, H. and Dodd, J., eds. (2005). Truthmakers—The Contemporary Debate. Oxford: Oxford University Press. Bennett, K. (2003). Why the Exclusion Problem Seems Intractable and How, Just Maybe, To Tract It. Noûs, 37, 471-497. Bennett, K. (2007). Mental Causation. Philosophy Compass, 2, 316-337. Braun, D. (1995). Causally Perspectives, 9, 447-475.

Relevant

Properties.

Philosophical

Campbell, K. (1990). Abstract Particulars. Oxford: Blackwell. Davidson, D. (1980). Essays on Actions and Events. Oxford: Oxford University Press. Ehring, D. (1997). Causation and Persistence. Oxford: Oxford University Press. Gibb, S. C. (2004). The Problem of Mental Causation and the Nature of Properties. Australasian Journal of Philosophy, 82, 464-475. Heil, J. (2003). From an Ontological Point of View. Oxford: Oxford University Press. Heil, J. and Robb, D. (2003). Mental Properties. American Philosophical Quarterly, 40, 175-196. Hochberg, H. (2004). Relations, Properties and Predicates. Relations and Predicates, eds. H. Hochberg and K. Mulligan, Frankfurt: Ontos Verlag,

54 17-53. Horgan, T. (1989). Mental Quasation. Philosophical Perspectives, 3, 4776. Keinänen, M. (2005). Trope Theories and the Problem of Universals. Helsinki: Department of Philosophy. Kim, J. (1999). Making Sense of Emergence. Philosophical Studies, 95, 336. Lowe, E. J. (2000). An Introduction to the Philosophy of Mind. Cambridge: Cambridge University Press. Lowe, E. J. (2003). Physical Causal Closure and the Invisibility of Mental Causation. Physicalism and Mental Causation: The Metaphysics of Mind and Action, eds. S. Walter and H. Heckman, Thoverton: Imprint Academic, 137-154. Macdonald, C. and Macdonald G. (1986). Mental Causes and Explanation of Action. The Philosophical Quarterly, 36, 145-158. Macdonald, C. and Macdonald G. (1995). How to be Psychologically Relevant. Philosophy of Psychology: Debates on Psychological Explanation, eds. C. Macdonald and G. Macdonald. Oxford: Oxford University Press, 60-77. Macdonald, C. and Macdonald G. (2006). The Metaphysics of Mental Causation. The Journal of Philosophy, 103, 539-576. Martin, C. B. (1980). Substance Substantiated. Australasian Journal of Philosophy, 58, 3-20. Martin, C. B. and Heil, J. (1999). The Ontological Turn. Midwest Studies in Philosophy, 23, 34-60. Maurin, A.-S. (2002). If Tropes. Dordrecht: Kluwer Academic Publishers. Maurin, A.-S. (2005). Same but Different. Metaphysica, 1, 129-146. Maurin, A.-S. (2007). Infinite Regress—Virtue or Vice?. Hommage à Wlodek, eds. T. Rønnow-Rasmussen, B. Petersson, J. Josefsson and D. Egonsson, URL: . Maurin, A.-S. (forthcoming). A World of Tropes?. Worldviews, Science and Us: Studies of Analytical Metaphysics. A Selection of Topics from a Methodological Perspective, ed. R. Vanderbeeken., Brussels: World

55 Scientific Publishers. Mellor, D. M. (1995). The Facts of Causation. London: Routledge & Kegan Paul. Noordhof, P. (1998). Do Tropes Resolve the Problem of Mental Causation?. Philosophical Quarterly, 48, 221-226. Robb, D. (1997). The Properties of Mental Causation. The Philosophical Quarterly, 47, 178-194. Robb, D. (2001). Reply to Noordhof on Mental Causation. Philosophical Quarterly, 51, 90-94. Robb, D. and Heil, J. (2003). Mental Causation, Stanford Encyclopaedia of Philosophy (Winter 2003 edition), ed. E. N. Zalta, URL: . Rodriguez-Pereyra, G. (2005). Why Truthmakers. Truthmakers, ed. H. Beebee and J. Dodd. Oxford: Oxford University Press, 17-31. Russell, B. (1956). Logic and Knowledge, ed. R. C. Marsh. London: George, Allen & Unwin. Simons, P. (1992). Logical Atomism and its Ontological Refinements: A Defense. Language, Truth and Ontology, ed. K. Mulligan, Dordrecht: Kluwer Academic Publishers, 157-179. Simons, P. (1994). Particulars in Particular Clothing: Three Trope Theories of Substance. Philosophy and Phenomenological Research, 54, 553-575. Smith, B. (2005). Against Fantology, Experience and Analysis. ed. M. E. Reicher and J. C. Marek. Vienna: öbv&hpt, 153-170. Strawson, P. F. (1959). Individuals. London: Routledge & Kegan Paul. Williams, D. C. (1953). On the Elements of Being I. The Review of Metaphysics, 7, 3-18. Wittgenstein, L. (1922). Tractatus Logico-Philosophicus, trans. D. F. Pears and B. F. McGuiness. London: Routledge & Kegan Paul. Yablo, S. (1992). Mental Causation. Philosophical Review, 101, 245-280.

BASIC ONTOLOGY, MULTIPLE REALIZABILITY AND MENTAL CAUSATION FRANCESCO ORILIA Università di Macerata [email protected] ABSTRACT. In basic ontology philosophers dispute inter alia about the nature of properties and events. Two main rival views can be identified in the current debate. According to universalism, properties are universals and events are structured entities involving as constituents (in a typical case) a particular, a property qua universal and a time. In contrast, according to tropism, properties are tropes, abstract particulars that can also be viewed as events. This paper analyzes the resources that these two doctrines can offer in an attempt to construct a nonreductive physicalist account of the mental that accommodates multiple realizability without falling prey to epiphenomenalism (in short, NENRP). The tropist resources allow for success by means of a strategy leading to a non-unitarian doctrine. This rules out the unitarian idea, according to which creatures with important physical differences can still have the very same experiences. This tropist strategy can be “simulated” in a version of universalism, with the same non-unitarian consequences. However, from the perspective of another, possibly more natural, version of universalism, one can perhaps find another avenue for NENRP, which brings with it a unitarian point of view. The resulting approach strongly suggests, in contrast to one based on tropism, the possibility of something like self-acquaintance as understood by philosophers such as Russell or Chisholm. In sum, the paper shows ways in which basic ontology matters in philosophy of mind.

1.

Introduction1

In basic ontology, or metaphysics,2 we dispute, inter alia, about the nature of properties and we make theoretical choices about them. There have been 1

I am especially indebted to Ausonio Marras for detailed comments which allowed me to improve on a previous version of this paper, both in style and content. Nevia Dolcini, Anna-Sofia Maurin and David Robb provided helpful comments too. I am also grateful to Ausonio Marras as well as to Simone Gozzano and Cynthia MacDonald for useful discussions on the topics discussed here. In particular, professor MacDonald has kindly clarified for me various aspects of MacDonald and MacDonald 2006. A forerunner of the present work was delivered at a meeting of the Italian Society for Analytic Philosophy (Orilia 2006a). I wish to thank those in the audience who provided comments and suggestions.

58 recent works that suggest that our decisions at this foundational level may have far-reaching consequences in philosophy of mind, consequences which are essentially bound to, or at least strongly suggested by, the foundational choices in question. Robb 1997 is a clear-cut example, as it accommodates multiple realizability and mental causation, by relying on what I shall call tropism, in essence the view that the properties of ordinary objects are tropes.3 For Robb’s account is at least at first glance precluded to the supporter of the rival view, universalism, according to which such properties are universals. More specifically, one is tempted to draw from Robb 1997 the lesson that tropism, contrary to universalism, is compatible with an independently appealing doctrine in philosophy of mind, which we could call non-epiphenomenalist nonreductive physicalism (NENRP, in short). This is an approach that endorses multiple realizability while eschewing the embarrassing idea that the mental is causally inert. Heil’s contribution to this volume (2008) somehow questions that there is such a lesson by demonstrating that the aspects of tropism that make it compatible with NENRP could be, so to speak, “simulated” in an appropriate version of universalism, Heil universalism, as I shall name it. This form of universalism has in fact the same sort of compatibility with NENRP enjoyed by tropism. Heil recognizes that basic ontological issues might have a bearing on philosophy of mind and accordingly tells us that we should “pursue philosophy of mind in a metaphysically self-conscious way” (Heil 2008, 14). But his point is that sometime, contrary to prima facie appearances, such a bearing may lack and in particular this is the case with the issue of tropism vs. universalism, at least as it relates to the topics discussed in Robb 1997. And thus Heil (2008, 28) claims the following: It turns out that the question whether properties are modes [i.e., tropes, in the terminology of this paper] or are universals is one the answer to which has less bearing than you might think on questions that have bedevilled philosophers of mind.

I shall try to show however that the issue of tropism vs. universalism has more bearing on philosophy of mind than what Heil seems disposed to

2

For present purposes I shall not assume any significant difference between the meanings of “ontology” and “metaphysics”. 3 As Robb notes, a version of his tropist approach can also be found in Heil 1992, 136139.

59 recognize,4 or at least that it has an important kind of bearing that is missed if the universalist, wrongly, as I shall argue, remains content with Heil universalism. With this goal in mind I shall revisit in the next five sections some well-known topics in ontology (sections 2-3) and philosophy of mind (sections 4-6). This will make this paper, and indeed this volume, more self-contained and will allow me to fix some terminology. The reader familiar with the themes in question can glimpse quickly through the pages that they occupy or perhaps even skip them. Then, I shall proceed as follows. First, I shall establish to my satisfaction that a tropist nonreductive physicalism along the lines of the one proposed by Robb can indeed meet the demand for non-epiphenomenalism, in spite of criticisms by Nordhof 1998 and MacDonald and MacDonald 2006. This will grant the compatibility of tropism and NENRP. Second, I shall verify that Heil universalism can similarly accommodate this demand. Third, I shall argue that Heil’s version may not be the most appropriate or natural type of universalism and thus I shall consider another form of this doctrine, monist universalism. Fourth, I shall indicate how the latter can be made compatible with NENRP. Finally, I shall compare the three different versions of NENRP that emerge from the discussion, one based on tropism, another on Heil universalism and lastly a third one relying on monist universalism. It will turn out that there is an important divide, with the first two of them on the one side and the latter version on the other side—a divide which is arguably significant from the point of view of philosophy of mind. 2.

Properties

Universalism and tropism are perhaps the main contenders nowadays as far as basic ontology go. In this section I shall explain how these two rival 4

Heil’s inclinations in this respect and the conception of the relation between language and reality that back them up are discussed in Maurin’s (2008) contribution to this volume. It should be noted that Heil does not assert that the issue of tropism vs. universalism has no bearing at all in philosophy of mind. In fact, he emphasizes that the acknowledgement of tropes can have the positive role of allowing for properties in the ontological inventory of a philosopher (of mind) who does not want to be committed to universals and moreover he hints (see the concluding remarks in his 2008) that acknowledging tropes rather than universals gives one an advantage in the treatment of intentionality.

60 doctrines are to be understood by focusing on those aspects that are specifically relevant for our purposes. The main point is that, according to universalism, the properties of ordinary objects (such as rocks, apples, dogs and people) are universals, whereas, according to tropism, such properties are tropes, understood as abstract particulars. Properties qua universals can be shared (in typical cases), i.e., they can be possessed simultaneously by different ordinary objects and are then somehow, roughly speaking, spread over the portions of space (or space-time) where these different objects are localized. Contrariwise, properties qua tropes cannot be shared. Each of them can be possessed only by one specific ordinary object and it is thus wholly localized in the spatial (or space-time) region occupied by this object. In this sense, tropes are particulars. But they are also abstract in that many of them can be localized (via the ordinary object that possesses them) in the same portion of space. In contrast, an ordinary object is a concrete particular in that only one such object can be localized in a certain portion of space. We can illustrate the contrast, by describing the reactions that two philosophers, a universalist and a tropist, may have in confronting a crucial intertheoretical datum, the crucial datum, as we shall call it. The datum is that different ordinary objects appear to present objective similarities that induce us to appeal to a single predicate to characterize them. Thus, for example, we may notice that two distinct coins have the same shape and accordingly we characterize both of them as “round”. By using the usual terminology in a pre-theoretical neutral way, both philosophers might agree that in some objective sense both coins exemplify (instantiate, have, possess) roundness, or to put it otherwise, that they both fall under a certain type, roundness, to which the predicate corresponds. However, according to the universalist, this type is a universal and the falling under it of the coins is simply their instantiating the universal in question. According to the tropist, on the other hand, the type is a (trope) resemblance class, a collection of mutually resembling (similar) tropes.5 Moreover, the falling under it of the coins is due to the fact that each coin instantiates a distinct trope belonging to the class in question. The notion of resemblance or similarity of tropes is a primitive one 6 and 5

Some tropists (perhaps with more nominalist inclinations) might prefer to speak of concepts or even predicates instead of classes of mutually resembling tropes, but we can neglect this, as far as we are concerned here. 6 A tropist may want to make an exception to this, when it comes to “mereologically complex” properties. See Robb 2008, note 15.

61 correspondingly it is a “brute fact” whether or not two tropes are similar.7 On the other hand, the similarity of two ordinary objects is explained in terms of similarity of tropes. It is important to note that trope resemblance can have degrees and thus we can distinguish between a perfect and a partial (approximate, inexact 8 ) one. In the former case, the similarity reaches its maximum, whereas in the latter this maximum is only approximated, to a higher or to a lower extent, as the case may be. Thus for example two red objects with exactly the same shade of red will have two red tropes that resemble each other perfectly, whereas one with a scarlet shade of red and another with a crimson shade of red will have two partially resembling tropes, a scarlet trope and a crimson trope, respectively. All these tropes can be assumed to be members of the class of partially resembling tropes corresponding to the predicate “red” and in turn all of them belong to an even wider class (involving a lower degree of resemblance) corresponding to the predicate “coloured”. It is also important to note that trope similarity (and thus membership in a resemblance class) is a necessary affair (it is an internal relation, as someone may want to put it): if trope t1 resembles trope t2 to a certain degree, this is so necessarily. One way to put the contrast between these two doctrines is this. For the universalist, universals fulfil two roles. They characterize objects and also unify them. For the tropist these two roles are fulfilled by two different role fillers: tropes characterize and resemblance classes unify (cf. Robb 1997 and Maurin 2008). Given this, from the tropist’s point of view “property” is ambiguous in that it may stand for the “characterizers”, tropes, or for the “unifiers”, resemblance classes. Correspondingly, in this perspective, we tend to use predicates such as “red”, “crimson” and “scarlet” ambiguously, to speak about, depending on the context, either tropes, on the basis of their belonging to a certain resemblance class, or ordinary objects, on the basis of their having tropes belonging to a certain 7

I take it that for the typical supporter of tropes the existence of these brute facts is an objective matter and accordingly the corresponding resemblance classes are usually called natural (see, e.g., Armostrong 1989). For present purposes we can assume that this idea is part of tropism. It should be noted that the very existence of the natural classes in question depends on the individual tropes’ being what they are, thereby resembling some tropes and failing to resemble others, as a matter of brute fact. 8 A.-S. Maurin has urged to me in correspondence that “approximate” or “inexact” may be more appropriate than “partial”. She may be right, but many trope theorists use “partial” (see, e.g., Robb 2008).

62 class (witness the above example, with respect to which expressions such as “red objects” and “red tropes” were both used). By the same token, “exemplification” and related words may be regarded as ambiguous insofar as they can be used either to indicate that an object exemplifies a certain trope or to indicate that the object relates to a certain class by virtue of exemplifying a trope in that class. Maurin 2008 thus embellishes “property” with the indexes “1” and “2” in order to distinguish between tropes (index “1”) and resemblance classes (index “2”). I shall do the same and I shall also use in the same fashion these indexes with “exemplification” and related words, when the need to disambiguate will be more pressing. Thus, for example, from the tropist’s perspective, a certain coin is round in that it exemplifies2 the property2 roundness (a resemblance class) by way of exemplifying1 a property1 (trope) belonging to the property2 roundness. It should be noted that in a sense a trope can be regarded as an instance of the property2 to which it belongs, a thin instance, we may say, as contrasted with the corresponding thick instance (to adapt a terminology differently used by Armstrong, e.g. in his 1997), namely the ordinary object that has1 the trope in question. We may then say that the trope instantiates0 the property2 to which it belongs.9 Traditionally, we distinguish between determinables and determinates. For example, red is a determinate of the determinable colour and in turn a determinable of the determinates crimson and scarlet (in general, following Yablo 1992, when F is a determinate of G, we can say that F determines G, or equivalently that F and G are in the determination relation 10 ). When a determinate fails to be a determinable of something else, we may call it an ultimate determinate. For example, an absolutely specific shade of red (presumably we do not have a predicate for it) or the properties square and rectangular (as determinates of quadrilateral) may be ultimate determinates. In a universalist perspective, determinables are “higher-level” universals determined by “lower-level” ones and an ordinary object exemplifies a determinable by virtue of exemplifying an ultimate determinate of the determinable in question. Thus, for example, it 9

Not all trope theorists may like this terminology. A.-S. Maurin, for example, has objected in correspondence to my use of “instantiation0”. Note however that this is not so different from Williams’ use of “exemplify” (Williams 1953, 117; cf. Gozzano 2008, section 2). 10 For a recent and interesting analysis of determination, see Funkhouser 2006. Funkhouser claims that his approach can be best applied to a tropist ontology, but he does not rule out its use in a universalist worldview.

63 is not possible for an object to be quadrilateral, if not by being square or rectangular or ..., as the case may be.11 The tropist can account for the determination relation by appealing to classes of partial resemblance. For example, a partial resemblance class such as the property2 red comprises the two subclasses crimson and scarlet, whose respective members have a higher degree of similarity. In this sense, crimson and scarlet can be said to be two different determinates of red. From such a perspective, the ultimate determinates are classes of perfect resemblance and an ordinary object can exemplify2 a determinable only by virtue of belonging to a determinate comprised in the determinable in question (thereby exemplifying1 a trope in the perfect resemblance class which is this determinate). Determination is a relation of (asymmetric) necessitation between two properties (Yablo 1992). This must be understood in the light of these definitions: A property F necessitates a property G iff necessarily, whatever has F has also G; F asymmetrically necessitates a property G iff (i) F necessitates G and (ii) possibly, something has G but does not have F. For instance, whatever is crimson must be red, although it may be the case that what is red is not crimson; it could be, say, scarlet. The necessity in question can somehow be considered metaphysical, if not logical, and accordingly such is the necessitation involved in the determination relation. There will be reasons to consider a necessitation arising from a necessity that is presumably just nomic or physical.12 11

Armstrong has tried to argue that the only universals that really exist are ultimate determinates and that accordingly there are no determinables (1978, 117). It is not clear to me that he has maintained this position over the years. For example in his 1989, ch. 5, section IX, he seems to admit determinables in at least some cases. At any rate, the idea that there are determinables is more widespread and can be taken for granted as part of universalism, as far as we are concerned here. 12 For an introductory comparison of universalism and tropism, see Armstrong 1989. Contemporary defences of ontological views based on universals can be found, e.g., in Bergmann 1967 and Armstrong 1978, 1978a, 1997. Armstrong is committed to a “sparse” conception, according to which not every predicate corresponds to a universal. I agree with this, although I would also emphasize that there is in any case a concept expressed by the predicate, to the extent that it is meaningful. For contemporary defences of ontological views based on tropes, see, e.g., Williams 1953, Campbell 1990 and Maurin 2002. Further details on the contrast between tropes and universals and additional references can be found in the other contributions to this volume. See in particular Gozzano’s contribution, for an extensive account of the nature of tropes. In addition to the rough outline offered here in this section, there are various things that should be noted for clarity’s sake, but that can for simplicity’s sake

64 3.

Events

In current philosophical literature there are many accounts of the entities that fill the theoretical roles of truthmakers and causal relata. At least some of these accounts contain, more or less explicitly, the claim that such entities coincide: there is just one role-filler for these two roles in that, at least in typical cases, the entities that make true sentences true are also the entities classifiable as causes or effects, as the case may be. Indeed, this is most plausible. We can say for example that Tom’s running at a certain time is what makes the sentence “Tom is running” (uttered at the appropriate time) true and we can also say that it is Tom’s running that causes Tom’s losing weight. I shall thus take this claim for granted. In discussions in which causality is most central the word “event” is typically used to designate these entities. On the other hand, when issues such as truth and predication gain centre stage, “state of affairs” and “fact” tend to be employed. Quite independently of the subject matter, however, it might be appropriate to reserve the word “event” (possibly, together with be left out as much as possible. Here they are. In general, an entity, insofar as it exemplifies a property, can be called a property-instance. The property-instances which are particulars, but are not tropes, events or the like (see below) may be called primary property-instances. I have focused on what may be considered the paradigmatic cases of such items, namely ordinary objects. But there are of course particulars not usually classified as ordinary objects, e.g., microphysical entities (photons, electrons, protons, etc.), which count as primary property-instances for both universalists and tropists. Although I shall avoid talk of relations as far as possible, it must be mentioned that property-instances can be linked to each other by relations. The typical universalist will acknowledge relations among universals. The tropist, typically, will take the ordinary relations between primary property-instances (such as spatial ones) to be tropes, but still a tropist might acknowledge that some relations are universals (I have in mind in particular the relations of (more or less perfect) resemblance among tropes). This does not mean that such a tropist is also a universalist. The essential mark of tropism, as I understand it here, is that it takes the properties of primary property-instances to be tropes (independently of whether or not some universals are also acknowledged). There may be however “mixed” tropistuniversalist ontologies. In order to account for some intriguing empirical data from quantum mechanics, I have elsewhere (2006) tentatively put forward the view that tropism and universalism might co-exist in the sense that the properties of ordinary objects are universals and those of some microphysical items (bosons) are tropes. And there are philosophers, such as Lowe (see his contribution to this volume), who acknowledge in their ontology, for systematic reasons, both tropes and universals as properties of primary property-instances. For present purposes we can ignore these approaches.

65 “process”) for those cases in which change is involved (as in Tom’s running) and “fact” and “states of affairs” or simply “states” for those in which it is not (as in Tom’s being human). Accordingly, it might also be appropriate to introduce a general term such as “occurrence” to designate in one fell swoop all the entities in question (Mourelatos 1978, Gill 1993). But as long as subtleties such as this are not crucial one may well for simplicity’s sake regard “event” and “states of affairs” as interchangeable. This is what I shall do here, although, following the tradition, I shall mostly use “event”, since issues concerning causation will be more prominent in what follows. Some regard events as particulars and others as universals (Lombard 1998). For present purposes we can assume the first option. There are two competing views about events (conceived of as particulars), one is structuralist and the other is non-structuralist, as we may say. According to the former, events are structured entities involving as constituents a property, a concrete particular (in typical cases) and a time. They exist inasmuch as the particular exemplifies the property at the time in question; this is their Existence condition, as Kim puts it (1993, 35).13 This view is typically associated to Armstrong (1997) in ontology and Kim in philosophy of mind (Kim 1993; in relation to Kim, it is often called the property exemplification view or similarly). According to the latter, usually associated to Davidson (1980), events are conceived of as basic (unstructured) particulars which fall under types. Events as conceived by Davidson are (arguably) truthmakers 14 and (most prominently) causal relata. Indeed, Davidson emphasizes the importance of causality in his account of events to the point of proposing that causation is the key to provide the identity condition for events, as follows: ICD event e is identical to event b iff a and b have the same causes and the same effects. 13

In keeping with the promise of avoiding talk of relations as far as possible, I neglect relational events which may be taken to involve various particulars and one relation instead of a single particular and a property. If relational events are taken into account a more complicated version of what I call below Kim Identity Condition is needed (Kim 1973). 14 A crucial point in Davidson’s account is the possibility of claiming that the same event makes a number of different sentences true, e.g., “Jones is very amused”, and “Jones is amused about what happened in the kitchen” (Feldman 1980, 151).

66 The principle ICD has been criticized because it appears to be circular (Feldman 1980, Quine 1985). But it is not essential to the view that events are basic particulars (and in fact Davidson came to abandon it; see his 1985). For present purposes we can then speak of the events of the nonstructuralist conception as Davidsonian events, without implying, by using “Davidsonian”, that they obey the implausible identity condition ICD. The central point of this conception, as contrasted with the rival structuralist standpoint, is that the former allows, whereas the latter does not, for talk of events in terms of constituents. In the structuralist conception it is appropriate to represent an event, as Kim does, by means of a correspondingly structured symbol such as “[x, P, t]”. This is taken to stand for the event which results from the particular x’s exemplifying the property P at time t. In Kim’s terminology (1993, 35), we can call P the constitutive property of the event in question. It should be remarked that P is not a property exemplified by the event [x, P, t], but rather by the particular x, although there are properties that the event instantiates by virtue of having P as constitutive property. For example, the event of Tom’s running at time t, [Tom, running, t], is a run (a running event, an event involving someone who is running), since it involves running as constitutive property. And since to run is to move, the event in question is also a movement. The event can also have properties that do not depend on its having the constitutive property that it has. For example, if t is in the Summer, the event is a Summer run and if Tom is wearing shoes it is a run with shoes. The non-structuralist view of events of course also recognizes that events (like any entity in the ontological inventory) have properties.15 Thus, in this conception as well, the event of Tom’s running at t is a movement, a run, and possibly a Summer run with shoes. In the light of the above, there are two ways of understanding the claim that an event is physical or mental, or more generally characterizable as being F, depending on whether we accept a structuralist or a nonstructuralist view. According to the first approach, an event is characterizable as F if it has a constitutive property which is of kind F (a determinate of F). Thus, for example, the event [Tom, running, t], is physical to the extent that running is a physical property. In contrast, the event [Tom, feeling pain, t] is mental to the extent that feeling pain is a 15

Davidson is a nominalist who prefers to say that events can be described in different ways rather than admitting that they really have properties, but for present purposes we can neglect this.

67 mental property.16 According to the second approach, the claim can only mean that the event as such falls under a certain property. For example, a run is physical and a pain is mental. In an approach in which (unstructured) events are identified with tropes (see below), we can say that a run is physical in that it belongs to a class of (partially) resembling tropes corresponding to the predicate “physical”, and a pain is mental in that it belongs to a class of (partially) resembling tropes corresponding to the predicate “mental”. Kim and Armstrong take the properties involved as constituents in events to be universals. Accordingly, they propose what could be called a universalist structuralist account of events. The universalist is quite naturally inclined to it and thus we shall assume, at least for present purposes, that the doctrine of universalism comprises this assumption. Kim has also proposed an identity condition for events which goes as follows: Kim Identity Condition [x, P, t1] = [y, Q, t2] iff (x = y & P = Q & t1 = t2). By taking this principle for granted, we get what we could call the Kim universalist structuralist view of events. Arguably, this is the standard standpoint in the structuralist camp and thus we may say that the typical universalist is a Kimian universalist in that he upholds Kim Identity Condition. As we shall see, however, there will be reasons to consider a version of universalism that rejects this identity condition for events. It should be noted that, once tropes are brought into the picture, we could have a tropist structuralist account of events, according to which an event is the exemplification by a concrete particular of a trope at a certain time. And if we include Kim Identity Condition in it we get, let us say, a Kim tropist structuralist view. However, with tropes in our ontological inventory, this option may not be the best one. It has been argued in fact that tropes as such could play the role of truthmakers (Mulligan et al. 1984) and causal relata (Williams 1953, Campbell 1981). Consider the two round coins again. One of them, coin a, has a certain round trope, r1, and the other, coin b, a different round trope, r2. Consider further two tokens of “this is round”, one uttered in relation to a and the other in relation to b. 16

Throughout this paper, in focusing on pain so as to have at hand a clear-cut example of mental property I freely use terms such “feeling pain”, “having pain”, “pain” and the like as equivalent, in order to talk about a certain property. However, I also use “pain” to talk about events; the context can disambiguate.

68 Clearly, they must have two different truthmakers. From the universalist’s standpoint we cannot identify these truthmakers with the properties corresponding to the two tokens of “round” involved in the two statements. For the two properties are one and the same universal. An appeal to a and b themselves seems necessary to distinguish two truthmakers and thus we are led to the view that the truthmaker is a structured entity involving both a property and an object exemplifying the property. But from the point of view of the tropist there are two available properties, namely r1 and r2. Thus, the token of “this is round” uttered in relation to a can be taken to have r1 as truthmaker and the token of “this is round” uttered in relation to b can be taken to have r2 as truthmaker. The two statements are then granted two distinct truthmakers, as it should be, without having to admit that a and b themselves, nor for that matter the property2, roundness, are constituents of the truthmakers in question. A similar trick can be played with causation. To adapt an example by Yablo 1992, consider two pigeons, p1 and p2, who are conditioned to peck at red. Pigeon p1 is presented with a red triangle and is thus caused to peck, while the other, p2, is similarly caused to peck by being presented with a red square. From the point of view of the universalist, one property, redness, is involved in both episodes of causation, but we cannot say that it is the cause in both cases, for clearly the pecking by p1 has one cause and the pecking by p2 another distinct cause. But the tropist can say that there are two properties at play, a red trope, r1, possessed by the triangle and a red trope, r2, possessed by the square. And the possibility of identifying the two causes with these two properties, respectively, is thus available. Given that events are, as we have seen above, the entities that play the roles of truthmakers and causal relata, we can, according to this approach, identify them with tropes. We can call this the non-structuralist tropist account of events. Indeed it seems to be the most natural choice, perhaps the standard one, and thus I shall take it as the default view in discussing tropism.17 Tropes, as contrasted with the events of the structuralist brand, are conceived of as basic, unstructured, particulars capable of having properties (to the extent that they belong to classes of resemblance), just like Davidsonian events. Thus, I think that tropes, inasmuch as the non17

The choice in question is not however the standpoint taken for granted in Robb 1997, where tropes and events are kept distinct. Robb hints (1997,181, not with my terminology of course) at the structuralist tropist view of events, but does not commit himself to it, either. As he has put it in correspondence, he is in that paper “largely neutral on the nature of events”.

69 structuralist tropist account of events is accepted, can be identified with Davidsonian events. 4.

Reductive Physicalism

The current Zeitgeist favours a physicalist outlook, which presupposes, inter alia, the idea that the physical domain is “causally closed”. This can be roughly put as follows: Closure If a physical property or event is causally brought about, it is brought about by physical properties or events. This principle of causal closure is in tension with the intuition that, as it has been said, “mind matters”, an intuition that leads us to uphold, at least prima facie, the following tenet: Influence Mental events and properties can have a causal influence on the physical world. The tension arises in view of our inclination to accept an independently plausible assumption about causation, namely: Exclusion There is, at least in typical cases, no causal overdetermination. At some point in the last century, before functionalism took the lead, a different doctrine featured as the most successful account of the mental in analytic quarters: the reductive physicalism proposed by Place, Smart, Feigl and others. It appeared to reconcile Closure and Influence, by being centred around this thesis: Type Identity Any mental property is identical to a physical property.18 18

As I see it, that two properties, e.g., having pain and having C-fibres firing, are identical means in practice that there are two predicates, “having pain” and “having C-

70 Clearly, Type Identity entails the following: Token Identity Any mental event is identical to a physical event (the mental event of x’s having a mental property M is identical to an event such as x’s having a physical property P). Of course, if mental properties and events are physical, they can have causal influence without violating Closure. The reconciliation of the latter with Influence thereby follows. 5.

Nonreductive Physicalism

In spite of this success, reductive physicalism has fallen in disgrace, mainly because of the multiple realizability argument put forward by Putnam and others. According to this argument, it is plausible to think that there are creatures, endowed with the mental properties that we humans possess, which simply do not have the physical properties that, according to the reductive physicalist, would be identical to the mental properties in question. Thus, for example, there could be a sophisticated robot and a Martian humanoid, who are made up of inorganic matters and thus do not have C-fibres. Accordingly, they cannot have the property of having Cfibres firing and yet, for all the evidence we have, they could be in pain. Thus, the property of having pain cannot be identical to the physical property of having C-fibres firing, contrary to what the reductive phyicalist proposes (on the basis of the neurophysiological evidence that correlates having pain and having C-fibres firing in humans). At most, we can say that pain is some sort of “higher-level” property which is “realized” by different “lower-level” physical properties in different creatures. 19 In fibres firing”, which, albeit expressing two different concepts, somehow correspond to one and the same property (a universal, from the point of view of universalism). 19 We need not concern us here with what realization precisely is. It is, we may say, a relation that links a realizable, higher-level, (mental) property to another, lower-level, (physical) property which works as “realizer”. But this of course is at best just useful terminology. Yablo 1992 makes the interesting proposal that realization is a determination relation, but there are good reasons, I think, to reject this idea (cf. Ehring 1996, Funkhouser 2006, section 4, Gozzano 2008, section 3), which will not then be endorsed here (note that, if Yablo’s thesis is not assumed, the terms “higher-

71 humans, the lower-level property would be C-fibres firing. What about the robot and the Martian? We can imagine, say, that the robot was designed so as to realize pain by having R-fibres firing and the Martian, as it turns out, realizes pain by having M-fibres firing. These science-fiction examples make the case in a particularly vivid way, but they are not strictly speaking necessary. It is enough to consider the fact that our brains are quite different from those of other animals to which we attribute at least some of the mental properties that we attribute to ourselves. This by itself casts doubt on the idea that we and these fellow animals can share physical properties that can be identified with the mental properties that we and the fellow animals are all capable of exemplifying. In sum, the following thesis came to be widely accepted: Multiple Realizability The same mental property can be realized by different physical properties in different species (or even, as we shall note below, in different individuals of the same species or in the same individual at different times). Now, Multiple Realizability immediately leads to a thesis that directly contradicts Type Identity, namely: Distinctness Mental properties and physical properties are distinct.20

level” and “lower-level” must be understood in different ways, depending on whether we are discussing the determination or the realization relation). There is a clear sense in which having pain is a property that admits of determinates, such as having a toothache, having a throbbing toothache, down to ultimate determinates for which we presumably have no adequate terminology. Correspondingly, we should assume that properties such as having C-fibres firing, having M-fibres firing (see below), etc. have corresponding determinates. For simplicity’s sake, however, I shall speak as if pain, having C-fibres firing, having M-fibres firing, etc., had no determinates. This inaccuracy will be immaterial for present purposes. 20 As is well-known, another argument for Distinctness was proposed by Davidson on behalf of his anomalous monism (see his 1980). It is based on the assumption that while physical properties grant the existence of strict physical laws, mental properties do not allow for strict psychological and psycho-physical laws, which is taken to imply that physical properties cannot be identical to mental ones. Although Davidson’s anomalous monism has raised much discussion, the assumption in question has never

72 It is tempting in the light of this argument to retreat to a relative reductive physicalism, according to which there is, contrary to our prima facie commonsensical expectations, e.g., no single pain. There are rather different species-specific kinds of pain: pain-for-humans, pain-forMartians, pain-for-robots, etc. (and similarly for other mental properties) (Lewis 1969, Kim 1972). The idea is that specific forms of Type Identity can still be true if mental properties are relative in this way (pain-forhuman is identical to having C-fibres firing, pain-for-Martians to having M-fibres firing, etc.). It has been argued however that, in view of the phenomenon of brain plasticity, even this is implausible (Horgan 2001). Take a physical property P that seems to be a good candidate for being identified with a human mental property M. It may happen that a subject who, because of a brain damage, is incapable of exemplifying the property P still exemplifies property M and this is so somehow by virtue of exemplifying another physical property P΄, the exemplification of which is compatible with the brain damage. This suggests not so much that M is identical to P, let alone P΄, but rather that M was at first realized by P, and later, after the brain damage, by P΄. In other words, M would be realizable in different ways in individuals of the same species and in the same individuals at different times. Lewis 1969 admits not only the possibility of distinct pains such as pain-for-humans and pain-for-Martians, but even that of different pains such as pain-for-Lewis and pain-for-Putnam. A philosopher who allows for this might end up distinguishing even between pain-for-Putnam-at-time-t1 and pain-for-Putnam-at-time-t2 in order to stick to Type Identity in spite of the phenomenon of brain plasticity. Perhaps this is not a very appealing option, as it runs too much against our commonsensical expectations. Be this as it may, Reductive Physicalism has increasingly lost consensus. A nonreductive physicalist paradigm has come to the fore, in an attempt to retreat to a form of physicalism compatible with both Distinctness and Closure. This physicalism is nonreductive in that it accepts Distinctness, thereby making room for Multiple Realizability. In addition to Distinctness, two main theses are typically appealed to in order to characterize this doctrine and make it a form of physicalism, as its appellation demands. One is Token Identity. The other is: obtained the widespread approval encountered by Multiple Realizability and indeed the former is much more dubious than the latter.

73 Supervenience Mental properties supervene on physical properties, in that, necessarily, for every system (organism, creature) x and mental property M of x, x has a physical property P such that necessarily whatever has P also has M.21 The nonreductive physicalist replaces Type Identity with at least one of them. Both in fact appear sufficient to rule out purely mental substances, entities that have mental properties without having physical properties, and to establish some sort of dependence of the mental on the physical.22 Before going ahead, some (partly terminological) considerations about Supervenience will be useful. Consider an individual x with mental property M. By Supervenience, it must be the case that x also has a corresponding physical property P. We can then say that in this case M supervenes on P. Intuitively, P is the physical property that realizes M. 23 It should be noted that, given the modal force of Supervenience, M and P are in a relation of necessitation. More specifically, in the light of Multiple Realizability, the necessitation in question is asymmetric: P asymmetrically necessitates M (Yablo 1992, 182). 24 Moreover, the 21

As is well-known, there are many species of supervenience and Kim has been most prominent in studying them (see his 1993). Here we are appealing to what Kim has called “strong supervenience” (1984). It is important to note that the variable “x” is not intended here to range over events, but rather over people, animals and the like (cf. Kim 2005, 33-34), so as to exclude that Supervenience immediately entails Token Identity. The idea is that these two principles should be seen as compatible (at least prima facie), but not such that the former entails the latter. 22 For present purposes this description of nonreductive physicalism suffices, although it is worth noting that there is no full agreement on how exactly theories of this kind should be characterized. For example, G. and C. MacDonald (2006, 541) speak of minimal physicalism, a view that espouses both Distinctness and Token Identity. On the other hand, Kim (1998, 12) calls in the same way an approach that, without Token Identity, couples Distinctness with something like Supervenience (actually Kim considers therein a weaker version of the latter principle, but he adds to it other principles that rule out certain “Cartesian” outcomes and that need not concern us). 23 In fact, once one accepts to talk in terms of “realization” of properties (as we are doing here) and buys Multiple Realizability, one typically also accepts the idea that an individual can have a mental property only by virtue of having a physical property that realizes it. 24 For, as Yablo notes (1982, 182), the phenomenon of multiple realizability can be taken to imply that “necessarily, for every mental property M, and every physical property P which necessitates M, possibly something possesses M, but not P”.

74 necessitation in question may well be nomic or physical, at any rate somewhat weaker than the (metaphysical or logical) necessitation involved in the determination relation.25 When a mental property supervenes on (is realized by) a physical property, something analogous can be said about corresponding mental and physical events. For when a mental event, such as Tom’s pain, occurs, we must admit that some creature exemplifies a mental property (in our example, Tom exemplifies being in pain). And this is so, presumably, even if we do not buy Kim’s property exemplification view of events. By Supervenience, then, there must also be a physical event resulting from the creature’s exemplifying a certain physical property on which the mental property supervenes. And the latter property, as noted, is such as to (asymmetrically) necessitate and be realized by the former. On this basis, we can also conveniently say that the mental event supervenes on (is realized by, asymmetrically necessitates) the physical one. The Acceptance of Token Identity still offers a way to address the tension between Closure and Influence. It does so, if we understand these principles, respectively, as follows: Closure 1 If a physical event has a cause, this cause is a physical event.26 Efficacy Mental events can at least sometime cause other events. Similarly, Supervenience also promises at least at first glance to keep Closure and Influence together. On the one hand, it offers a way to guarantee Closure, and more specifically its more specialized version Closure 1. Take an event which is a good candidate for having a mental cause, Tom’s yelling, as it follows Tom’s pain. By Supervenience, the pain 25

I take this to be the standard view among the physicalists who accept Supervenience. Similarly, in reductive physicalism the identity of a mental and a physical property is normally taken to be a matter of simply nomic or physical necessity. In fact, as clearly explained in Marras 2005, the necessity to be appealed to in these issues had better be nomic or physical, despite recent contrary opinions. 26 Kim distinguishes a weaker and a stronger notion of closure (see, e.g., Kim 1998). The latter somehow implicitly contains a version of Exclusion, which is however best viewed as a separate principle (Kim 2005, 51). Here we are thus appealing to the former.

75 supervenes on a physical event, say, an episode of C-fibres firing in Tom, and this physical event can be taken to be a cause of the yelling. On the other hand, Influence can also be accommodated, one may think, provided it is spelled out in a certain manner. This manner presupposes some bit of terminology, in the light of which a property P of an event c can be said to be relevant to c’s being a cause of a certain effect e (Braun 1995). Take a well-known example used by Braun (adapted from Dretske 1989). Wilma is a soprano who sings the word “shatter” at a high pitch and amplitude, in such a way that a nearby glass is caused to shatter. There is then a singing event with the property of involving a sound with high pitch and amplitude as well as that of involving a word token which means shatter. Intuitively, however, only the first of these properties is relevant to the event’s causing the glass to shatter. In other words, the singing event causes, qua having the former property, the shattering event. In general we can say: c, qua P (or by virtue of being P), causes e, where P is a certain property. By employing this terminology, Influence is then spelled out as follows: Relevance Mental properties are at least sometime causally relevant, i.e., there can be an event c which, qua M, causes another event e, where M is a mental property. Now, Supervenience at least prima facie suggests that mental properties are causally relevant and that therefore the reductive physicalist who endorses it has no trouble in holding on to both Closure and Influence (where the latter is understood as Relevance). This can be seen by noting that it is tempting to provide the following counterfactual analysis of causal relevance: an event c, qua P, causes an event e iff (i) c has the property P, (ii) c causes e and (iii) c would not have caused e, had it not had property P. Consider Tom’s pain at t, followed by Tom’s yelling at t΄. By Supervenience, the pain supervenes on a physical event, say, a firing of Tom’s C-fibres at t. In line with Closure, this causes the yelling. But Supervenience also tells us that the firing could not have occurred without being a pain, i.e., a mental event. And thus Relevance seems to be insured, given the counterfactual analysis of causal relevance. In fact, there appear

76 to be decisive objections to this analysis,27 but it is useful in letting us see a reason that can easily incline toward Supervenience. It should be noted that the terminology introduced in order to spell out Relevance also suggests this additional version of Closure: Closure 2 If a physical event e has a cause, then there is a physical event c and a physical property P such that c, qua P, causes e. In other words, the physical property P of c is relevant to c’s causing e. By a similar line of reasoning, Supervenience may seem to make Influence (in the form of Relevance) compatible with Closure 2 as well. Consider again Tom’s pain and the ensuing yelling. Even if admitted that there is a mental property of the former event, being a pain, which is causally relevant to the yelling, the latter may well have been caused by a physical event whose existence is granted by Supervenience, the one on which the pain event supervenes. And this physical event may surely have a physical property, say, being a firing of C-fibres, which is causally relevant to the yelling. 6.

Troubles for nonreductive physicalism

We shall now move on to some problems that the reductive physicalist must confront. They have to do in some way or another with a threat of epiphenomenalism. Once Influence is split into Efficacy and Relevance, we can say that epiphenomenalism shows up in any approach in which we are forced to deny at least one of them. To see how this could happen in nonreductive physicalism, we should focus on two different ways of cashing out Exclusion, depending on which version of Closure is privileged. In relation to Closure 1, we get: Exclusion 1 If an event c is a cause of an event e, then, in typical cases, no other event is a cause of e. With Closure 2 in mind, on the other hand, we land on: 27

Accordingly, Braun 1995 proposes an alternative based on a distinction between essential and contingent properties of events.

77 Exclusion 2 If an event c, qua P, causes an event e, then, in typical cases, it is false that c, qua P΄, causes e, where P and P΄ are distinct properties. Dretske’s soprano example can be used to motivate both. Mary, ignorant of the laws of acoustics, may propose, in seeing the shattering of the glass, that it was caused by some supernatural event (simultaneous with the singing). Suppose she is told that the high pitch and amplitude involved in the singing are enough to make it a cause of the shattering. If she does not withdraw her appeal to a supernatural event distinct from the singing, we would think that she is being unreasonable. Why? Because we presuppose Exclusion 1 in providing causal explanations (and we see no reason to view this causal episode as an “untypical” one involving overdetermination). Suppose Mary withdraws her original hypothesis and admits that the singing caused the shattering. Nevertheless, she insists that, although the high pitch and amplitude involved in the event had a role in the event’s causing the shattering, a role is also played by the fact that it involved the uttering of a word with the meaning shatter (she argues that, had the soprano uttered a word with the same pitch and amplitude but with a different meaning, the glass would not have shattered). Surely, we would still find this an unreasonable proposal and (quite apart from the implausibility of the idea that meanings can have this sort of causal power) this is presumably because in causal explanations we presuppose not only Exclusion 1, but also Exclusion 2: if the singing, qua involving a certain pitch and amplitude, caused the shattering, it cannot be the case that it caused it also by virtue of involving the uttering of a word that means shatter (again, there is no reason to think that the episode in question is “untypical” in that it involves overdetermination). Let us now see how, by resorting to one or the other of these principles, one can try to show that the nonreductive physicalist is guilty of epiphenomenalism. We shall consider two kinds of nonreductive physicalist. One who wishes to adhere at all costs to the Kim structuralist universalist account of events and another who is prepared to give it up. There is an important preliminary step to be made explicit, which has independent ontological interest, regardless of whether one worries about epiphenomenalism or not. Here it is. On the basis of his own structuralist universalist account of events, Kim has argued that Token Identity collapses into Type Identity and thus the nonreductive physicalist, having

78 rejected the latter thesis, is not free to endorse the former (Kim 1998, 60).28 Kim’s argument can be reconstructed as follows. Suppose events are viewed as structured items whose constitutive properties are universals and which obey Kim Identity Condition. Consider an arbitrary mental property M, a certain universal. We can assume that M is exemplified, if not in this world, in some nomically possible world w. If it is exemplified in w, then there must be in w a mental event [x1, M, t1]. But then, by Token Identity, there is also in w a physical event [x2, P, t2] to which the former is identical. This is physical in that P is a physical property. Now, by Kim Identity Condition, it follows, inter alia, that M = P. Since M was arbitrarily chosen we can conclude that any mental property is identical to a physical property, i.e., Type Identity is true. This argument leads the supporter of nonreductive physicalism to a dilemma. Either she rejects Token Identity or she had better find an account of events different from the one provided by Kim. Consider now the nonreductive physicalist committed to the Kim structuralist universalist account of events. She has to react to the above argument. Suppose she does this by telling us that she has no problem in rejecting Token Identity; she can still uphold Supervenience and be a happy with it. As Kim has also argued in various works,29 there is a problem with this move in that Distinctness, Supervenience, Closure 1 and Exclusion 1 provide a route to epiphenomenalism. For they jointly lead us to reject Efficacy and, we may add, Relevance. The basic idea is this. Take a mental event m which seems to be a good candidate for being causally efficacious and more specifically to be such that it causes, qua M, another event e, where M is a mental property. For example, m could be Tom’s pain at time t. Then, the event e putatively caused by the pain could be, say, Tom’s yelling at time t΄. By Closure 1, this effect must have a physical cause and, given Supervenience, the plausible candidate is the physical event on which the mental one supervenes. In our example, we might say it is a certain firing of Tom’s C-fibres occurring at t. But then, by Exclusion 1, the pain cannot be a cause, and a fortiori cannot be an event which, qua pain, causes something else. This is so, unless the pain is identical to the Cfibres firing in question. But, given Kim’s view of events, Distinctness rules this out, for the reasons explained above. To put it otherwise, the pain could be identical to the firing of the C-fibres only if Token Identity were 28 29

A related point is made in Kim 1972. See, e.g., his 1993, 1998, 2005.

79 true. But this, from the perspective of Kim’s account of events, requires Type Identity, which is precisely what Distinctness rules out. Let us now turn to the nonreductive physicalist who, possibly in response to the argument we have just seen, does not endorse the Kim universalist structuralist view of events. This gives her elbow room to uphold Token Identity, which in turn grants her Efficacy (for reasons explained above). The trouble now is that, given Distinctness and Closure 2, Exclusion 2 immediately leads to the denial of Relevance. For Closure 2 asserts that, if an event e is caused by a mental event c, this can happen only if c has a physical property by virtue of which c causes e. This implies that c is physical, which, given Token Identity, may well be the case. However, by Exclusion 2 and Distinctness, we must admit that it is not by virtue of some mental property that c causes e. Hence, Relevance must be rejected. 30 We have seen that the nonreductive physicalist can uphold Token Identity by retreating from Kim’s account of events. We now see that, by so doing, she lands on a form of epiphenomenalism as a result of losing Relevance. There is another problem worth noticing. Token Identity (and Efficacy with it) becomes an option once we set Kim’s account of events aside. But this move leaves one with the burden of providing an alternative to Kim’s account, hopefully an alternative that can be independently motivated. This is particularly important if the problem with Relevance can somehow be defused. Let us take stock. If the reductive physicalist accepts Kim’s account of events, she must abandon Token Identity and retreat to Supervenience. But in doing so, she must reject Efficacy and Relevance (given Exclusion 1 and Closure 1). If, on the other hand, the reductive physicalist rejects Kim’s account of events, she can uphold Token Identity. This allows her to maintain Efficacy but she is forced to give up Relevance (given Exclusion 2 and Closure 2), and, at any rate, she must tell us which positive account of events, compatible with Efficacy, she is prepared to put forward in order to replace Kim’s account.31 30

This and the above argument attributed to Kim are two versions of an “exclusion argument” whose origins can be traced back at least to Malcolm 1968 (Robb and Heil 2003). The two versions are not always sharply distinguished, although sometime they are (see, e.g., Yablo 1992, 180). 31 This section illustrates, inter alia, how Kim’s well-known attacks on nonreductive physicalism are dependent on his view of events. This is a point noted by Marras in some of his works (e.g., in his 1993) and explicitly taken up in Marras and YliVakkuri 2008.

80 7.

Tropism and NENRP

We have noted that someone who does not adhere to Kim’s account of events should provide an alternative picture. It could be said that this alternative picture has been with us for years: it is the one due to Davidson. But Davidsonian events, I argued, can best be seen as tropes. Alternatively, tropes can enter the scene, by adopting a structuralist tropist conception of events. In sum, in providing this alternative picture one could resort to tropism. This is no surprise in view of section 3, above. Now, Robb 1997 has argued that, by adopting tropism, we can quite naturally be nonreductive physicalists without acquiescing to epiphenomenalism. 32 Robb’s proposal is discussed in Heil’s, Maurin’s, and Gozzano’s contributions to this volume and thus additional information on it can be gained from those sources. Here I will summarize it in a way that is convenient for my present purposes. The gist is this. We should confront the crucial datum of section 2, as regards a mental predicate such as “feels pain”, by appealing to the idea that tropes can partially resemble each other. The idea is that we correctly apply this predicate to Tom the human, Roby the robot and Marsie the Martian, insofar as there are three partially resembling tropes, h, r and m, instantiated1 by Tom, Roby and Marsie, respectively. In other words the predicate corresponds to a class of partially resembling tropes, comprising h, r and m. This can be seen as the mental property2 of pain, exemplified2 by Tom, Roby and Marsie. In this sense, all these tropes are mental. On the other hand, these three tropes can each belong to classes of perfectly resembling tropes, with respect to which we deploy predicates such as “having C-fibres firing”, etc. We thus have three different physical properties2 (each of which is exemplified2 by Tom, Roby and Marsie, respectively) to which the tropes belong. In this sense, they are also physical. We thus have Distinctness and Multiple Realizability. For example, there are different ways of instantiating2 pain: by instantiating2 the property2 having C-fibres firing (i.e., by instantiating1 a trope belonging to the latter property2), by instantiating2 the property2 having R-fibres firing, etc. All this is compatible with Token Identity. This is immediately clear if events just are tropes, as the non-structuralist tropist conception has it. Consider again the event of Tom’s being in pain at time t. This is just 32

In the context of the problem of mental causation, a tropist point of view analogous to that of Robb 1997 is also presented in Ehring 1996. Ehring however does not make the claim that tropism can be used to fully eliminate epiphenomenalism from nonreductive physicalism. More on this in note 41 below.

81 trope h. This trope belongs to the perfect resemblance class of having Cfibres firing and thus it is physical.33 What I have outlined is a view which we could call tropist nonreductive physicalism. As noted above, once Token Identity is secured, so is Efficacy. But what about Relevance? On a superficial reading, it may seem that Robb claims in his 1997 that tropist nonreductive physicalism also secures the latter, but this is so only because he considers a thesis, different form what I have called “Relevance”, which he however calls with the same name. It is a thesis also discussed in Maurin’s and Heil’s contributions to this volume,34 wherein they agree (implicitly in Heil’s case) that Robb’s tropist approach indeed accommodates it. In this paper I shall call it “Robb Relevance”. Here it is: Robb Relevance Mental properties are (sometimes) causally relevant to physical events. By this, Robb simply means that an effect may well be caused by an event somehow involving a mental property. And he thinks that his tropist proposal grants this. For example, an event such as Tom’s pain does involve a mental property. This is so, if by “property” we mean property1, i.e., trope. And in fact (to continue with the above example), the pain in question is trope h (given the identification of tropes with events35). Now, the pain can cause something, since h is not only mental, but also, as explained above, physical. In other words, Robb Relevance is gained by appealing to properties1 of concrete individuals. But Relevance has to do with properties of events and thus addressing the former does not amount to addressing the latter. Nordhof 1998 has thus criticized Robb for not going far beyond Davidson: Robb and Davidson are in the same boat, in having troubles with Relevance; neither really grants us that a mental event, qua mental, can cause something. Robb has replied that, by bringing tropes explicitly into the picture, he goes beyond Davidson by explaining how (mental) properties are involved in causation (2001, 91). This may well be true, but, as he admits, “we have independent reasons (examples 33

Suppose on the other hand that we settle on the structuralist tropist account. In this case the event in question is [Tom, h, t]. This is physical in that its constitutive property belongs to the perfect resemblance class of having C-fibres firing. 34 Still under the heading “Relevance”, in Maurin’s case. 35 Alternatively, given the tropist structuralist account of events, the pain involves the trope h, by having it as a constituent, since the pain is [Tom, h, t].

82 such as the soprano) for thinking that the qua problem arises for events” (1997, 191). He adds that there is no similar qua problem for properties and thus for tropes (Heil and Maurin agree with this in their contributions to this volume). But this may be true at best inasmuch as tropes are properties. The trouble is that, as we have seen, tropes can also be viewed as events. And thus the qua problem arises for tropes insofar as they are events. In fact, mutatis mutandis, the “qua talk” involved in Relevance makes sense in tropism just as in universalism: we can claim that the event f, qua instantiating0 F, i.e., by virtue of being a trope belonging to property2 F, can cause another event. 36 But then we have to address Relevance. More precisely, we have to address it in a form appropriate to tropism. I leave to the reader a formulation based on the structuralist tropist conception of events and focus simply on the version based on the default option, according to which events are tropes: Tropist Relevance There can be an event (trope) c which, qua instantiating0 M, causes another event e, where M is a mental property2. Robb addresses neither Tropist Relevance nor a relative of it based on a different (but still tropist-friendly) view of events.37 By capitalizing on this, C. and G. MacDonald (2006, 555-56) argue that a tropist approach lacks the resources to salvage Relevance. They contrast tropist approaches 36

From the structuralist perspective, the qua problem arises for events, conceived of as involving tropes as constituents. Given this, we can claim that a certain event [x, f, t], by virtue of encompassing a trope f exemplifying0 property2 F, causes another event. 37 Robb has informed me in correspondence that he does not look at Tropist Relevance with favour on two counts. First, he is not committed to the view that events are tropes and, second, he regards the idea that types are causally relevant as a category mistake. According to Robb, “this is clearest, if types are classes: Nothing causes anything in virtue of being a class”. As to the first point, let us emphasize that we are discussing here the merits in philosophy of mind of tropism cum the identification of tropes with events, independently of whether or not Robb is inclined to accept this identification. As regards the second point, it should be noted that Tropist Relevance does not assert that a class causes anything, but simply that a trope’s belonging to a certain resemblance class is relevant to the trope’s causing something. A trope c belongs to a resemblance class M (and thus instantiates0 M) somehow by whatever “brute fact” makes it the case that it and the other members of the class are naturally (and not as a result of an arbitrary convention) grouped in the class in question and not in others. If this is kept in mind, it seems to me perfectly legitimate to say that a trope can cause something by virtue of belonging to a certain class, as opposed to another.

83 (Robb’s in particular) with a version of universalism that they put forward, which is claimed to have all the virtues of tropist approaches (as they are used in an attempt to support nonreductive physicalism), plus one. Namely, the resources to accommodate Relevance. I shall discuss their proposal in more detail below. For the time being, it is important to note that it crucially relies on the one hand on the acceptance of Supervenience (or something similar) and on the other hand on the rejection of Exclusion 2. The latter is replaced by something like: Exclusion 2* If an event c, qua P, causes an event e, then, in typical cases, it is false that c, qua P΄, causes e, where P and P΄ are distinct and independent properties. (See MacDonald and McDonald 2006, 544.) This principle differs from Exclusion 2 because of what is provided by the addition of the word “independent”, which I have thus italicized for emphasis. Indeed, without a word like that the principle cannot be true.38 We can resort again to Yablo’s pigeon example to see this. If a pigeon is caused to peck by the redness of a triangle, surely we want to say that the redness is causally relevant for the pecking, yet we do not want to deny that the determinate shade of red of the triangle is also relevant. This seems to be due to the fact that the determinable redness and the determinate shade in question, albeit distinct properties, are not independent. But how should we spell out “independent”? In essence, according to MacDonald and MacDonald 2006, if two properties stand in a necessitation relation (whether nomic/physical or logical/metaphysical), they are not independent. Thus, in particular, according to this proposal, if a mental property M supervenes on (is realized by) a physical property P, then M and P are not independent.39 For present purposes, we can be content with this bit of knowledge about what independence amounts to. With Exclusion 2* in view, the above argument against Relevance cannot get off the ground. Take a mental event m which causes e. True, by Closure 2, this can happen only if c has a physical property P by virtue of which m causes e. However, given Supervenience, P may well be the physical property on which M, the mental property that makes m mental, 38

Unless we interpret “distinct” loosely, which may be intended by some of those who discuss the problem of mental causation by deploying something along the lines of my Exclusion 2; Yablo 1992 seems to me a case in point. 39 Yablo 1992 makes a similar proposal, but in his view realization is determination.

84 supervenes. Since Supervenience tells us that P and M are in a necessitation relation [M asymmetrically necessitates (is realized by, supervenes on) P], Exclusion 2* cannot prevent us from considering M a property such that m, qua M, causes e. That is, Relevance is not endangered. Now, if this is right, (Tropist) Relevance is similarly compatible with tropist nonreductive physicalism, for this doctrine implies Supervenience, more precisely the tropist version of it, namely: Tropist Supervenience Mental properties2 supervene on physical properties2, in that, necessarily, for every system (organism, creature) x and mental property2 M of x, x instantiates2 a physical property2 P such that necessarily whatever instantiates2 P also instantiates2 M. This can be seen by adapting to the way tropism has been characterized here the argument given by Robb 1997, 188.40 In sum, once Exclusion 2 is given up in favour of its more plausible relative, Exclusion 2*, and (Tropist) Supervenience is given the appropriate prominence in tropist nonreductive physicalysm, we see clearly that this approach need not reject (Tropist) Relevance. For with (Tropist) Supervenience, mental properties2 are dependent on physical properties2 and thus we cannot appeal to Exclusion 2* to reject Relevance.41 In sum, the tropist nonreductive physicalist may well be a non-epiphenomenalist

40

This argument crucially depends on the principle, noted in section 2, according to which, if two tropes are similar (to a certain degree) they are so necessarily. 41 Ehring 1996 admits that in a tropist ontology Token Identity (a “property instance identity thesis”, in his terminology) and thus Efficacy can be secured, but, he notes, Relevance is not thereby secured (p. 465). He also considers something like Exclusion 2* and entertains the possibility that by means of it Relevance can also be regained. However, he concentrates only on the option of cashing out the notion of dependence involved in Exclusion 2* in terms of the determination relation. Since he argues against Yablo 1992 that mental properties are not determinables of physical one, he does not present in the end the tropist ontology as a way of fully amending nonreductive physicalism from epiphenomenalism. However, if, following MacDonald and MacDonald 2006, we are more liberal and admit that an appropriate relation of supervenience is sufficient to give us the dependence appealed to in Exclusion 2*, then tropism can secure (Tropist) Relevance (and thus fully avoid epiphenomenalism) in the way suggested above.

85 nonreductive physicalist, in other words a supporter of what I have dubbed NENRP in the introduction. 8.

Heil universalism and NENRP

Tropist nonreductive physicalism relies on a crucial aspect of the tropist ontology. Namely that a predicate may apply to a concrete individual for two different reasons. It might apply because the individual instantiates a trope belonging to a class of perfectly resembling tropes to which the predicate corresponds or because the individual instantiates a trope belonging to a class of partially resembling tropes to which the predicate corresponds. In the latter case, it may well happen that the members of the partial resemblance class are grouped into subclasses of perfect resemblance with predicates corresponding to them. An example of the former kind is given, presumably, by “round” and an example of the latter kind by “having a weight in between 5 and 10 kilos”. The tropes in the partial resemblance class corresponding to this predicate are grouped in turn into classes of perfect resemblance for which we can have corresponding predicates, such as “having a weight of exactly 5 kilos”, “having a weight of exactly 6 kilos”, and so on. Similarly, it is argued, “pain” corresponds to a class of partially resembling tropes involving subclasses of perfect resemblance captured by predicates such as “having C-fibres firing” and “having M-fibres firing”. This paves the way for Distinctness and Multiple Realizability. In his contribution to this volume, Heil claims in effect that tropist nonreductive physicalism is not essentially tropist, in the sense that the universalist can “simulate” the above mentioned crucial aspect of tropism, by resorting to the idea that there can be resemblances among universals. She could say for example that “round” applies to two distinct round coins because they both exemplify the universal roundness, whereas “coloured” applies to both coins because they respectively exemplify two distinct universals, say, a green one and a yellow one, so to speak, which resemble each other. In the former case, the predicate corresponds to a universal, whereas in the latter case it only corresponds to a family of distinct but similar universals. Analogously, “having C-fibres firing” and “having Mfibres firing” could be taken to correspond to two different but mutually resembling universals, i.e., having C-fibres firing and having M-fibres firing, and “pain” to the family of mutually resembling universals that

86 comprise both. There is however no universal to which the predicate “having pain” corresponds, according to this picture. At most, “having pain” can be taken to correspond to a family of similar universals. This many-membered family, having pain, is of course distinct from each member of the family (or for that matter, from the one-membered families which have as unique members the various members of having pain, i.e., having C-fibres firing, having M-fibres firing, etc.). However, having pain can be, so to speak, multiply realized, in that each of its different members can be exemplified by various concrete individuals. In this sense, Distinctness and Multiple Realizability are guaranteed, pretty much in the way they are from the tropist’s point of view. Let us say that a universalist who follows this road accepts Heil universalism. In practice, Heil universalism consists in bringing in families of similar universals in order to account for the crucial datum of section 2. Note that, once this is done, it is appropriate for the universalist to distinguish between, say, “propertya” and “propertyb” and “exemplificationa” and “exemplificationb”, just like the tropist distinguishes between “property1” and “property2” and “exemplification1” and “exemplification2”. For example, one should assert that both Tom and Marsie feel pain in that both exemplifyb the propertyb having pain, which is a family of similar universals. And this happens because there is a propertya, having C-fibres firing, exemplifieda by Tom, and another propertya, having M-fibres firing, exemplifieda by Marsie. With this in mind, we can see clearly that the Heil universalist can buy Token Identity and Supervenience (and thus Efficacy and Relevance) just like her tropist colleague. As regards Token Identity, she can say that an event such as Tom’s being in pain, i.e., Tom’s exemplifyingb the propertyb of having pain, is simply Tom’s instantiatinga the propertya of Cfibres firing, a physical event. But it is not Tom’s instantiatinga a universal that could also be exemplifieda by a Martian without C-fibres, because there is no such universal. Moreover, she can hold on to her own version of Tropist Supervenience, by replacing “property2” and instantiation2”, with “propertyb” and “instantiationb”, respectively. Thus, for example, the event of Tom’s being in pain, qua involving a propertya, having C-fibres firing (which belongs to a mental propertyb, a family of similar “mental universals”, the having pain family), may well cause another event e. In sum, a Heil universalist can accept NENRP just like a tropist.

87 9.

Standard universalism

But is it a good idea for a universalist to be a Heil universalist? Surely, there can be resemblances among universals and thus Heil universalism may seem tempting.42 But the basic point of this doctrine, as I understand it, is that there are universals that resemble each other as a matter of brute fact, just as it is the case with tropes in a tropist approach, where resemblance of tropes is a primitive notion. However, not having to resort to a primitive notion of resemblance has been seen as an advantage that the universalist should try to preserve over the tropist: Such unanalyzable, primitive, resemblance of universals I regard as a fall-back position for the Realist about universals. It may in the end have to be accepted, at least for some cases. But it is an uncomfortable compromise, true to the superficial appearances, but lacking the deep attractiveness of a theory that always takes resemblance to involve some degree of identity.

Armstrong 1989, 105. As is evident from this quotation, Armstrong strives to understand resemblance of universals, without taking it as primitive. For the obvious thing to say from the point of view of the universalist is that, in general, the resemblance of two entities is not a brute fact, because it is due to the sharing of a number of universals by the entities in question. Thus, in particular, if the entities in question are universals, their resemblance should be understood by viewing them as lower-level universals that share some higher-level universals. 43 Let us look at some concrete examples. Two different determinate shades of red resemble each other in that they share the higher-level property of being red (say, power of radiating light 42

Indeed in Orilia 2006a, I myself invoked resemblance of universals in order to reproduce the tropist account of multiple realizability from the universalist’s point of view. However I also expressed therein some perplexities, which I shall try to make explicit in the following. 43 As I understand him, Armstrong admits that this is what happens in some cases, in particular when functional properties and laws of nature are involved (1989, 105), although he also thinks that this is not necessary in other cases, e.g., those regarding necessary truths of determination, such as that red is a colour (1989, 100) or quantity properties such as having a mass and being an ounce in mass (p. 101). But we can leave these subtleties aside for present purposes.

88 with a wavelength within a certain range). Each such shade resembles any other colour shade more than it resembles, to take an extreme case, the microphysical property of having negative spin. This is so because they all have, whereas the latter does not, the higher level property of being red. Similarly, being sugar and being salt are similar, because they share the higher-level property of solubility; being nephrite and being jadeite, because they share the higher-level property of being jade; being sapphire and being ruby because they share the higher-level property of being made of molecules of Al2O3. This position, which in contrast to Heil universalism appeals to higher-level universals to ground the resemblance of lower-level ones, could perhaps be dubbed with some legitimacy standard universalism. At any rate, this is how I shall call it here. From its perspective, the nonreductive physicalist may well agree that two properties such as having M-fibres firing and having C-fibres firing resemble each other. But not because they do so as a matter of brute fact. Rather, because they share the property of being realizers of pain. And this of course implies that there is a mental property, having pain, instantiated on the one side by creatures with M-fibres firing and, on the other side, by creatures with C-fibres firing. For without the mental property, the resemblance between the two physical properties would be groundless. 10.

Monist universalism and NENRP

Following Heil universalism, a reductive physicalist can say that the predicate “pain” applies to an individual, not because this individual shares with others a certain universal to which the predicate corresponds, but because the individual exemplifies a property, such as having C-fibres firing, which belongs to a family of mutually resembling universals, a family which includes having M-fibres firing. According to this picture, when Tom is in pain, there is just one event, Tom’s havinga the propertya of having C-fibres firing. The event of Tom’s havinga pain is out of the picture, for there is no propertya of being in pain. Hence, the issues of whether the event of Tom’s havinga pain is causally efficacious and the propertya of havinga pain is causally relevant do not arise. The road toward NENRP is paved. But what about the standard universalist? The road for her is not paved in the same way. She is forced to accept, we may say, two propertiesa: on the one hand, havinga pain (which can be instantiateda by

89 Tom, Roby and Marsie) and, on the other hand, havinga C-fibres firing (which cannot be instantiateda by Roby and Marsie). Now, this prima facie suggests that the standard universalist is also committed to distinguish, in the case of Tom’s pain, between two events, Tom’s havinga pain and Tom’s havinga C-fibres firing. In other words, the standard universalist seems to be bound to the denial of Token Identity. She can be, it would seem, a nonreductive physicalist cum Supervenience, but not cum Token Identity. By (legitimately) replacing Exclusion 2 with Exclusion 2*, the standard universalist who wants to be a nonreductive physicalist can claim (like the tropist) that Supervenience is enough to grant relevance. But for the physicalist the key to Efficacy is nothing less than Token Identity. Does this mean that the nonreductive physicalist, if a standard universalist, cannot coherently avoid the rejection of Efficacy and must acquiesce to epiphenomenalism? MacDonald and MacDonald 2006 appears to offer a way out of this predicament. The MacDonalds note that a red box exemplifies the property of being coloured “just by exemplifying the property of being red” and put forth the view that similarly an event e can exemplify a mental property M “just by” exemplifying a physical property P (p. 563). This happens when P is a lower-level property that realizes a higher-level mental one, namely M. In an attempt to illustrate this position, let us consider once more the event of Tom’s pain at t. The idea seems to be this. Having pain is realized in this case by a physical property that Tom exemplifies at t, having Cfibres firing. Therefore, the event in question has the mental property of being a pain (or the having of a pain), by virtue of being a firing of Cfibres. Similarly, “a mental event can exemplify the property, being a thinking of Vienna, just by exemplifying the property, say, being neurochemical event α” (2006, 563). The MacDonalds claim that this is compatible with the Kim structuralist universalist account of events, if it is recognized that only one member of the pair of physical and mental properties involved in these cases is a constitutive property of the event in question. 44 And in their physicalist perspective the chosen one is the physical property. Thus, in Tom’s example, the constitutive property is having C-fibres firing and the single event which is both a mental and a physical event is then [Tom, having C-fibres firing, t]. 44

In discussing the argument against nonreductive physicalism in Kim 2007, Marras and Yli-Vakkuri 2007 in effect consider this option (the alternative (e) in their section 7) and present it as an option that Kim might accept. They however underline that it involves the rejection of Kim’s identity condition for events (see below on this).

90 As the foregoing discussion suggests, the MacDonalds thus think that they can consistently maintain Token Identity (and thus Efficacy) in spite of being nonreductive physicalists committed to the Kim structuralist universalist account of events. They also accept Supervenience (more precisely, a close relative of it; see p. 565) and, as noted above, Exclusion 2* (rather than Exclusion 2). This allows them to claim that they hold a view that makes room for Relevance. Here is how they summarize their position (p. 565): (1) mental properties of persons supervene on their physical properties, and so (2) mental properties of events supervene on their physical properties. This is consistent with the view that an individual event can be an exemplifying of both a mental and a physical property (of a person), can be an instance of both a mental property and a physical property (of an event), and can be an instance of a mental property just by being an instance of a physical property (of an event).

But I am not quite convinced. Let us focus again on the event of Tom’s pain at t. As I understand them, the MacDonald’s would admit that, insofar as there is this pain, Tom exemplifies at t not only having C-fibres firing, but also having pain at t. In spite of this, they deny that there is the event [Tom, having pain, t] and acknowledge only the event [Tom, having C-fibres firing, t]. Hence, they seem committed to a denial of Kim’s existence condition for events: Tom exemplifies having pain at t and yet [Tom, having pain, t] does not exist. In general, according to the MacDonald’s, as I understand them, when an object, x, exemplifies two properties, P and Q, at a given time, t, and it makes sense to say that x exemplifies Q, “just by” exemplifying P, we should deny that Q is a constitutive property of an event, thereby acknowledging at most the event [x, P, t]. In our example, given that having pain is a higher-level property realized by the lower-level property of having C-fibres firing, it makes sense to say that Tom exemplifies the former “just by” exemplifying the latter and thus we should at most acknowledge the event [Tom, having Cfibres firing, t], thereby rejecting [Tom, having pain, t]. However, it seems problematic to me that we can discriminate between the two properties in question in this way. After all both are exemplified by Tom and thus, to the extent that events are exemplifications of properties by objects at times, I find it implausible to deny the existence of [Tom, having pain, t] (and, more generally, to reject Kim’s existence condition). If there is this event,

91 of course, given Distinctness and Kim Identity Condition, it cannot be identical to the event [Tom, having C-fibres firing, t]. In sum, Token Identity must go. But another way of making Distinctness compatible with the structuralist universalist conception suggests itself. Rather than saving Kim Identity Condition as the MacDonald’s try to do, drop this principle, by allowing that more than one property can be a constitutive property of an event.45 The idea is that, even though F and G are distinct, the instantiation of F by a concrete particular at a given time and the instantiation of G by the same particular at the same time may well be the same event, provided that F and G are linked by a necessitation relation such as supervenience, realization or determination. In sum, the suggestion is to replace Kim Identity Condition with a principle along these lines: Monist Identity Condition [x, P, t1] = [y, Q, t1] iff x = y & t1 = t2 & (P = Q or P and Q are in a relation of necessitation). A standard universalist who accepts this principle may be called a monist universalist. It seems to me that the monist universalist can coherently embrace NENRP. For, beside buying Relevance via Supervenience (in the way indicated above), she can also get Token Identity and thus Causal Efficacy by virtue of her acceptance of Monist Identity Condition. For instance, consider the event e of Tom’s being in pain. By Supervenience, there must also be the event e΄ of Tom’s having P, 45

The MacDonalds play with this option. They try to motivate it by appealing to Lombard’s view (1986), according to which events always involve changes. I think this way of supporting the option in question is misleading, because Lombard’s is not a view which implies that an event has more than one constitutive property. According to him, an event, e, is the exemplification of a dynamic property D at a time t by an individual x. For this exemplification to occur it must be the case that t is an interval comprising “sub-times” t1, t2, t3, etc., at each one of which x exemplifies a property in a corresponding sequence of static properties, S1, S2, S3, etc., where each such static property is somehow “contrary” to the next one (Lombard 1998, 289). This may give the impression that there are many constitutive properties of the event e, namely S1, S2, S3, etc., as well as D. But in fact there is at most one constitutive property among these, namely D. For S1, S2, S3, etc. are not constitutive properties of the event, but rather, one may say, of various states which somehow are involved in the event. Be this as it may, in the end the MacDonalds drop the option in question (see note 39, p. 561 in their paper).

92 for some physical property P, a property (say, having C-fibres firing) that realizes being in pain. But, in view of Monist Identity Condition, e = e΄. In sum, we have Token Identity. There are then three kinds of NENRP that are worth comparing for the purposes of this paper: the tropist brand, the Heil universalist brand, and the monist brand. It will emerge that, from the point of view of philosophy of mind, there are significant differences among them, differences ensuing from their being rooted in different doctrines of basic ontology, in the way explained above. 11.

Unitarianism vs. anti-unitarianism

According to unitarianism, Tom and fellow creatures such as Marsie and Roby can feel exactly the same way. For instance, Tom can experience precisely what it feels for Marsie and Roby to be in pain. In other words, the commonsensical expectation according to which there is just one pain (and similarly for other mental properties) is, in the unitarian’s opinion, fulfilled. The standard universalist can be unitarian in the following sense. Tom, Roby and Marsie can feel pain in the same way in that they can instantiate one and the same property of having pain, precisely in the sense in which they can all have the same height, weight or colour. The monist universalist who supports NENRP can uphold unitarianism, by virtue of being a standard universalist. According to anti-unitarianism, Tom and his fellows can only approximately feel the same way: the above mentioned commonsensical expectation is not fulfilled, just as suggested by the doctrine of relative reductive physicalism, considered in section 4. As noted, according to it, there is no single pain, but at best different kinds of pain: pain-for-humans, pain-for-Martians, pain-for-robots, etc. From the tropist’s point of view, anti-unitarianism can be cashed out as, e.g., the fact that Tom, Marsie and Roby can at best exemplify pain tropes which only partially resemble each other (although they can exemplify perfectly resembling height, weight and colour tropes).46 Similarly, from the Heil universalist’s standpoint at best 46

A.-S. Maurin has suggested in correspondence a different way in which a tropist may render the anti-unitarian position: “Could not the tropist say that each of Tom, Marsie and Roby exemplify a trope, and that these tropes exactly resemble each other ‘mentally’ (e.g. functionally), although they only approximately (if at all) resemble each other ‘physically’ (e.g. neurophysiologically)?” It seems to me that, if we follow

93 they can exemplify similar but distinct universals (although they can exemplify the very same height, weight and colour universals). From the tropist’s point of view, the commonsensical expectation would be, e.g., that the predicate “having pain” corresponds to a class of perfectly resembling tropes, and the fact of the matter that the class corresponding to the predicate is instead only one of partial resemblance. On the other hand, from the point of view of the Heil universalist, the expectation would be that of a single universal (or one-membered class of universals) corresponding to the predicate and the reality that of a multimembered family of mutually resembling universals. Thus, the tropist and the Heil universalist, insofar as they support NENRP, in a sense can be seen as pursuing, in their different but symmetric ways, versions of relative reductive physicalism. 47 They are both reductivists and non-reductivists, though not in an incoherent way. Let us focus on the tropist to clarify this. She is reductivist to the extent that she admits, in line with relative reductive physicalism, that mental properties2 such as pain-for-humans and pain-for-Martians (classes of perfectly similar tropes) are identical to physical properties2 such as having C-fibres firing and having M-fibres firing. As noted, in the light of the phenomenon of brain plasticity, the relative reductive physicalist may want to resort to mental properties such as pain-for-Lewis-at-time-t1 and painfor-Putnam-at-time-t2. Similarly, the tropist can say that the mental properties2 that can be considered classes of perfect resemblance are more numerous, but smaller in size, than we might have thought. For example, there would be no single (large) perfect resemblance class corresponding to the predicate “pain-for-humans”, but many smaller perfect resemblance classes, e.g., one comprising a pain-for-Lewis-at-t1 trope and another comprising a pain-for-Putnam-at-t2 trope. However, the tropist is nonreductivist in claiming that there is a (very large) mental property2 such as pain (a class of partially resembling tropes), which is not identical to this road, we have to complicate tropism, by taking resemblance to be a triadic relation involving two tropes and something like a “respect” in which they may be similar. Now, these “respects” may turn out to be very close to the universals that the tropist wants to avoid and, at any rate, it is not obvious that the tropist has any theoretical advantage in appealing to them. 47 This connection between relative reductive physicalism and the tropist version of NENRP is noted by Robb and Heil, as they present the latter as a way to salvage what I have called the commonsensical expectation (see theirs 2003, section 6.3, 14). In view of the above considerations however the latter position cannot be said to (fully) meet this expectation and the same goes for the Heil universalist version of NENRP.

94 (smaller) physical properties2 such as having C-fibres firing (or, given brain plasticity, the one comprising the pain-for-Lewis-at-t1, but not the pain-for-Putnam-at-t2 trope). Something similar, mutatis mutandis, can be said from the perspective of Heil universalism. In sum, both the tropist and the Heil universalist, insofar as they support NENRP, are bound to be anti-unitarian and also lean toward relative reductive physicalism. In contrast, the monist universalist can support a unitarian version of NENRP, which need not have anything to do with relative reductive physicalism. 12.

Self-acquaintance

Another important difference emerges if we reflect on the intuitive idea that we are somehow acquainted with our own mental states in a way in which we are not with respect to external objects (at least from the perspective of a representationalist account of perception). For example, if Tom and Mary are both in pain at the same time, Tom is acquainted with his pain and Mary with her pain, but not vice versa. One could think that this simply means, from a universalist perspective, that Tom is acquainted with a universal, say, having paint, with which Mary is not (cannot be). On the other hand, however, she is acquainted with a different one, say, having painm, precluded to Tom. But this may at best be in tune with Heil universalism. From the perspective of standard universalism, it seems more appropriate to say that there is one universal, being in pain, that occurs as constituent in two distinct events. Tom is acquainted with one of them, his being in pain (at the time in question), and Mary is acquainted with the other one, her being in pain (at the time in question). Similarly, from a tropist perspective, one could say that Tom is acquainted with a certain pain event, i.e., a trope exemplified1 by Tom, and Mary with a different pain event, i.e., a trope exemplified1 by Mary, although the two events (tropes) may well resemble each other perfectly. Yet, there is a difference which might be taken to have far-reaching consequences in philosophy of mind. Events are, for the (default) tropist, basic particulars and thus admitting an acquaintance with one of them does not bring with it any temptation to say that this acquaintance involves in turn an acquaintance with the constituents of the trope. The trope, qua basic, has no constituents. In contrast, for the standard universalist, events

95 are structured in that they involve constituents, namely a particular, a property and a time. In one of the events under scrutiny now the particular is Tom himself and in the other Mary herself. Recall that monist universalism is a form of standard universalism. By virtue of this, in the light of what we have just seen, it is a point of view that invites the conclusion that each of us can be acquainted not only with the mental universals that one exemplifies, but also with a certain particular, oneself (and, at a given time, with the time in question).48 In contrast, since Heil universalism and tropism are different from standard universalism, they do not invite the same conclusion. It is interesting to note in this connection that well-known supporters of universals, such as Roderick Chisholm (1969) and Bertrand Russell (at some point of his career) have acknowledged this sort of self-acquaintance. For example here is what Russell (1912, ch. 5) says: We have acquaintance in sensation with the data of the outer senses, and in introspection with the data of what may be called the inner sense—thoughts, feelings, desires, etc.; we have acquaintance in memory with things that have been data either of the outer senses or of the inner sense. Further, it is probable, though not certain, that we have acquaintance with Self, as that which is aware of things or has desires toward things.

13.

Some doubts

Perhaps Monist Identity Condition is criticizable. A critic could say that it is the redness of the cloth that infuriates the bull (or, as we are told, the movement), not the cloth’s having that particular shade of red (movement). Or she might say that it is the being jade of the stone that elicits a feeling of aesthetic pleasure in a certain person, and not its being nephrite or its being jadeite. It seems, in other words, that we have two states of affairs and not just one, in situations in which, by Monist Identity Condition, we should say that there is just one event. Moreover, the critic could argue that when we distinguish between a necessitating property F and a necessitated property G, and we also admit that both of them are instantiated at t by a single individual, x, then we admit ipso facto that there are two items, the instantiantion of F by x at t and the instantiantion of G by x at t. If we say 48

For more details on this option and for its repercussions in philosophy of language, see Orilia 2007.

96 that there is one single event, the critic could then advance the suspicion that we are complicating the ontology by introducing a third item, the event, in addition to the two exemplifications, rather than simplifying it, by making the two exemplifications identical to one and the same item, the event. If this is so, we might have an ad hoc violation of Ockham’s razor, simply motivated by a desire to secure Efficacy. If Monist Identity Condition becomes unappealing in the light of such (or other) criticisms, the standard universalist might have to eschew monist universalism. If so, perhaps, in order to save Efficacy she could end up reconsidering, instead of Supervenience, some sort of mental causation incompatible with physicalism. In other words, NENRP, as supported by monist universalism, could have an Achilles’ heel which could lead altogether to its abandonment in favour of the old dilemma between some kind of non-physicalist position and epiphenomenalism (on the assumption that reductive physicalism is not viable). The tropist and the Heil universalist forms of NENRP, despite other weaknesses that they may have, do not have this one. 14.

Conclusion

As suggested by Heil 2008, there is indeed a version of universalism, Heil universalism, as I have called it, which is compatible with nonepiphenomenalist non-reductivist physicalism (NENRP) pretty much in the way tropism is. But this fact can hardly be used to back up the claim that the issue of tropism vs. universalism is irrelevant for philosophy of mind. For Heil universalism is not the most natural form of universalism (or at least is not the only form of this doctrine) and the universalist can legitimately prefer what I have called monist universalism and try to use it in an effort to support NENRP. However, if some doubts regarding the identity condition for events are left aside, this results in a version of NENRP quite different from the versions based on tropism and Heil universalism. For only the former is compatible with the unitarian position according to which physically different creatures can feel or have experiences in exactly the same way and only the former invites, in a particularly straightforward sense, the idea that there can be an acquaintance with the self of the kind adumbrated by philosophers such as Russell or Chisholm. The results presented here thus contribute, I hope, to the general thesis that basic ontology matters in philosophy of mind.

97 REFERENCES Armstrong, D. M. (1978). Nominalism and Realism. Universals and Scientific Realism, Volume I. Cambridge: Cambridge University Press. Armstrong, D. M. (1978a). A Theory of Universals. Universals and Scientific Realism, Volume II. Cambridge: Cambridge University Press. Armstrong, D. M. (1989). Universals, An Opinionated Introduction, London: Westview. Armstrong, D. M. (1997). A World of States of Affairs. Cambridge: Cambridge University Press. Bergmann, G. (1967). Realism. A Critique of Brentano and Meinong. Madison: University of Wisconsin Press. Braun, D. (1995). Causally Perspectives, 9, 447-475.

Relevant

Properties.

Philosophical

Campbell, K. (1981). The Metaphysics of Abstract Particulars. Midwest Studies in Philosophy, 6, 477-488. Campbell, K. (1990). Abstract Particulars. Oxford: Blackwell Publishers. Chisholm, R. (1969). On the Observability of the Self. Philosophy and Phenomenological Research, 30, 7-21. Reprinted in Self-Knowledge, ed. Q. Cassam, Oxford: Oxford University Press, 1994, 94-108. Davidson, D. (1980). Essays on Actions and Events. Oxford: Oxford University Press. Davidson D. (1985). Reply to Quine on Events. Lepore and McLaughlin 1985, 172-176. Dretske, F. (1989). Reasons and Causes. Philosophical Perspectives, 3, 4776. Ehring, D. (1996). Mental Causation, Determinables and Property Instances. Noûs, 30, 461-480. Endicott, R. (2005). Multiple Realizability. The Encyclopedia of Philosophy ed. D. Borchert, Thomson Gale, Macmillan Reference, 2nd Edition, 427-432. Feldman, F. (1980). Identity, Necessity and Events, Readings in the Philosophy of Psychology, Volume I, ed. N. Block. Cambridge: Harvard University Press, 148-158.

98 Funkhouser, Eric (2006). The Determinable-Determinate Relation. Noûs, 40, 548-569. Gill, K. (1993). On the Metaphysical Distinction between Properties and Events. Canadian Journal of Philosophy, 23, 365-385. Reprinted in Thought, Language and Ontology. Essays in Memory of Hector-Neri Castañeda, eds. F. Orilia and W. J. Rapaport, Dordrecht: Kluwer, 135151. Gozzano, S. (2008). Tropes’ Simplicity and Mental Causation. This volume. Heil, J. (1992). The Nature of True Minds. Cambridge: Cambridge University Press. Heil J. (2008). Modes and Mind. This volume. Horgan, T. (2001). Multiple Reference, Multiple Realization and the Reduction of Mind. Reality and Humean Supervenience: Essays on the Philosophy of David Lewis, eds. G. Preyer and F. Siebelt, Landham, MD: Rowman and Littlefield, 205-221. Kim, J. (1972). Phenomenal Properties, Psychophysical Laws, and the Identity Theory. The Monist, 56, 177-192. Kim, J. (1973). Causation, Nomic Subsumption and the Concept of Event. The Journal of Philosophy, 70, 217-236. Reprinted in Kim 1993, 3-32. Kim, J. (1984). Concepts of Supervenience. Philosophy and Phenomenological Research, 45, 153-176. Reprinted in Kim 1993, 5378. Kim, J. (1993). Supervenience and Mind. Cambridge: Cambridge Universisty Press. Kim, J. (1998). Philosophy of Mind. Boulder: Westview. Kim, J. (2005). Physicalism or Something Near Enough. Princeton: Princeton University Press. Lepore, E. and McLaughlin, B., eds. (1985). Action and Events. Perspectives on the Philosophy of Donald Davidson. Oxford: Blackwell. Lewis, D. (1969). Review of Art, Mind and Religion. The Journal of Philosophy, 66, 23-35. Lombard, L. B. (1986). Events. A Metaphysical study. London: Routledge and Kegan Paul.

99 Lombard, L. B. (1998). Ontologies of events. Contemporary Readings in the Foundations of Metaphysics, eds. S. Laurence and C. MacDonald, Oxford: Blackwell, 277-294. Lowe, E. J. (2008). Tropes and Perception. This volume. Macdonald, C. and Macdonald G. (2006). The Metaphysics of Mental Causation. The Journal of Philosophy, 103, 539-576. Malcolm, N. (1968). The conceivability of Mechanism. Philosophical Review, 77, 45-72. Marras, A. (1993). Psychophysical Supervenience and Nonreductive Materialism. Synthese, 95, 275-304. Marras, A. (2005). Consciousness and Reduction. British Journal for the Philosophy of Science, 56, 335-361. Marras, A. and Yli-Vakkuri, J. (2008). The “Supervenience Argument”: Kim’s Challenge to Nonreductive Physicalism. This volume. Maurin, A.-S. (2002). If Tropes. Dordrecht: Kluwer. Maurin, A.-S. (2008). Does Ontology Matter?. This volume. Mourelatos A. (1978). Events, Processes and States. Linguistics and Philosophy, 2, 415-434. Mulligan, K., Simons, P., and Smith, B. (1984). Truth-Makers. Philosophy and Phenomenological Research, 44, 287-320. Nordhof, P. (1998). Do Tropes resolve the Problem of Mental Causation?. The Philosophical Quarterly, 48, 221-226. Orilia, F. (2006). Quantum-mechanical Statistics and the Inclusivist Approach to the Nature of Particulars. Synthese, 148, 57-77. Orilia, F. (2006a). Universals, Tropes and Philosophy of Mind. Presented at Cervelli, persone e Società, VII Congresso Nazionale, Società Italiana di Filosofia Analitica, 28-30 September 2006, Università Vita-Salute S. Raffaele, Cesano Maderno (Milan). Orilia, F. (2007). Self-reference and Self-knowledge. The Soochow Journal of Philosophical Studies, 16, Special Issue, International Conference on Analytic Philosophy, 257-281. Quine, W.V.O. (1985). Events and Reification. Lepore and McLaughlin 1985, 162-171. Robb, D. (1997). The Properties of Mental Causation. The Philosophical

100 Quarterly 47, 178-194. Robb, D. (2001). Reply to Noordhof on Mental Causation. The Philosophical Quarterly, 51, 90-94. Robb, D. (2008). Zombies from Below. This volume. Robb, D. and Heil, J. (2003). Mental Causation. Stanford Encyclopaedia of Philosophy (Winter 2003 edition), ed. E. N. Zalta, URL: . Russell, B. (1912). The Problems of Philosophy. London: Home University Library. Williams, D. C. (1953). On the Elements of Being I. The Review of Metaphysics, 7, 3-18. Yablo, S. (1992). Mental Causation. Philosophical Review, 101, 245-280. Reprinted in Philosophy of Mind. Classical and Contemporary Readings, ed. D. Chalmers, Oxford: Oxford University Press, 179-196 (page references to the latter).

The “Supervenience Argument”: Kim’s Challenge to Nonreductive Physicalism AUSONIO MARRAS University of Western Ontario [email protected] JUHANI YLI-VAKKURI McGill University [email protected] ABSTRACT. Jaegwon Kim’s “supervenience argument” purports to show that epiphenomenalism about the mental follows from premises that any nonreductive physicalist should find acceptable. Kim regards his argument as a reductio ad absurdum of nonreductive physicalism. We reconstruct and evaluate the latest version of Kim’s argument. We argue that the premises of Kim’s argument are much less innocent than they may appear. In particular, we single out for criticism an unstated assumption about the identity conditions of events, and we argue that this assumption could be seen as all by itself implying that nonreductive physicalism is false, thus begging the question against that position. It is also dubious, we argue, whether Kim’s unstated assumption is even consistent with one of the stated assumptions of his argument, “the principle of causal exclusion”, given a standard understanding of causal overdetermination. We conclude with some polemical remarks about the conception of causation presupposed by Kim’s argument—a conception that appears to depart from that at work in science and commonsense discourse.

1.

Introduction1

Many philosophers have worried that physical causation may “exclude” mental causation—physical and mental causes “compete” for efficacy, and because of some principle (the causal closure of the physical domain, for example), physical causes inevitably “win”. This picturesque language is common, but explicit arguments, which identify prima facie plausible principles from which epiphenomenalism about the mental would logically follow, are less so. Jaegwon Kim’s “supervenience argument” represents perhaps the most influential attempt to construct such an argument. In Kim’s case, the argument is presented as a reductio ad absurdum of 1

We would like to thank Ian Gold for helpful comments on an earlier draft of this paper.

102 nonreductive physicalism. The present paper is an attempt to get clear on just what Kim’s supervenience argument is, and how, or whether, it works. We will only discuss Kim’s most recent formulation of the supervenience argument, or arguments—Kim gives us two “versions” of the argument (“Completion 1” and “Completion 2”). This formulation is found in ch. 2 of Kim’s most recent book, Physicalism, or Something Near Enough (Kim 2005; henceforth “PSNE”), in which, according to the chapter title, “The Supervenience Argument [is] Motivated, Clarified, and Defended”. We will attempt to give a reconstruction of the arguments that renders them deductively valid, so that every assumption on which we believe Kim relies is made explicit. We will show that the premises of Kim’s arguments are not nearly as innocent as they seem, and that one assumption in particular, which concerns the identity conditions of events, can only be viewed as begging the question against nonreductive physicalism, as that position has traditionally been conceived. Whatever else they may be, Kim’s arguments then are not reductios of nonreductive physicalism. Nor do they succeed in posing a problem about the possibility of mental causation by showing that we must either accept epiphenomenalism or reject one or another of a set of prima facie plausible metaphysical principles. Kim takes himself to have shown that the claims he calls his “substantive premises”, which are indeed prima facie plausible—at least relative to prevailing assumptions about causation, on which we will comment in the final section of this paper—are not consistent with the claim that mental events have causal efficacy. But what he has in fact shown is the inconsistency of a larger set of claims, some of which have no particular prima facie plausibility, even relative to the prevailing assumptions. We think the aforementioned assumption about the identity conditions of events is the least plausible among these. 2.

The “substantive premises”

Kim thinks his arguments show that the following claims cannot all be true: Irreducibility No mental property is identical with a physical property.

103 Supervenience All properties strongly supervene2 on physical properties.3 In other words: necessarily, for all properties P, all objects x, and all times t: if x has P at t, then, for some physical property P′, x has P′ at t and, necessarily, for all y and all times t′ , if y has P′ at t′, then y has M at t′. Closure If a physical event has a cause that occurs at a time t, then it has a physical cause that occurs at t.4 Causal Efficacy Mental events sometimes cause other events. These assumptions, and Kim’s arguments, concern two different kinds of entities: properties and events. Like Kim, we assume that events are concrete instances5 of properties—not “instances” in the sense of objects that have the properties, but instances in something like the sense of havings of properties by particular objects at particular times.6 We take the following schema (at least when suitably restricted) to be a platitude: the

2

We will abbreviate “strongly supervenes” to “supervenes” in the discussion to follow, as we are not discussing any other varieties of supervenience. 3 In PSNE the principle only says that all mental properties strongly supervene on “physical/biological” properties, but Kim, and we, are interested in formulating some minimal physicalist commitments, and a physicalist had better think that all properties strongly supervene on physical properties. 4 Curiously, the principle Kim calls “Closure” is not a closure principle. To say that the physical domain is causally closed, in the usual technical sense, would of course be to say that every cause of a physical event is also a physical event—not an implausible principle, to our mind. Kim, however, thinks that to assume the physical domain to be causally closed in the literal sense would be to beg the question that is at issue in his argument: it would be “like starting your argument with mind-body causation already ruled out, at least for nonreductivists” (PSNE, 51). We think this is incorrect, for reasons that will become evident in sections 5 and 6, but for now we will join Kim in assuming only the weaker principle that every physical event with a cause has a synchronous physical cause, and calling it “Closure”. 5 Kim prefers the term “exemplification”. See Kim (1976). 6

We shall not consider the possibility that events might be construed as “tropes”. For discussion see papers by Francesco Orilia and others in this volume.

104 event that is the having of property P by object x at time t exists if and only if x has P at t.7 According to Kim, Irreducibility, Supervenience, and Closure are shared commitments of all nonreductive physicalists. Thus Kim is posing the dilemma: either reject nonreductive physicalism or reject Causal Efficacy, viz., accept epiphenomenalism. But things aren’t quite so simple, as there is a further premise in the argument which is not a characteristically nonreductive physicalist assumption: Exclusion No single event can have more than one sufficient cause occurring at any given time, unless it is a genuine case of overdetermination.8 Kim says this is a “general metaphysical constraint” (p. 22), so presumably he thinks we should accept it whether we are nonreductive physicalists or not. Whether Exclusion is true or not is not perfectly obvious, and we’ll return to this matter in section 7. For now let us simply note that the appeal of the principle is perhaps due to the fact that it sounds tautologous: it sounds a lot like the claim that every event has at most one sufficient cause occurring at a given time unless it has more than one sufficient cause occurring at that time. 3.

Other general metaphysical constraints

Though the assumptions Kim names and calls his “substantive premises” (p. 41) end here, his argument requires several other assumptions that, like Exclusion, are not clearly nonreductive physicalist commitments. We assume that he regards them, too, as “general metaphysical constraints”. One such assumption is what we will call:

7

If P is an n-place relation, then x is an ordered n-tuple of objects. A restriction to the schema in terms of the notion of metaphysical contingency is contemplated in note 27. We find this restriction acceptable but not necessary. Our schema is similar to what Kim (1976, 35) calls the “existence condition”, but not quite the same: see note 27. 8 This is Kim’s formulation exactly (PSNE, 42). Like Kim, we omit the word “sufficient” in the discussion to follow, but when we use the word “cause”, this should be understood as having an implicit “sufficient” in front of it.

105 No Overdetermination For all properties F and G, if F is a supervenience base of G, then no event is causally overdetermined by (events that are instances of) F and G. Like Exclusion, this principle has a tautologous sound to it. When we say that an event is causally overdetermined by two other events, we mean that it has two causes that are independently sufficient for its occurrence. “Independent” here means, surely, at least that it would have been possible for each to occur without the other (and that if one had occurred without the other, it would have brought about the same effect). But it is a straight logical consequence of the definition of supervenience that instances of properties and their supervenience bases are not “independent” in this sense, as the instantiation of the base necessitates the instantiation of the supervening property. Our evidence that Kim makes use of No Overdetermination is in the section titled “Why overdetermination is not an option”, in which Kim answers the question, Why do instances of mental properties and their supervenience bases not overdetermine their effects? His answer (p. 48): The usual notion of overdetermination involves two or more separate and independent causal chains intersecting at a common effect. Because of Supervenience, however, this is not the kind of situation we have here. In this sense, this is not a genuine case of overdetermination ...

Since Kim thinks it is “because of supervenience” in the case under consideration, we assume he thinks that supervenience always precludes overdetermination, and that is what No Overdetermination says. In addition, Kim must also assume some other principles relating supervenience to causation. We believe that he is assuming these: Supervenience-Causation (SC) I If c causes e and e′ is a supervenience base of e, then c causes e′. Supervenience-Causation (SC) II If c causes e and c′ is a supervenience base of c, then c′ causes e. These principles are perhaps more clearly expressed by means of a visual aid: see Figures 1 and 2. (We represent causation with a single arrow and supervenience with a double arrow pointing from the supervenience base

106 to the supervening event.) SC I and SC II respectively tell us that when the top single arrow in each of Figure 1 and 2 occurs, so does the bottom single arrow. Figure 1: Supervenience-Causation I e c e′ Figure 2: Supervenience-Causation II c

e c′ One further remark about what SC I and SC II mean is in order. The principles speak of supervenience bases of events, though the usual notion of supervenience is defined for properties. What we mean by “e′ is a supervenience base of e”, when e and e′ are events, is that e and e′ are simultaneous events involving the same object(s) and e′ is an instance of some property which is a supervenience base of a property that is instantiated by e. More precisely: The event in which x instantiates P at t is a supervenience base of the event in which y instantiates Q at t’ if and only if x = y, t = t’, and P is a supervenience base of Q. We believe that Kim introduces SC I at pp. 39-41, where he expresses the principle by saying that an instance of a property can only cause an instance of a supervenient property “by causing its supervenience

107 base” (p. 40).9 The grounds for this claim have to do with the alleged fact that there are “two seemingly exclusionary answers” to the question “What is responsible for, and explains, the fact that [e] occurs on this occasion?” These two answers are: (a) Because c caused e. (b) Because e′, a supervenience base of e, is instantiated on this occasion. Kim says that a “tension [is] created by (a) and (b)” and that a “simple and natural way of dissipating” this tension is to conclude that c is a cause of e. It is not clear to us why these answers are “seemingly exclusionary”, or even that they are seemingly exclusionary (they don’t strike us as seemingly exclusionary),10 but we will leave it to others to reconstruct and evaluate the reasoning that leads Kim to accept SC I, because an adequate reconstruction of it would require a separate paper11 and we are not, at any rate, singling this assumption out for criticism. The reasoning supporting SC II is more transparent. SC II is introduced, obliquely, at p. 41, where Kim states that “There are strong reasons for thinking that [in the kind of situation represented in Figure 1] [c'] is a cause of [e']”. These reasons are:

9

This does not have exactly the same meaning as our SC I, but we assume that whatever additional import Kim’s “by” has is irrelevant to the arguments; the assumption that the cause of an event e is also the cause of any event subvening e suffices for the purpose for which Kim needs the principle he expresses using the word “by”—namely, deriving step (3) in the two versions of the argument below. Since Kim’s principle differs from our SC I only in being stronger than it (“x does A by Bing” entails “x Bs”), and the weaker SC I can fill the same role in the argument, we think we can safely ignore Kim’s “by”. 10 See Marras 2007, section 3, for a discussion of related issues. 11 At the bottom of p. 39 and the top of p. 40 Kim gives a highly impressionistic argument involving counterfactuals and modal operators for the conclusion that c’s occurrence must have had “something to do with” e'. The argument is open to several different interpretations, which we cannot give here for lack of space. Here the conclusion we are invited to draw is that this “something” is the causal relation. The final step in the argument, then, appears to be a kind of inference to the best explanation.

108 [c'] is (at least) nomologically sufficient for [c], and the occurrence of [c] on this occasion depends on, and is determined by, the presence of [c'] on this occasion. Since ex hypothesi [c] is a cause of [e], [c'] would appear to amply qualify as a cause of [e] as well.

The idea appears to be this: c supervenes on c'; therefore c' is at least nomologically sufficient for c (this depends on reading the innermost modal operator in the definition of supervenience as “at least” nomological necessity). By assumption, c is a cause of e. Any event that is nomologically sufficient for the occurrence of a cause of an event is also a cause of it; therefore c' is a cause of e'. The final “general metaphysical constraint” required by the argument, according to our reconstruction, is: Closure-Overdetermination If a physical event p has a cause c that occurs at t, then p has a physical cause p' that occurs at t such that p is not overdetermined by c and p'. Kim does not express Closure-Overdetermination anywhere in the text, but we attribute it to him because 1) it very neatly fills a gap in one of his arguments, and 2) it is highly plausible given Closure, which he does accept. For suppose that a physical event p has a cause c that is not physical. Then by Closure p has a physical cause p' synchronous with c. Is p overdetermined by c and p' ? If it is, then every event that has a nonphysical cause is overdetermined, and this seems absurd. It is of course logically possible that the physical causes posited by Closure sometimes but not always overdetermine their effects jointly with their concurrent nonphysical causes, but this seems bizarre and arbitrary. The most natural conclusion is that Closure-Overdetermination is true.12

12

Compare Papineau (1993, 22-23), who also endorses this principle, effectively giving it the same justification. Papineau correctly observes that ClosureOverdetermination together with the claim that every mental event causes some physical event implies that we must either accept the token (not type) identity of mental and physical events or reject Exclusion (he does not use these labels, but this is what his argument amounts to).

109 4.

The arguments

Before giving us his arguments, Kim announces (p. 39): Properties as such don’t enter into causal relations; when we say “M causes M*”, that is short for “An instance of M causes an instance of M*” or “An instantiation of M causes M* to instantiate on that occasion.” Also for brevity we suppress reference to times.

We will follow Kim in adopting these abbreviating conventions whereby “an instance of property M” and the like will be replaced by simply “M” and the like, as if properties were identical to their instances. In reading the argument to follow, it will be useful to pretend that the world is such that there are just two times, call them “t” and “∆t”, and that anything that is a cause of anything else occurs at t, and anything caused occurs at ∆t— references to times thus becomes unnecessary. Kim offers two versions of his argument, and we will discuss them in order. Our exposition differs from Kim’s in two respects: First, we present his “versions” of the argument as two distinct arguments, rather than as a “Stage 1” followed by two alternative “completions”, as he does. Secondly, we will eliminate certain unnecessary steps, which we will comment on after we give each reconstruction. We will use sorted variables, so that “P”s, supplemented with primes as needed, are variables for physical properties and “M”s, likewise supplemented, are variables for mental ones. 4.1. Kim’s first argument In the first argument we introduce for a reductio the assumption that a case of mental-to-mental causation occurs: (1) M causes M'. By Supervenience, we conclude that: (2) There is a physical property P' such that P' is a supervenience base of M'.

110 By SC I we conclude from (1) and (2) that: (3) M causes P'.13 Again, by Supervenience: (4) There is a physical property P such that P is a supervenience base of M. By SC II, it follows from (3) and (4) that: (5) P causes P'. But Irreducibility tells us that: (6) M ≠ P. From No Overdetermination and (4) we get: (7) P' is not causally overdetermined by M and P. Applying Exclusion to (5), (6), and (7), we conclude that: (8) M does not cause P'. (8) contradicts (3), and we have a refutation of (1). In other words, we have derived (9) from our assumptions. (9) It is not the case that M causes M'. Since M and M' were arbitrary mental properties, we must conclude the universal closure of (9):

13

This is not exactly what Kim’s step (3) says: it says “M causes M* by causing its supervenience base P*” (p. 40, Kim’s italics), but see our note 9.

111 (10) For all mental properties M and M', it is not the case that M causes M'. In other words, there is no mental-to-mental causation. To show that there is no mental-to-physical causation, we need a separate argument, which can be obtained from the argument just given by deleting lines (1) and (2) and taking (3) (“M causes P' ” ) to be the reductio hypothesis. These arguments together commit the nonreductive physicalist to a pervasive epiphenomenalism: mental properties do not cause any properties, mental or physical, to be instantiated.14 The argument just given fits into a larger reductio: one against the “substantive premises” of section 1 and the “general metaphysical constraints” of section 2. The conclusion (10) directly contradicts Causal Efficacy. We seem to have a demonstration that the “substantive premises” and the “general metaphysical constraints” cannot all be true. Supposing the metaphysical constraints to be off the table, we seem to be forced to reject either Causal Efficacy or one of the other “substantive premises”, which characterized nonreductive physicalism—in other words we must either accept epiphenomenalism or reject nonreductive physicalism. The above reconstruction differs from Kim’s presentation of the argument in one important respect: in Kim’s presentation, Exclusion tells us that “we must eliminate either M or P as [P']’s cause”, whereafter Closure is called upon to tell us that it must be M that gets “eliminated” (pp. 42-43). But as can be seen above, Exclusion together with the preceding three lines yields the conclusion that M is not a cause of P, contradicting (3) [in Kim’s numbering, (8) (p. 41) contradicts (5) (p. 43)]. Citing Closure after a contradiction has been derived cannot contribute anything to the argument. Closure, then, is not actually needed as a premise in the first version of Kim’s argument.15 14

Or so Kim seems to think. If you think, as we do, that there are properties that are neither mental nor physical (say, biological properties), then you won’t agree that a pervasive epiphenomenalism is established yet. However, the argument goes through no matter what property—biological, chemical, aesthetic, whatever—is considered in place of M′. 15 It might be suggested that, instead of SC II and No Overdetermination, Kim is tacitly appealing to Closure-Overdetermination, as we believe he is in his second argument (see below), to derive steps (5) and (7). However, there is no textual evidence for this. At pp. 41-42, where these steps are derived, Kim makes no mention of a causal closure principle (except parenthetically at the top of p. 42, to announce that Closure will later be applied to “disqualify M as a cause of [P′]”—the step that

112 4.2. Kim’s second argument Let us now turn to the second version of Kim’s argument. What follows is more paraphrase than reconstruction. Our lines (1) and (3)-(6) come from Kim more or less verbatim except for a change in the numbering, and as noted in note 7 above; the remaining lines are close paraphrases of lines in Kim’s argument. Lines (1)-(3) are what Kim calls “Stage 1”, and lines (4)(7) he calls “Completion 2”. These taken together comprise Kim’s second argument. As before, we make for a reductio the assumption that there is a case of mental-to-mental causation: (1) M causes M'. Again we know by Supervenience that: (2) There is a P' such that P' is a supervenience base of M'. From (1) and (2) it follows by SC I that: (1) M causes P'. By Closure, (3) implies that: (4) P' has a physical cause—call it P—occurring at the time M occurs. But Irreducibility tells us that: (5) P ≠ M. Next Kim asserts a line without justification [he introduces it only with “hence” (p. 44)]: according to us is unnecessary). What Kim does cite here is the fact that P is “at least nomologically sufficient” for M (quoted above), and it is clear that he thinks this is so because M supervenes on P. This, and Kim’s later remarks about supervenience precluding overdetermination (p. 48), is to us decisive evidence that SC II and No Overdetermination are the operative assumptions here.

113 (6) P' is not overdetermined by P and M. Now, from (4), (5), (6) and Exclusion it follows that: (7) M does not cause P'. (7) contradicts (3), and once again we must conclude that M did not cause M'. As M and M' were arbitrary mental properties, the conclusion must be that no mental-to-mental causation occurs ever, and the conclusion that no mental-to-physical causation occurs either follows as in the second variant of the first argument. In addition to the large problem shared by both versions of Kim’s argument discussed in the next two sections, there are two minor problems specific to the second version, as presented by Kim. The first problem is that not only is line (6) not justified, but it can’t be justified from assumptions explicitly made by Kim, his “substantive premises”. This is why we attributed Closure-Overdetermination to Kim. It would enable him to complete the argument as follows.16 M (4) P′ has a physical cause—call it P—occurring at the time M occurs— and P′ is not overdetermined by P and M. [From (3) by ClosureOverdetermination.] 16

There are, to be sure, other ways to complete the argument on Kim’s behalf, which are not implausible. The alternatives that come to mind, however, would have Kim appealing to the supervenience of M on P, whereas he insists that his second “completion is simpler than Completion 1” in part because “Supervenience is not needed as a premise” here (p. 44). The main alternative we can think of for ClosureOverdetermination is:

Closure-Supervenience If a physical event p has a cause c occurring at a time t, then p has a physical cause p’ such that 1) p′ occurs at t and 2) p′ is a supervenience base of c. This principle could be given a justification similar to the one we gave ClosureOverdetermination: it could be argued that if the principle is not true, absurdly pervasive overdetermination will follow (unless mental events happen to be identical to physical events—a possibility we will consider later). For lack of space, we leave working out the exact justification for Closure-Supervenience, as well as how Kim’s second argument could be completed using it, as an exercise for the reader.

114

(5) P ≠ M. [From Irreducibility.] (7) M does not cause P′. [From (4) and (5) by Exclusion.] The second problem is not so easy to solve on Kim’s behalf. It is that Kim says that in his second argument “Supervenience is not needed as a premise” (p. 44), but this is not so: the second argument shares what Kim calls “Stage 1” with the first argument, and Supervenience is needed in Stage 1. Without Supervenience, we cannot conclude that M′ has a physical supervenience base P′. According to our reconstruction, the difference between the two arguments is different from what Kim asserts: they differ with respect to two premises: the first argument makes use of SC II and No Overdetermination whereas the second does not; the second argument makes use of Closure-Overdetermination whereas the first one does not. It is, of course, entirely possible that our reconstruction does not capture Kim’s intentions (see note 15), but our goal has been to produce a reconstruction that is as close to the letter of Kim’s exposition as possible. 5.

The main logical problem

Here is the most serious problem with Kim’s arguments. That the arguments appear to have a valid form is only an artifact of the abbreviating convention used by Kim—unabbreviated, the arguments are either invalid or enthymematic; we suppose the latter. We followed Kim in using the same variables for both properties and events, but now we will adopt a more explicit nomenclature, with uppercase letters (“M” and “P”, possibly supplemented with primes) for properties and lowercase letters for their instances (“m” and “p”, possibly supplemented with primes). Using this convention and adding material presumably elided by Kim, the steps of our first reconstruction are as follows, where the added material is in boldface. (1) An instance m of M causes an instance m′ of M′ [Assumption]. (2) There is an instance p′ of a physical property P′ such that p′ is a supervenience base of m′. [From (1) by Supervenience.]

115 (3) m causes p′. [From (1) and (2) by SC I.] (4) There is an instance p of a physical property P such that p is a supervenience base of m. [From (1) by Supervenience.] (5) p causes p′. [From (3) and (4) by SC II.] (6) M ≠ P. [By Irreducibility.] (7) p′ is not causally overdetermined by m and p. [From (4) by No Overdetermination.] (8) m does not cause p′. [From (5), (6), and (7) by Exclusion.] It is clear that (8) cannot be derived from (5), (6), and (7) by Exclusion: for (8) to be so derivable, (6) would have to say “m ≠ p′ ”, not “M ≠ P′ ”. Nor does (8) follow from previous steps by any of the other assumptions that we have made explicit so far. However, one interesting conclusion does follow, namely: (8′) m = p. The argument is straightforward: since m and p are simultaneous causes of p′ [by (3) and (4)], and m and p do not jointly overdetermine p′ [by (7)], it follows by Exclusion that m = p. Since we assumed nothing about m except that it is a cause of some other mental event, what we have here is a proof that any mental event that causes another mental event is identical to a physical event. If we assume, instead of (1) that m causes a physical event, we can show that any mental event that causes a physical event must also be identical to a physical event. So we have shown that any mental event that causes any other event— mental or physical17—must be identical to a physical event. We have, in other words, a new argument for the token identity theory (or token physicalism) first proposed in Davidson 1970. Unlike Davidson’s argument for the theory, however, this one does not assume what Davidson called the “nomological character of causality”—that causes and effects must be related by deterministic laws—which many philosophers now find implausible, so perhaps this argument is more compelling than Davidson’s.

17

Again, this dichotomy may not be exhaustive, but the argument goes through no matter what kind of property we consider in place of M.

116 But clearly Kim does not think he is giving us an argument for the token identity theory. He thinks he is showing the inconsistency of Causal Efficacy with the assumptions he attributes to the nonreductive physicalist and certain “general metaphysical constraints”. So what has gone wrong? It would be most implausible to suggest that Kim has simply confused properties with their instances because of his abbreviating convention. Rather, we suggest that Kim is making use of an unstated premise which he likely thinks of as yet another “general metaphysical constraint”. It’s not difficult to see what this hidden premise might be, as Kim makes it explicit in his earlier work on events. It is an assumption about the identity conditions of events. To express this assumption we must first introduce a bit of notation from Kim 1976: let [x, F, t] denote the event (if any) in which object x possesses property F at time t.18 Using this notation, Kim (1976, 35) states what he calls the “identity condition”: Identity Condition [x, P, t] = [y, Q, t′] just in case x = y, P = Q, and t = t′. The Identity Condition implies that if F and G are distinct properties and f and g are instances of F and G respectively, then f ≠ g. Given this, we can conclude from steps (1), (4), and (6) of the first argument that m ≠ p, and we can complete the argument: M (7) m ≠ p. [From (1), (4), and (6) by the Identity Condition.] (8) p′ is not causally overdetermined by m and p. [From (4) by No Overdetermination.] (9) m does not cause m′. [From (5), (7), and (8) by Exclusion.] With the introduction of the Identity Condition, however, the logical problem has been transformed into a dialectical problem for Kim.

18

Again, if F is an n-place relation, x is an ordered n-tuple of objects.

117 6.

The dialectictal prolem

The dialectical problem is that if Kim does include the Identity Condition among his assumptions, he can no longer claim to have shown that Causal Efficacy is inconsistent with nonreductive physicalism plus his metaphysical constraints. The Identity Condition implies that each event is an instance of exactly one property 19 —which is expressed using Kim’s notation as: (F)

If [x, F, t] = [y, G, t′], then F = G.

Let us call any view of events that is committed to (F) fine-grained, and let us call any view of events committed to (F)’s denial coarse-grained. Let us now consider the question: What is nonreductive physicalism? According to one popular answer, nonreductive physicalism just is the conjunction of token physicalism with the denial of type physicalism.20 But to have a finegrained conception of events is to assume that token physicalism implies type physicalism, 21 which is—if you accept that popular answer—well nigh synonymous with the statement that nonreductive physicalism is false. On this conception of what nonreductive physicalism is, Kim’s supervenience argument, if it is to be viewed as an argument against nonreductive physicalism, reduces to the claim that the nonreductive physicalist gets the identity criteria of events wrong, and the various assumptions about causation and supervenience do no work at all. This is not, of course, the only possible characterization of nonreductive physicalism,22 but it is certainly the classic one. Recall that token physicalism was an essential component of the positions defended in the two ur-documents of nonreductive physicalism—Davidson 1970 and Fodor 1974. Psychophysical supervenience played no role in their arguments, and it only became a physicalist staple due to Kim’s subsequent 19

Namely, the property that Kim calls the event's “constitutive” property (as explained in section 7 below). 20 Perhaps more accurately, the conjunction of token physicalism with Irreducibility, which is stronger than the denial of type physicalism. The falsity of type physicalism would only require the existence of one property that is not a physical property. 21

As Fancesco Orilia also notes in his contribution to this volume (section 6). He also points out that Kim (1996, 60) explicitly recognizes this implication of his account of events. 22

There are others that appeal to the idea of “realisation” or “constitution”, and that are compatible with a fine-grained conception of events. Cf. Boyd (1980, 82-87; 101103), Cummins (1983, 22-23); Papineau (1993, ch. 1).

118 work. And although Kim regards psychophysical supervenience as a minimal requirement for any form of physicalism (Kim 1998, 15),23 at the same time he no longer regards it as a sufficiently robust relation on which to base a physicalist theory of mind, particularly since supervenience can be defined over “multiple domains” (Kim 1988), rendering it compatible with various forms of parallelism (including epiphenomenalism). Notice, however, that if psychophysical supervenience is defined over a single domain of events, it entails token physicalism, for every mental event will be a physical event. At any rate, token physicalism remains a highly plausible candidate for a “minimal” physicalist commitment, and it still enjoys wide acceptance as a position that purposively distinguishes itself from type physicalism. Consequently, to construct an argument against nonreductive physicalism, one of whose premises effectively says that token physicalism implies type physicalism, is to adopt a rather odd dialectical strategy. We will consider in a moment the likely response that classic (token physicalist) versions of nonreductive physicalism lack the resources to account for mental causation (at least to the extent that they are silent about the causal role of properties), and so cannot be taken seriously as versions of nonreductive physicalism. But first there is a further problem with Kim’s argument that needs to be considered. 7.

A further problem

The further problem is that the Identity Condition does not appear to be consistent with Exclusion—at least not if we understand “overdetermination” in a particular way, which seems to us natural. Exclusion implies that if an event has two distinct synchronous causes, then it is overdetermined by them. Brutus’s killing of Caesar (call this event “BKC”) and Brutus’s murdering Caesar (call this “BMC”) are synchronous events that have many common effects. But clearly BKC and BMC cannot overdetermine their common effects if we assume, as we have done, that overdeterminers must be independent at least in that each 23

Unsurprisingly, since on his fine-grained conception of events, physicalism must minimally require that mental events, albeit distinct from physical events, somehow “depend on” and be “determined by”, physical events. And “realisationist” versions of nonreductive physicalism obviously imply supervenience, since the realisation relation implies the supervenience relation (as Kim recognizes in his 1998, 23-24).

119 could have occurred without the other (the murder could not have occurred without the killing). Now, because murdering and killing are two distinct properties (relations), the Identity Condition implies that BKC ≠ BMC. But because BKC and BMC are synchronous events with at least one common effect, and BKC and BMC do not overdetermine any of their effects, Exclusion implies that BKC = BMC. 24 This is not an outright logical inconsistency: no contradiction follows from the set {Exclusion, Identity Condition} alone. To avoid having to choose between Exclusion and the Identity Condition, and thus giving up his supervenience argument, Kim can argue that: (a) BKC and BMC are not synchronous, (b) BKC and BMC have no common effects, (c) BKC and BMC do, after all, overdetermine all of their common effects, or (d) the property of killing is the same as the property of murdering. None of these seems very promising to us. There is, however, a fifth alternative: (e) Kim could reply that the event description notation “[x, P, t]” used in the formulation of the Identity Condition is not to be understood in terms of the commonsense idea of an object having a property at a time, but in terms of a technical notion of a “constitutive property” of an event. The idea would be that the “P” in “[x, P, t]” always denotes a special, unique property “constitutive” of the event denoted by “[x, P, t]”, and that not every property x has at t can be a constitutive property of an event. It should be clear how this distinction would enable Kim to dodge the objection about BKC and BMC: the application of the Identity Condition in our argument for the distinctness of BKC and BMC would simply be invalid, because the argument does not have premises stating that killing and murdering are constitutive properties and hence we are not entitled to conclude that there exist any such events as [〈Brutus, Caesar〉, kills, t] and [〈Brutus, Caesar〉, murders, t] for us to apply the Identity Condition to. (Of course, if the Identity Condition and Exclusion are both to be maintained, it had also better be the case that at least one of killing and murdering is not a constitutive property, or else the argument could be supplemented with additional true premises to yield a refutation of the Exclusion/Identity Condition combination. We do not, however, have any arguments to show that murdering and killing are constitutive properties.) We believe that Kim would choose alternative (e) for coping with the present problem—the distinction between constitutive and other properties of events is, in fact, drawn by Kim in his 1976. Although in that paper Kim 24

As one would expect under a coarse-grained conception of events.

120 resists the identification of such events as BKC and BMC,25 he is open to the identification of other events involving the instantiation of distinct properties by an object.26 But choosing alternative (e) comes at a price: To begin, Kim would have to reject the principle that [x, P, t] exists if and only if x has P at t (an unrestricted form of what Kim 1976, 35, calls the “existence condition” for events). 27 But if this principle is rejected, the 25

On p. 44 Kim indicates that he would not identify Brutus’s stabbing Caesar with Brutus’s killing Caesar, and on p. 43 he says that “no stabbings are killings and no killings are assassinations”. These are similar enough to our examples. 26 On p. 44 Kim concedes that Brutus’s stabbing Caesar and Brutus’s stabbing Caesar with a knife are the same event and he accordingly denies that the relation x stabs y with a knife is a constitutive property of the event denoted by “Brutus’s stabbing Caesar with a knife”, though it clearly is a property of the pair 〈Brutus, Caesar〉. 27 Kim intends his “existence condition” schema to be understood with the restriction that only predicates that ascribe constitutive properties may be substituted for “P” (Kim 1976, 34-37). Our own attitude about what events exist is liberal: we do not think our existence schema “The event in which x has P at t exists iff x has P at t” (section 2) is in need of restricting. Events in which properties of every kind are instantiated are required for the semantic analysis of natural language. Only some events can enter into the causal relation—only the metaphysically contingent ones, we suppose, viz. events e such that it is metaphysically possible both for e to have occurred and for e to not have occurred. For discussions of causation, our existence schema could be restricted so that only predicates that ascribe contingent properties (properties such that possession of them is a metaphysically contingent matter) may be substituted for “P”, and the semanticist’s events could be called something other than “events” (“schmevents”, for example), but this is a question of word choice, not substance. Such a restriction is intelligible inasmuch as the notion of metaphysical possibility is, and, it seems to us, it would correctly disqualify as events those schmevents that cannot enter into the causal relation. Note that the question whether to restrict the schema (E1) The event in which x has P at t exists iff x has P at t, which we used in section 2, is different from that of whether to restrict the schema (E2) [x, P, t] exists iff x has P at t, which is what Kim is concerned with in his 1976. The latter is a partial definition of a new piece of notation (“[”, “]”), call this the canonical event description notation. Ruling out a particular predicate “R” as an allowable substituend for “P” in (E2) would not have the consequence that there is no such thing as the event in which x has R at t; it would only have the consequence that that event, if it exists, is not denoted by “[x, R, t]”—it might be denoted by some other canonical event description. Presumably Kim

121 meaning of Kim’s event description notation is no longer clear: to understand it we would need an account of just what it is that makes some properties “constitutive” and others not, and none has been provided by Kim. We do not claim that such an account cannot be provided—the salient point is, rather, that for anyone who is worried about the coherence of the Exclusion/Identity Condition combination, as we are, the acceptability of Kim’s supervenience arguments will depend on his ability to produce such an account. What’s more, if Kim chooses (e), a new logical problem is introduced into the argument. The problem is that option (e) renders not only our argument against the coherence of the Exclusion/Identity condition combination but also Kim’s own supervenience arguments invalid, for exactly the same reason: if the Identity Condition is to be applied to m, M, p, and P in the supervenience arguments, the arguments will require two additional premises to remain valid: 1) that M is a constitutive property of m, and 2) that P is a constitutive property of p. Perhaps it will turn out that both claims fall out of the correct account of constitutive properties—even though since writing his 1976, Kim has indicated that, on pain of having to revise his property-exemplification account of events, he may have to deny that mental properties can be constitutive properties of events!28 However that may be, it should be clear that if option (e) is chosen, some further work on Kim’s part will be required to make the supervenience arguments convincing.

would not impose the same restriction on the ordinary language (E1) as on the quasiformal (E2), since in 1976 he wants to allow events to instantiate properties which are not constitutive of them (see our note 24). The restriction Kim proposes for (E2) differs from the one just contemplated for (E1) in that the former is not made using terms of which we have a prior understanding but by using the technical term “constitutive property”, which Kim does not explicitly define (see pp. 36-37). 28 Kim notes that “a revision of the standard property-exemplification account of events (essay 3) [Kim 1976]” may be called for, “especially if mental properties, in spite of their multiple realisability, are accepted as legitimate event-generating properties. For on the standard account two property instances count as distinct events if the properties instantiated are distinct ... . Considerations advanced in Kim (1992) concerning disjunctive properties may be reason enough for excluding mental properties as constitutive properties of events” (Kim 1993, 364-65, note 5, our italics).

122 8.

A problem about mental quausation?

Kim is well aware, of course, that there are token physicalists who are not type physicalists. Does he think his supervenience argument has anything to say to them? Surprisingly, he does. In footnote 9 in ch. 2 of PSNE and elsewhere (e.g. Kim 1998, ch. 4) Kim hints that his argument could be reconstructed so as to have bite against nonreductive token physicalist positions. Unfortunately, however, Kim never gives us an explicit reconstruction. Though it would be obviously unfair to criticize an argument never explicitly formulated, we would still like to register the reason for our scepticism that a plausible argument along the lines Kim hints at can be made. Kim says (PSNE, 42, n. 9) that in the reconstructed argument, “An M-instance causes a P-instance” must be understood with the proviso “in virtue of the former being an instance of M and the latter an instance of P”.

The argument would then be one about quausation (to use Terence Horgan’s (1989) term): causation qua something. 29 But then we would need a quausal exclusion principle to replace Exclusion. What might it be? The trouble is that Kim’s suggestion does not determine a unique translation of Exclusion into quausal terms, and we are left wondering what principle he might have had in mind. The most straightforward rendering of Exclusion into quausal terms that we can think of is: Quausal Exclusion 1 Except in cases of overdetermination, if F and G are distinct properties, and an instance c of F causes an event e in virtue of c’s being an instance 29

The closest Kim comes to providing an argument is in the following: “Suppose that a certain event, in virtue of its mental property, causes a physical event. The causal closure of the physical domain says that this event must also have a physical cause. We may assume that this physical cause, in virtue of its physical property, causes the physical event. The following question arises: What is the relationship between these two causes...?” (1989, 280). Kim goes on to suggest that, barring overdetermination and given closure, the physical cause excludes the mental one unless the mental and the physical properties of the two causes are identified. The implication here is that the mere “token identity” of two causes as coarse-grained events won’t give us a solution to the problem; what needs to be identified is “that in virtue of which” the cause causes what it does.

123 of F, then it is not the case that c causes e in virtue of c’s being an instance of G.30 This principle, however, has no intuitive support that we can find, and we think it would be unfair to attribute it to Kim. Similar principles have been ably criticized by Fodor (1989), Jackson and Pettit (1990; see (B) at p. 110 for a straightforward counterexample), Heil and Mele (1991), and Yablo (1992), and we have little to add to those critiques beyond pointing out that a problem analogous to the one considered earlier about BKC and BMC would arise again: given Quausal Exclusion 1, where e (for example, Calpurnia's grieving) is an effect of Brutus' action of killing/murdering Caesar, BKC causing e in virtue of BKC’s being a murdering of Caesar would exclude BKC’s causing e in virtue of BKC’s being a killing of Caesar—surely something we should have no reason to claim. Let us instead look to Kim’s other writings to see if a quausal exclusion principle might be derived from other claims he has made. In Kim 1988 we find a “principle of explanatory exclusion” (PEX), which states: “No event can be given more than one complete and independent explanation” (p. 239). There Kim also concedes that PEX is “something that many will, I’m afraid, consider absurdly strong and unacceptable” (ibid.). Absurdly strong or not, if we also assume Kim’s “explanatory realism”,31 we could try to make a case for the following principle on the basis of PEX: Quausal Exclusion 2 Except in cases of overdetermination, if F and G are distinct and independent properties, and an event c causes another event e in virtue of c’s being an instance of F, then it is not the case that c causes e in virtue c’s of being an instance of G. For lack of space we will not attempt to show just how Quausal Exclusion 2 might be derived from PEX; the interested reader may consult the 30

This is essentially the principle attributed to Kim by Marras, 2000, p 145, as principle (Q*). Note that is addition to a quausal exclusion principle, the reconstructed argument would also require analogous revisions of Closure and/or of SC I and SC II, as well as of any premise to the effect that a mental (physical) event causes another event. (E.g., Closure would become Quausal Closure: “If a physical event e has a cause that occurs at time t, then it has a physical cause c that occurs at t and that causes e in virtue of some physical property of which c is an instance”.) 31 Which is itself far from obvious. See Marras 1998 for discussion.

124 argument of Macdonald and Macdonald (2006, 544-545), which uses PEX to justify a principle very similar to Quausal Exclusion 2. Rather, our main point about Quausal Exclusion 2 is that, however plausible or implausible PEX may be, and whatever difficulties might be involved in using PEX to justify Quausal Exclusion 2, if Quausal Exclusion 2 is the principle that will be used in the new supervenience argument, then there is a straightforward reply. The reply has been around in the mental causation literature, in various forms, for a few years.32 It is simply this: Quausal Exclusion 2 cannot be used to argue that the causal efficacy of mental properties is “excluded” by that of their supervenience bases, because such an application of Quausal Exclusion 2 would require mental properties and their supervenience bases to be independent. By the very definition of supervenience, there is a necessary connection between mental properties and their supervenience bases, so mental properties are not independent of their supervenience bases. The “quausal” formulation of the mental causation problem aims to highlight the point that mental causation “ultimately involves the causal efficacy of mental properties” (Kim 1998, 37), and Kim’s challenge to the token physicalist is to explain how a mental event, even if token-identical with a physical event, can cause what it does in virtue of the mental properties it instantiates (where these are distinct from the physical properties it instantiates), and why the causal efficacy of the mental properties is not preempted by the efficacy of the physical properties. However, to merely complain that token physicalism, as such, lacks the resources to account for the efficacy of mental properties33 is not to the point: the relevant issue is whether token physicalism can coherently be supplemented with additional “substantive premises” (e.g., about the physical realisation of higher-level properties, or their implementation in physical mechanisms, etc.)34 so as to give mental properties a causal role— a role which token physicalism as such does not preclude. 32

See Macdonald and Macdonald 2006, 566, Bennett 2003, and Papineau 1993, ch. 1, sections 6 and 7. 33 This was, essentially, Kim’s (1993) complaint against Davidson’s anomalous monism. While granting that Davidson was “arguably right” in denying that anomalous monism entails the causal inertness of mental properties, he nonetheless insists that anomalous monism “fails to provide mental properties with a causal role” (p. 20). As argued in Marras (1997), this was merely a failure of omission: Davidson’s monism was not intended to provide a theory of the causal efficacy of mental properties. 34 See, e.g., Fodor 1989, McLaughlin 1989, Jackson and Pettit 1990, Hardcastle 1998.

125 9.

Is there a problem about token identity?

Why has the fact that Kim’s supervenience argument begs the question against nonreductive physicalist positions that assume token identity not received wide attention? There are two fairly common beliefs that might be responsible for this. One is that “the problem of how events should be individuated” “plays no essential role in questions about epiphenomenalism” (Shapiro and Sober forthcoming, n. 3)—and this is just what Kim appears to think. The other is that there is something wrong with token physicalism, and that nonreductive physicalism ought to be formulated in terms of a Kim-style fine-grained conception of events. Our response to the first belief is implicit in what we said in the previous two sections: the supervenience arguments Kim actually gives us depend essentially on an assumption he makes about the identity criteria of events, and it is questionable whether Kim can construct a convincing alternative to the actual supervenience arguments without making that assumption. Clearly, then, the way we individuate events does play an essential role at least in the questions about epiphenomenalism raised by Kim. Our response to the second belief is that there quite simply are no good arguments against token physicalism; or, less tendentiously and more to the point, that there are no good arguments against token physicalism that are not equally good arguments against type physicalism. 35 (This is unsurprising, since type physicalism entails token physicalism; so if the latter is shown to be false, so is the former.) The arguments against token physicalism are notoriously controversial,36 and at any rate, they are of no help to a friend of type physicalism like Kim since, as just remarked, if any

35

Kripke’s well-known modal arguments against the identity theory, for example, can equally be directed against the type and the token identity theory. So can any of the familiar arguments based on the conceivability/possibility of zombies and/or disembodied minds. Such arguments must of course be distinguished from arguments against nonreductive (versus reductive) physicalism, either of the token identity variety or of the realisation variety. These latter arguments need to show not that token identity is false, but that the denial of type identity is false. Typically these aim to show that the multiple realisability thesis is false, or that it does not stand in the way of a reductive account of mental properties. See, e.g., Kim 1992, Richardson 1979, Bickle 1998. 36 For a recent, forceful critique of such arguments see Papineau 2002.

126 of these arguments defeat token physicalism, they defeat type physicalism.37 37

Consider, for example Burge’s (1979) argument against token physicalism. The argument is something like this: Tokens of those mental states that are broadly individuated cannot be identical to tokens of the physical states that realize them because the realizing properties are plausibly intrinsic ones. Suppose, for example, that Joe’s belief that p at time t is realized by Joe’s brain state S (“Joe’s S” for short). If the content of Joe’s belief is broadly individuated, then there is a possible world w in which Joe is intrinsically exactly like he is in the actual world, so he has S at t in w, but in which Joe’s does not believe that p at t because of some difference in his environment. Now it would seem that on the basis of this we can argue: (1) It is possible Joe’s belief that p occurs but Joe’s S does not occur (premise). (2) It is possible that Joe’s belief that p ≠ Joe’s S (from (1) by Leibniz’s law and propositional modal logic). (3) Joe’s belief that p ≠ Joe’s S (from (2) because objects that are possibly distinct are actually distinct). The argument can clearly be reformulated so as to apply not just to Joe’s (token) belief that p and to Joe’s S, but also to the types (properties) believing that p and being in S. After all, “It is possible that Joe’s belief that p occurs but…” is necessarily equivalent to “It is possible that believing that p is exemplified by Joe but…” That having been said, we note that the argument is not particularly convincing when applied (as intended) against token physicalism. That is, while it is true that if the argument is successful in refuting token physicalism, it is also successful in refuting type physicalism, we do not think that the argument is successful against token physicalism. The trouble is that, as a matter of modal logic, in order for the move from (1) and (2) to (3) to be valid, or, equivalently, in order for ‘‘a ≠ b → a ≠ b’ to be a logical truth, ‘a’ and ‘b’ must be rigid designators. However, the token events at issue in the argument are denoted by definite descriptions, which are not rigid—at least not rigid de jure. Definite descriptions may, however, be rigid de facto if they happen to pick out their designata by their essential properties (e.g. ‘the even prime’ is rigid de facto). Burge of course recognizes this—he claims that tokens of beliefs have their contents essentially (Burge 1979, 75), but this is far from obvious. In fact, the very externalist considerations Burge brings to bear on token physicalism seem to militate against this conclusion. It seems to us entirely natural to describe the moral of Burge (1979) by saying things like: “There are counterfactual circumstances in which Joe’s belief that p at t would have had a different content”, “If the stuff in the oceans had been XYZ, then (the present token of) my belief that water is wet would have had a different content than it in fact does”. (Note that the argument would be valid if the descriptions in (1) were read as having wide scope relative to the modal operator. However, if (1) were read this way, it would receive no support from Burge’s

127 The reasons that have led some philosophers to prefer “token realisationism” to token identity physicalism are far from persuasive, and/or are largely motivated by a prior commitment to a fine-grained conception of events. Boyd (1980), for example, claims that his realisationist version of nonreductive physicalism enables him to cope with Kripke’s arguments against identity theories. But he himself shows, in the same paper, how identity theories are quite able to resist Kripke’s arguments; so what makes realisationism preferable to token identity? And Papineau (1993, 24) explicitly acknowledges that his reason for preferring realisationism is that, qua terms of causal relations, events are best viewed as fact-like entities, or as structured, fine-grained events à la Kim. If so, how we individuate events again does, pace Kim, play a role in the mental causation debate. 10.

Concluding (polemical) remarks

Our critique of Kim’s exclusion argument aimed to show that the reasons he gives for holding that nonreductive physicalism entails epiphenomenalism are not convincing, resting as they do on a conception of events that rules out from the start a classic and perfectly coherent version of nonreductive physicalism that we believe still merits consideration—a conception of events that is, furthermore, only dubiously consistent with another central premise of his arguments. To have accomplished this aim is not, of course, to have provided a positive account of how mental causation is possible within the confines of nonreductive physicalism thus construed. To provide such an account one externalist considerations: the latter only show that the mental properties of a person can come apart from the physical properties that, in the actual world, realize them. They do not show that the events in which these properties are instantiated in the actual world are distinct in some other world.) If, on the other hand, the argument is made using definite descriptions that designate properties instead of events, there is no problem about rigidity. Any property—at least any property we have a predicate for—can be designated by a rigid definite description that makes use of the predicate we would normally use to ascribe the property. For example “the property of being red”, or “the property red” are rigid: in no world does “the property of being red” designate any property other than redness. Similarly, “the property of believing that p” in no world designates anything other than the property of believing that p.

128 would have to explain, minimally, how mental events can have causal efficacy not merely in virtue of their being token-identical with physical events, but also in virtue of the mental properties they instantiate. Although the question of how mental causation is possible is surely a legitimate one, it is not one that we need to address here; it suffices to have shown that there is no compelling argument that Kim (or anyone else, to our knowledge) has provided for the conclusion that a satisfactory account of mental causation cannot be provided within the bounds of nonreductive physicalism. Still, while the question of how mental causation is possible is legitimate and important, there is something perplexing about the almost exclusive attention that it has received from Kim and others for so many years. There is no prima facie reason to suppose that the question of how mental causation is possible is more special, or more difficult, than the question of how biological, or chemical, or, indeed, physical causation is possible.38 Moreover, Kim’s worry that physical causation may somehow “exclude” mental (or other higher-level) causation presupposes that there is physical causation, and that it is somehow less problematic than mental or other higher-level causation. These presuppositions, however, cannot be taken for granted. In fact, we think this stance gets matters backwards: if the reality of any causation can be taken for granted, we think the reality of causation at higher levels, including the mental, should be, whereas the reality of physical causation should be considered an open question.39 The reason is this. The concept of causation we are interested in is not a philosophers’ invention; it does not get its content from metaphysical principles—Closure and others—of the sort that are discussed in the literature on mental causation, but from its use in science and commonsense understanding. And if we are interested in finding out what really causes what, we should listen to what science in particular has to say about this matter. Now it happens that the concept of cause is found only in the special sciences, and never in fundamental physics. That being so, we should accept the reality of higher-level causation (the kind that the special sciences talk about), including mental causation, and remain noncommittal, until some arguments are provided, on the reality of physical causation. 38

That Kim thinks that the problem of mental causation poses special difficulties is evident in ch. 3 of his 1998 book, especially in view of his claim that the supervenience argument “does not generalize” to other domains beyond the mental. 39 For a similar stance and a thorough discussion of the issues, see Ladyman and Ross 2007, ch. 5.

129 Unless physics changes dramatically and its practitioners begin to talk about causation, these arguments will have to be philosophical arguments. Perhaps the most straightforward argument for physical causation is one that assumes token physicalism and the reality of causation at some higher level, call it L (L could be, but doesn’t have to be, the mental level): all Lcauses are events, and all events are physical, so some physical events are causes. But this still leaves open the question of whether physical events cause other events in virtue of being instances of particular physical properties. These may seem like strange things to say, but consider the fact that physical events are instances of physical properties. We assume that physical properties are those properties that are ascribed by predicates that occur in quantum physics (or in whatever the correct lowest-level theory turns out to be), or by open sentences containing only such predicates. Instances of physical properties are virtually never cited as causes, at least not under their physical descriptions. Thus it takes an argument to show that instances of physical properties can be causes. If we assume that all instances of higher-level properties, some of which are commonly cited as causes both in (higher-level) science and common sense, are identical to instances of physical properties, then it is clear that there is physical causation all around us; but this still leaves all the physical events that are not identical to instances of any higher-level properties unaccounted for (presumably there are many such physical events), as well as the causal powers of the physical properties themselves. Trying to identify the causes and effects, and the causally efficacious properties, in the world as described by physics is not an easy task; it would require us to have a correct and informative analysis of the conditions under which one event can be said to cause another, as well as of the conditions under which a property instantiated in a cause can be said to be causally efficacious, and it’s not obvious that we have such an analysis.40 Not only is the reality of physical causation less obvious than the reality of higher-level causation, but arguments similar to those found in the mental causation debate could be used for “excluding” physical causation. The problem could be put in terms of a Malcolm 1965-style explanatory exclusion principle: given that we can have, in principle, a complete covering-law explanation for every physical event in the 40

However, there are some promising developments: recent attempts to explicate causation in terms of the outcomes of counterfactual interventions (Woodward 1997, Pearl 2000) might provide us with the means to do just this.

130 vocabulary of quantum physics, and that these explanations make no use of the concept of cause, what role is there left for physical causation to play? Physical causation is “excluded” by the initial conditions and the laws of physics, which determine, without positing any metaphysical glue binding one event to another, the occurrence (or objective probability of the occurrence) of every physical event. Not that we think this is a good argument, but its similarity to some of the arguments used to motivate epiphenomenalist worries about mental events and properties should raise some questions about the seriousness of those worries. REFERENCES Bennett, K. (2003). Why the Exclusion Problem Seems Intractable, and How, Just Maybe, to Tract It. Noûs, 37, 471-497. Bickle, J. (1998). Psychoneural Reduction: The New Wave. Cambridge: MA: MIT Press. Boyd, R. (1980). Materialism without Reductionism. Readings in the Philosophy of Psychology, Volume I, ed. N. Block. Cambridge, MA: Harvard University Press, 67-106. Burge, T. (1979). Individualism and the Mental. Midwest Studies in Philosophy, 4, eds. P. French, T. Uehling H. and H. Wettstein, 73-121. Reprinted in Externalism and Self-Knowledge, eds. P. Ludlow and N. Martin, Stanford: CSLI, 21-84 (page references to the latter). Cummins, R. (1983). The Nature of Psychological Explanation. Cambridge, MA: MIT Press. Davidson, D. (1970). Mental Events. D. Davidson, Essays on Actions and Events, Oxford: Oxford University Press, 1980, 207-227. Fodor, J. (1974). Special Sciences. J. Fodor, RePresentations. Cambridge, MA: MIT Press, 1981, 127-145. Fodor, J. (1989). Making Mind Matter More. Philosophical Topics, 17 , 59-80. Hardcastle, V. (1998). On the Matter of Minds and Mental Causation. Philosophy and Phenomenological Research, 58, 1-25. Heil, J. and Mele, A. (1991). Mental causes. American Philosophical

131 Quarterly, 28, 61-71. Horgan, T. (1989). Mental Quausation. Philosophical Perspectives, 3, 4676. Jackson, F. and Pettit P. (1990). Program Explanation: A General Perspective. Analysis, 50, 107-117. Kim, J. (1976). Events as Property Exemplifications. Action Theory, eds. M. Brand and D. Walton, Dordrecht: Reidel, 159-177. Reprinted in Kim 1993, 33-52 (page references to the latter). Kim, J. (1988). Supervenience for Multiple Domains. Philosophical Topics, 16, 129-150. Reprinted in Kim 1993, 109-130. Kim, J. (1992). Multiple Realization and the Metaphysics of Reduction. Philosophy and Phenomenological Research, 52, 1-26. Reprinted in Kim 1993, 309-335 (page references to the latter). Kim, J. (1993). Supervenience and Mind. Cambridge: Cambridge University Press. Kim, J. (1998). Mind in a Physical World. Cambridge, MA: MIT Press. Kim, J. (2005). Physicalism, or Something Near Enough. Princeton, NJ: Princeton University Press. Ladyman, J. and Ross, D. (2007). Everything Must Go: Metaphysics Naturalized. Oxford: Oxford University Press Macdonald, C. and Macdonald G. (2006). The Metaphysics of Mental Causation. Journal of Philosophy, 53, 539-577. Malcolm, N. (1968). The Conceivability of Mechanism. Philosophical Review, 77, 45-72. Marras, A. (1997). The Debate on Mental Causation: Davidson and His Critics. Dialogue, 36, 177-195 Marras, A. (1998). Kim’s Principle of Explanatory Exclusion. Australasian Journal of Philosophy, 76, 439-451. Marras, A. (2000). Critical Notice of Mind in a Physical World. Canadian Journal of Philosophy, 30, 137-160. Marras, A. (2007). Kim’s Supervenience Argument and Nonreductive Physicalism. Erkenntnis, 66, 305-327. Orilia, F. (2008). Basic Ontology, Multiple Realizability, and Mental Causation. This volume.

132 McLaughlin, B. (1989). Type Epiphenomenalism, Type Dualism, and the Causal Priority of the Physical. Philosophical Perspectives, 3, 109-136. Papineau, D. (1993). Philosophical Naturalism. Oxford: Blackwell. Papineau, D. (2002). Thinking About Consciousness. Oxford: Oxford University Press. Pearl, J. (2000). Causality. New York: Cambridge University Press. Richardson, R. (1979). Functionalism and Reductionism. Philosophy of Science, 46, 533-558. Shapiro, L. and Sober E. (forthcoming). Epiphenomenalism—the Do’s and the Don’t’s. Studies in Causality: Historical and Contemporary, eds. G. Wolters and P. Machamer, Pittsburgh: University of Pittsburgh Press. Woodward, J. (1997). Explanation, Invariance, and Intervention. PSA 1996, vol. 2, 26-41. Yablo, S. (1992). Mental Causation, Philosophical Review. 101, 245-280.

Tropes’ Simplicity and Mental Causation SIMONE GOZZANO Università di L’Aquila [email protected] ABSTRACT. In this paper I first try to clarify the essential features of tropes and then I use the resulting analysis to cope with the problem of mental causation. As to the first step, I argue that tropes, beside being essentially particular and abstract, are simple, where such a simplicity can be considered either from a phenomenal point of view or from a structural point of view. Once this feature is spelled out, the role tropes may play in solving the problem of mental causation is evaluated. It is argued that no solution based on the determinable/determinate relation is viable without begging the question as regards the individuating conditions of the related properties. Next, it is shown that Robb’s solution, much in the spirit of Davidson’s anomalous monism, entails abandoning the assumption that tropes are essentially simple, a consequence that I find not acceptable. My conclusion is that these entities are of no help in solving the problem of mental causation, and that a universalist approach should be preferred.

1.

Introduction1

In the past few years, a number of authors have argued that tropes can play a crucial role in solving many of the conundrums connected with the causal status of mental properties. In particular, if one adheres to Davidson’s anomalous monism, as Robb 1997 does, these entities, more than properties conceived as universals, are appropriate to meet the many requirements that are set to face the issue of mental causation. My aim in this paper is to argue that tropes are not suitable for such a task. In particular, the key assumption that allows tropes to carry the burden of the solution is, in my view, incompatible with one of the crucial features of tropes themselves, that is, their simplicity. In order to clarify all this, the first step is that of qualifying tropes by making explicit their essential features.

1

For comments on a previous draft, and a far ancestor, of this paper I express my gratitude to Francesco Orilia and David Robb. I thank Vicky Michela for the precious editing she provided.

134 2.

Tropes

The notion of trope has not had a smooth history in philosophy. It could be traced in notions such as individual accident in Aristotle and in the Scholastics, mode in Locke (see Lowe, this volume) and property of monads in Leibniz. However it received greater attention in the last century, when Williams 1953 decided to use the same word Santayana used to pick the essence of an occurrence. Williams’ end, however, was the opposite: he took “trope” to pick the occurrence of an essence. Campbell (in both his 1981 and 1990 works), following Stout 1921, has defined tropes as abstract particulars, thus interrupting a tradition that contrasted properties, taken as abstract entities, with particulars, considered as concrete items. His idea was that the two contrasts, abstracts vs. concretes and universals vs. particulars, were logically independent one from the other, and so it was conceptually admissible to scrutinize other possible intersections beyond the usual two abstract universals and concrete particulars. It is with this spirit that abstracts particulars have been isolated as crucial items to face many of the dilemmas concerning properties taken as universals (and Williams has even mentioned “concrete universals”, such as Socratesity). One of the main reasons to introduce tropes in metaphysics has been that of placing universals, so to say, down at ground zero. Campbell 1981 takes tropes to have moderated the metaphysical scandal of imagining entities, as universals are taken to be, that are scattered through space and time while enjoying the paradoxical form of being wholly present wherever and whenever they are instantiated. On the contrary, tropes are not repeatable entities: any trope completely exists in a specific space-time location, hence they are particulars. To compare universals and tropes let us consider red. If red is taken as a property, a universal, it could be considered as wholly existing in each singular instance even if no instance is necessary for its existence, provided that there is at least one instance. On the other hand, if red is taken as a trope, we have to interpret it as thisred-now, and it is thoroughly realized in a specific space-time location, in such a way that it cannot be repeated. What is more, tropes have to be space-time located. This is a substantive issue: for instance, Armstrong 1989 thinks that unistantiated universals, possibly defended by Plato in the Republic, should not be accepted, being the upshot of a semantic fallacy resulting from the assumption that every predicate gets its meaning from a prior existing universal. But the possibility of unistantiated tropes is self-

135 contradictory, given their intrinsic space-time nature. This might entail that postulating tropes presupposes something like a realistic stance on the space-time structure, a point I wish to leave aside. At the same time, and here we come to the second distinguishing feature of tropes, this-red-now can be space-time compresent with other tropes. For instance, following Williams, consider a lollipop: it is red, sweet and round. So, this-red-now, this-sweet-now and this-round-now all share the same space-time location. Now, if two or more concrete items were in the same space-time location they would be identical; but this-rednow, this-sweet-now and this-round-now are not identical, having quite different individuation conditions. Because they are in the same space-time and are not identical, they must be abstract. There is a way to challenge this argument: this statue and the clay it is made of are in the same (region of) space-time but they are not identical: the clay, but not this statue, could have been of a different shape. However, between this-red-now and thissweet-now there is not a relation of constitution, like one that holds for the statue and the clay. So, we may counter the above argument by saying that compresent entities are abstract, provided that they are not related by a constitutional relation.2 This second feature of tropes, abstractness, received a cognitive reading in Campbell’s 1981 paper. He thinks that we may get the lollipop’s redness by abstracting it away from its other tropes, for instance its flavor and shape. This does not make the red in question a purely mental feature: that red is perfectly objective, but it needs a cognitive act to isolate it. Hence, the abstractness of tropes, according to Campbell, is not the consequence of space-time compresence, but rather of an act of abstraction, or mental focusing. In the following, I will mainly insist that the space-time compresence is the crucial feature for abstractness, even if nothing will crucially depend on such an assumption. A third feature of tropes is, as Maurin 2002 insists, their simplicity. This seems to be the result of their being both particular and abstract. In fact, particularity and abstractness are individuation conditions for tropes in that tropes are specific features of a given space-time location. As Campbell uses to say, any abstract feature of a space-time location 2

Simons 1994 notes that an entity is abstract if it is not in space and time. For him, then, tropes are not abstract in this sense; rather they are abstract because they exist only inasmuch as they depend on something else to exist. Abstractness would be parasitic on concreteness, though, thereby excluding a possible world comprising solely abstract entities.

136 “monopolizes” it in that, in that very space-time location, you cannot have a second feature for the same dimension. For instance, if the lollipop is red, in that space-time position you cannot have a different feature in the colour dimension, while you can have some feature for the taste dimension, i.e. sweet. The colour dimension in that space-time location, then, is monopolized by that shade of red. Here “dimensions” can be taken as characterizing qualities, of any kind. This view somewhat entails the old substratum-property theory, which one may want to avoid in favor of a bundle view (Simons 1994, Robb 2005). Moreover, it seems that simplicity would not play any substantive role over and above particularity and abstractness. So, what is the distinguishing feature of simplicity vis-à-vis particularity and abstractness? The concept of abstractness, as we saw, requires the concept of compresence and this, in turn, presupposes that what is compresent cannot be further decomposed, otherwise compresence would require compresence, in an endless regress. Because the compresents that cannot be further decomposed are qualities, then these qualities have to be simple. Simplicity, however, is not conceptually connected to abstractness just in case the latter is interpreted as compresence. It may be thus connected also in case abstractness is taken as the result of an act of abstraction, or conceptual focusing, as Campbell suggests. If I focus on this colour, I abstract such a trope away from all other tropes this entity may have. If I get to a trope such that a further act of abstraction leads to a different result, then the trope I got to was not the simplest one. So, when one gets to a trope such that no further act of abstraction is available then one is abstracting a simple trope. In sum, on both construals the full abstractness of a trope requires the simplicity of the trope. We may look at this issue from a different perspective. According to Williams, tropes participate in two kinds of combination groups: on the one side they concur in the sum that constitutes concrete objects (as said, this lollipop is the sum of at least this colour plus this flavour plus this shape) giving rise to the bundle theory of particulars—this is our notion of compresence; on the other side, each trope falls into the set or class of all tropes that have with it the relation of being precisely similar. Here is how Williams 1953, 117, expresses the point: “Speaking roughly … the set … of tropes precisely similar to a given trope … is the abstract universal or ‘essence’ which it may be said to exemplify”. This is not to say that tropes are the instantiation of universals, rather that some generality is in order in the case of tropes as well. In this way, this-red-now and this-sphere-now

137 are two compresent tropes pertaining to two different classes of similarity. It is in this framework that objects are thought of as bundles of compresent tropes. However, because of the space-time nature of tropes, simplicity may turn out to be a somewhat complicated concept. Here is Ehring: Unlike universals, tropes cannot characterize more than one object at the same time, but tropes can persist over time. We may now add that tropes will be either simple or compound. A simple trope does not have tropes as proper parts. A minimumcharge trope, if there are such minimums, is an example of a simple trope. A compound trope includes another trope as a proper part.

Ehring 1997, 117.3 An object having an electric charge that is not a minimum charge is considered, by Ehring as well as by Campbell, as having compound tropes of the conjunctive kind. Compound tropes can also be of the structural kind. These are individuated by way of their relation to other tropes. Being one meter in length, says Ehring following Armstrong’s description of structural universals, individuates the property that something has as a result of having two adjacent half-meter length tropes (or more than two adjacent tropes of lesser length): “The length of this particular ruler is a structural trope composed of length tropes adjacent to each other” (ibid., 118). However, it seems to me that if the distinction between conjunctive and structural composite tropes is to be established, it deserves a firmer ground. Consider a musical chord of C major: it is the result of playing C, E and G together. As such, it is a compound trope, resulting from the compresence of simpler tropes—C, E and G. However, such a trope, while composite, is phenomenally simple; it is perceived as a single musical item. Similarly, violet is a compound trope formed by red plus black. At the same time it is phenomenally simple, because it is so perceived and taken. Now, neither in the electric charge case, nor in the ruler length, is there any substantial difference in the constituent tropes. This part of the charge is not any different from that part of the charge, if the distinction 3

Hochberg 2004 manifests a deep dissatisfaction with the idea of locating tropes’ simplicity in their having no parts. Its polemical target is Maurin 2002.

138 makes any sense, and the difference between this half-length and that half length of the ruler is, at most, spatial—one is to the left of the other with respect to me—and, as such, it plays no role in the overall length of the ruler. In the case of the chord or violet, the composing tropes are substantially different—C, E and G, red and black—allowing for variation—a stronger playing of the C key on the piano keyboard results in the C major chord having a different sound from that resulting from striking with more intensity the E key, so that the difference results in the composed trope by virtue of differences in the relations between the composing tropes. So, I would take both the charge and the length cases to be representative of conjunctive composite tropes, while the cases of chords and non-primary colours representative of structural composite tropes. In both cases, the analyses can be established only by assuming that there are simple tropes, which form the basis of composition. We may call these structurally simple tropes, while the tropes they give rise to, like C major or violet, phenomenally simple. We will consider a role for such distinction in the following. Furthermore, consider a chair, red and thus-and-so shaped. In its space-time location we have the red trope and the chair-shaped trope. However, let’s further imagine, while the red is homogeneous in all the chair’s parts, it is just the same shade of red, the shape is not. The shape of the seat is different from that of the back: inverting their position with respect to the legs, for instance, would result in a different overall shape, let’s say in one uncomfortable chair. Hence, the shape of the chair is a composite trope, because we can distinguish parts of it that are different along the individuating dimension—shape—and structural. However, the structure of the chair’s shape is more articulated than the musical chord, because in the former case its composing tropes cannot be recombined without variations in the resulting composite trope. It makes no sense to say that the C chord composed of C, G and E is different from that composed of G, E and C, if the respective keys on the keyboard are played simultaneously. So, we have a compound chair-shaped trope, having simple tropes (shape of the seat and shape of the back) as its base of composition. Simplicity, then, is in my view a crucial feature of tropes, allowing us to understand the basic compositional grammar of tropes themselves. Admitting tropes’ simplicity is a necessary step for considering tropes as the alphabet of being.

139 The last point that should be considered with regard to tropes vis-àvis universals is the resemblance relation, taken as a way to regain the type-token distinction. Tropes as such are token properties, while on the type side one must consider classes of resembling tropes, either naturally or nominalistically considered. Laws of nature, for instance, even if expressed in terms of universals, would in fact refer to classes of precisely resembling tropes. Here the notion of resemblance has to be considered a primitive one, so that judging something as resembling something else must be considered as a direct apprehension in an act of acquaintance. For instance, we judge two patches of red as being precisely similar to each other simply by observing them, and taking such observation as a selfstanding justification of any statement of similarity (this is Williams and Campbell’s position). And tropes, taken as the respects in which objects resemble each other, are “realistically conceived universals” (Campbell 1981, 134). However, resemblance as the feature through which tropes are collected together in classes, should not be considered only in its perceptual construal. In fact, we may say that all electrons have precisely similar charge tropes without taking this similarity as being grasped by an act of acquaintance. In this case, the similarity judgment is driven by causal considerations (see Simons 1994), so that tropes can be placed in the same resemblance class also when they have the same causal role, that is, if they have similar causes and similar effects.4 Ehring 1997 has maintained that the metaphysical task of solving the problem of causation is a major one for which tropes are invoked. He thinks that tropes can replace states of affairs, facts or events as causal relata for any single causal relation.5 So, properties-as-universals and tropes differ in the way in which they cope with causation. Properties figure as the vehicles for causal interactions between events. An event c is said to be a cause if and only if there is at least one property that determines the occurrence of another event e. The stone is the cause of the shattering of the window if and only if there is at least one property of it, its force or its shape or …, that determines the shattering; the mental event of deciding is the cause of the 4

One may wonder whether a resembling b in virtue of a’s causal roles resembling b’s causal roles could bring to some sort of regress. Moreover, if properties are placed into lawlike relations, these would connect sets of resembling relations, becoming second order relations. I leave these points aside. 5 On this issue see section 4 in Orilia 2008.

140 raising of the arm if and only if there is at least one property of it, being a desire to do such and such, that determines the raising. However, since no causal relation in the world is exactly determined solely by the properties called for in its description or explanation, we need to hedge such epistemic statements with provisos and caeteris paribus clauses. Tropes, on the other hand, bearing within the epistemic statements the specific conditions in which the causal relation took place, make the extra ingredient typical of the caeteris paribus conditions dispensable: it has been precisely this rock throwing that has determined this glass shattering. Another throwing would have resulted in a different shattering. This makes tropes subject to very thin and subtle individuation conditions, that is, their unrepeatableness and space-time location determines a singular causal relation. Now, how can tropes help to solve the problem of mental causation? 3.

The problem of mental causation

The causal efficacy of mental properties seems to be jeopardized by the acceptance of the principle of causal closure of the physical domain along with the denial of overdetermination. The principle says that if a physical event has a cause at t, it has a physical cause at t, while denying overdetermination amounts to rejecting the idea that events may systematically have two or more causes that are independently and not jointly sufficient. It can be easily seen why adopting the principle creates a tension for the causal efficacy of mental events and properties. If the raising of my arm is a physical event, as is reasonable to think, then it must have had a physical cause at some previous time t. Now, the mental event which is my decision to raise my arm either is a physical cause or it is not. If the first option is accepted, then the mental is efficacious just because it is physical; if it is the second one that is endorsed, then the mental is causally idle or the principle should be abandoned. If one wants to keep the principle, then the mental does not play any causal role by itself in either case, unless one considers my mental property as concurring with some physical property of my brain in causing my raising the arm. However, such concurrence overdetermines the cause of my raising the arm since it appeals to two different causes. The difficulty with this position is that it seems unnecessary to imagine such a metaphysical

141 richness in the physical domain. One possible retort is to observe that mental overdetermination is not concurrence (two singularly sufficient causes) but compresence (mental and physical properties working together), a retort somewhat aired by Tim Crane. According to him, if we believe that mental and physical states are linked by psychophysical laws—a claim which is defensible on independent grounds—then overdetermination would not be a coincidence: it would be a matter of natural law that the mental and the physical causes both bring about the effect.

Crane 1995, 19. How are these laws supposed to work? If mental and physical properties go hand in hand in causing, then each of them is singularly necessary while they both are jointly sufficient for the causing. This entails that my being in pain cause my taking the aspirin if and only if I have such and such brain state. So both properties are causally idle by themselves. Such an option, however, is open to two lines of reply. First, the physical cause in question is not the right one (and this is the reason why it results as not sufficient); second, how is the mental giving the “extra-bump” to the physical effect if not in physical terms? It seems then, that overdetermination should be excluded as a viable metaphysical option. A last option is taking the mental and the physical properties to be identical. However, since properties figure in laws as types, this solution would be tantamount to so called typeidentity theory of mind. A straightforward way to meet causal closure without overdetermination while saving mental causation is to identify the mental with the physical. However, the type-identity theory has been seriously attacked since the sixties.6 The token identity theory, on the contrary, when 6

Adopting a kimian metaphysics, the identity of mental and physical events follows from the identity of mental and physical properties. According to Kim 1976, in fact, events are structured entities comprising an object having a property at a time. So, once the identity-conditions for the object and the time are secured, a viable option, the identity theory reduces to property identity in that having pain now is identical to having brain state B now if and only if the property of having pain is type identical to the property of having state B. Davidson 1969, vice versa, takes event to be nonstructured entities. An event is mental if individuated through a description in which mental predicates occur; it is physical if physical predicates are used. On Kim’s construal of the individuation-conditions for event identity see Orilia 2008.

142 applied to properties, states that a mental properties occurring in spacetime is identical to a physical property, one that occurs at the same time and at the same location. Famously, Donald Davidson defended the token identity theory by holding three different principles: mental events causally interact with physical events; events related as cause and effect are covered by strict deterministic laws; there are no strict deterministic laws that cover mental events. He argues that the three principles, apparently at odds with each other, are consistent. He does this by stressing that, while causation is an extensional relation (any causal relation holds no matter how events are described) predictions and explanations, which are possible in virtue of laws, are intensional, thus crucially depending on the way in which events are described. It is in this respect that a sort of “ambiguity” with regard to the individuation of events finds its place: laws establish correlations between types of events; since mental types, given their holistic and normative character, are quite distinct from physical types, there is no way of establishing laws comprising them at this level. On the other hand, given the extensional nature of causal relations, there is no problem in identifying mental and physical events as tokens. Therefore, Davidson relies on the type-token “ambiguity” of events: mental events are type distinct from physical events, thus safeguarding their epistemological autonomy, but they are token identical to them, thus allowing their causal efficacy. Davidson’s solution has been charged with epiphenomenalism. The problem is that causation is guaranteed by subsumption under a law, but such a subsumption is possible only by considering physicalistic descriptions of events, and a description is physicalistic in that it takes into account just physical properties picked out by physical predicates. So, an event is causally efficacious only inasmuch as it is individuated through its physical properties. As Kim, in his 1989, 35, has argued, “on anomalous monism, events are causes and effects only as they instantiates physical laws, and this means that an event’s mental properties make no causal difference”. Davidson’s reply has centred on the irrelevance of descriptions as to causation:

143 if causal relations and causal powers inhere in particular events and objects, then the way those events and objects are described, and the properties we happen to employ to pick them out or characterize them, cannot affect what they cause.

Davidson 1993, 8. And also: For me is events that have causes and effects. Given the extensionalist view of causal relations, it makes literally no sense […] to speak of an event causing something as mental, or by virtue of its mental properties, or as described in a way or another.

Ibid, 13. However, it seems that the problem is still there. For, as Kim 1993 protests, the causal efficacy is captured by the instantiation of a law, and since mental predicates cannot be mentioned in strict causal laws, because strict laws describable by means of mentalist vocabulary are deemed not to exist, the presence of mental properties in a given event guarantees nothing more than their relevance. So, mental properties are at most relevant but not efficacious with respect to causal relations. It is important to consider the reason Davidson mentions as a source of confusion in his critics. Why have there been so many confusions and bad arguments in the discussion of AM, AM+P, and supervenience?7 The main source of confusion, I think, is in the fact that when it comes to events people find it hard to keep in mind the distinction between types and particulars.

Davidson 1993, 15.

7

By “AM” Davidson means anomalous monism, by “P” the premises that (i) mental events are causally related to physical events and (ii) that singular causal relations are backed by strict laws, which are the new formulations for the firsts two principles already mentioned.

144 Davidson thinks that the causal efficacy of properties manifests itself if it makes a causal difference in the powers of individual events, and that the idea of identifying it with the causal efficacy of physical properties is the result of confusing particulars with types of events. Consider the following example by Sosa 1984. Someone is killed by a loud shot. The loudness, however, is irrelevant to the death: had the shot be silent it would have killed the victim anyway. Mental events or properties, Sosa argues, are analogous to the loudness of the shot, hence they are causally inefficacious. Davidson points out that the counterfactual is ambiguous: even if the silent shot would have resulted in one death, […] It would not have been the same shot as the fatal shot, nor could the death it caused have been the same death. The ambiguity lies in the definite description ‘the shot’: if ‘the shot’ refers to the shot that would have been fired silently, then it is true that that shot might well have killed the victim. But if ‘the shot’ is supposed to refer to the original loud shot, the argument misfires, for the same shot cannot be both loud and silent. Loudness, like a mental property, is supervenient on basic physical properties, and so makes a difference to what an event that has it causes. Of course, both loud and silent (single) shots can cause a death; but not the same death.

Ibid., 17. In this passage Davidson is clearly adopting what we may call a trope view of events.8 A similar strategy has been pursued by David Robb 1997, along the lines suggested by Davidson9, and by Ehring 1999. Robb’s basic idea is, again, to trade in the type/token “ambiguity”, this time applied to properties, so as to have one reading of property at the general, type, level and another at the implementation, token, level. In Robb’s terms, the problem of mental causation is how to reconcile the following three principles: Distinctness: mental properties are not physical properties;10 Closure: every physical event/property has in its causal history only physical events/properties; 8

On this see Orilia 2008, section 3. But see John Heil 2003 as well. 10 On some interpretation of the principle the same holds for events. 9

145 Relevance: mental properties are (sometimes) causally relevant to physical properties. Robb’s idea is to construe “properties” as types in Distinctness, in order to differentiate the mental and the physical, and to read them as tropes in Closure and Relevance, warranting in this way their causal relevance without violating the principle of causal closure (Robb 1997, 187-8). Here is how Robb expresses the point: “Although second-order mental types and the first-order physical types that realize them are distinct, their tropes are the same” (ibid., 190). The difficulty in this approach, however, is the following. Consider a mental trope, call it m. In order for this trope to be a mental trope it has either to be the referent of a predicative expression concerning mental properties or to manifest a causal pattern of interactions typical of mental properties. In either case, if m is to be a mental trope, it must be subsumed within a “second-order mental type”. But this very trope has to be, at the same time, a physical trope. In order to determine what type of property a given trope is, one has to refer to the class or type it belongs to. So, if m has to be counted also as physical, then m belongs to a physical type too. The outcome is that m belongs both to a physical and to a mental type. If this is the case, either m is not a simple trope, because if types are different and the trope belongs to both then it is the concurrence of two tropes, or mental and physical tropes are type-identical, thus violating the distinctness condition because what makes mental and physical properties different is their belonging to distinct types. So, the crucial question here is whether a mental and a physical trope can be identical solely at the token, or realization, level. This problem can be formulated in a somewhat different way: can an event, taken as a trope (in line with my reading of Davidson), instantiate more than one type of property? Or, in Robb’s terms, can we have one and the same trope participating in two different resemblance classes? According to Cynthia and Graham MacDonald 1986, 147, having suspicions about such manoeuvres is a sort of dualist prejudice. Events can instantiate more than one property, because ‘property’ here is ambiguous between properties and their instances. … it may well be the case that one and the same event is both an instance of the property, being a desire for a drink, and an instance of another, physical property, say, being a brain event,

146 where being an instance of the former just is being an instance of the latter.

Ibid., 148. Shifting the focus on properties, Ehring 1999, 21, affirms: “The same property instance can be picked out by way of its membership in multiple classes, under different types”. However, in a recent and thoughtful paper, the MacDonalds reject the trope view, proposing in its place an exemplificationist view of events. In particular, they argue that the trope solution to the problem of mental causation is effective as long as one considers the causal efficacy of events. If the attention is, instead, on the causal relevance of mental properties, the trope route is not viable. In its place, it is preferable to consider properties as universals exemplified by events, these intended, à la Kim, as triples comprising properties had by object at a time. Moreover, the MacDonalds insist on the “ambiguity” of the concept of property, according to the trope view. Properties can be either instancing, hence individual tropes, or universal, so classes of perfectly resembling tropes. Now, how can we make sense of the idea that one and the same trope participates in two classes of resembling tropes, or that one event exemplifies more than one property? The MacDonalds answer, applicable to Robb’s theory of tropes, is given in the context of their exemplificationist account: Crucially, this amounts to the claim that there is just one exemplifying of two properties, one mental, and one physical, by an object at a time. That this is possible is apparent from determinate/determinable examples, such as being coloured and being red … Unlike the determinate/determinable property relation, the relation between mental and physical properties is not both metaphysical and conceptual.

C. MacDonald and G. MacDonald 2006, 561. They conclude that in their view the relation between mental and physical properties is just metaphysical.11 Leaving aside the view espoused by the

11

However, from a subsequent footnote we learn that “we do not think that the relation between mental and physical properties is as determinable/determinate relation, as

147 MacDonalds, I want to scrutinize the possibility of having such a rich view of events and the idea that tropes can belong to more than one class of resemblance. Let me start with the idea that one trope can belong to two or more resemblance classes in virtue of the determinable/determinate relation. Stephen Yablo 1992 has argued that mental and physical properties can be thus viewed, with the mental as a determinable and the physical as a determinate. Thus, my having pain is determinable inasmuch as it can be realized by this or that neural state, and my having those C-fibres firing is a maximally determinate property because it cannot be further determined. If the mental and the physical were in such a relation, there would not be any causal competition. However, it seems to me that mental and physical properties give rise to two independent series of determination relations. Consider my having pain now. Is it a maximally determinable state or is it a somewhat middle way? For instance, the determinable of my having pain now is my having an unpleasant sensory feeling now, which has as a determinable my having a conscious state now. Running on the other thread of the relation, my having pain now could have as a determinate my having pain in my left hand now, which in turn could have, as a maximally determinate trope, my having a throbbing, intense pain on the upper left corner of the palm of my left hand now. My having C-fibres firing now, vice versa, finds a determinable in my having a peripheral nervous activity now, which, in turn, has as a determinable my having some metabolic activity now. As with the case of pain, we can establish more determinate properties as well until a maximally determined one such as my having the release of this neurotransmitter from these cells now. These two chains of determinations, I think, cannot be identified or crossed without begging the question of the mind-body relation. If we did it, by taking for granted that the two chains are simply different ways of referring to the same phenomenon, we would “solve” the problem by an unexplained fiat. The other alternative is to say that the determination relation should not be conceived as a conceptual relation, but rather as an empirical one. However devising the way in which determination can be construed in non-conceptual terms is a task that has not been faced yet. An option would be that of construing this empirical relation in terms of constitution. standard cases of this relation involve conceptual entailment of the determinable from the determinate property” (ibid., 563, n. 43).

148 If this route is taken, we have to renounce to tropes’ being abstract in the sense previously exposed. In fact, we saw that abstractness was the result of space-time compresence of two type different properties. Red and round are type different and compresent resulting thereby abstract. The lump of clay and the statue are also type different and compresent, but there is a constitution relation, so the lump of clay is the concrete constituent of the statue. If the mental state is constituted by the physical state, the latter is not abstract anymore, hence not a trope anymore. Putting this point in my previous terminology, the determination relations are within trope dimensions, not across them, even if there are dimensions that are regularly coinstantiated. If my hearth rhythm increases whenever my kidney activity increases, and vice versa, nobody would consider the property of having an increasing kidney activity as a determinable of the property of having an increasing heart activity, to adapt a famous example by Quine. On the event side, the problem is not to have an event instantiating more than one property at a time, but the relation between the properties instantiated. For instance, consider a rake: it has a wooden brown handle and an iron black toothed bar. If I detach the handle from the toothed bar, this very event is also the separation of wood from iron, of a brown thing from a black thing, of a less flexible shape from a more flexible shape, and so on. I can even have multiple purposes and intentions in separating one part of the rake from the other. Such an event, though, would not constitute a metaphysical achievement or a cue on the relations between wood and iron, handle and toothed bar, brown and black. One may want to stick with the determinable/determinate relation, after rejecting my previous argument of the two chains of determination, by saying that it confuses the proper determinables for any given trope. If so, a second argument can be marshalled, which focuses on tropes’ simplicity as a feature that is not compatible with their belonging to two or more resemblance classes. Consider, for instance, a red trope: anything that is red is coloured as well. However, any trope is a maximally determinate entity: you cannot have entities in space and time that are “generic” in the sense of being coloured but not of some specific colour. This red is a maximally determinate shade of red. It is coloured as well, true, but such a determination relation is conceptual, while the relation between tropes does not seem to be of this kind, if tropes are maximally determinate entities. In fact, the compresence of two or more tropes is a matter of contingent and empirical fact. If being coloured is considered as a trope side by side with

149 red, then its compresence with red would be necessary and conceptual. But tropes were postulated to give us a metaphysically firmer grip on reality, and it seems to me that such a grip would be lost in case such purely conceptual tropes were admitted. Renouncing the idea that tropes are maximally determinate coincides with giving up one of the crucial features that distinguishes tropes from universals. For instance, taking red—as the generic determinable of scarlet, crimson, …—as a trope would make it wholly present in many places at the same time or in the same places at different times, raising again the “metaphysical scandal”, as Campbell considered this feature when applied to universals. At the same time, if red is not taken as a maximally determinate shade of colour, it becomes a categorical name for a certain range of light, the one that goes from, say, crimson to scarlet. So, it is a categorical mistake to take the generic red to be a trope as specific as scarlet, analogous to those mistakes described by Ryle 1949 in taking the name “University” to be referring to another and specific College building, one different from those visited so far. A final argument against the identification of physical and mental tropes takes these to be maximally determinate. The gist of the argument is that having two maximally determinate tropes does not entail having two simple tropes in the same way. A mental trope, being captured by an inner act of acquaintance, is definitively phenomenally simple; I grasp it. But it does not mean that it is structurally simple. My being in pain, now, could correspond to a very complex neuronal activity, which is structurally complex. As a matter of fact, this is what neuroscience tells us. C-fibre and A-δ fibres concur in giving us this sensation which is pain. We take it to be simple, but it is so only under a phenomenological perspective, not from a structural one. So, the two tropes cannot be identified, because one is not further decomposable, hence simple on both readings of simplicity, whereas the other is decomposable, and so is complex, even though they are both maximally determinate. Since simplicity is a crucial feature for tropes identification, the purported identity fails. 4.

A little coda

One may think that an inevitable consequence of the extensionality of the causal relation is that of having many tropes instantiated at once, a point raised by Davidson and taken up by Robb and the MacDonalds. Such a

150 view, moreover, would sidestep the so-called qua issue, that is, the vexed question of “is it in virtue of this or that aspect that the trope was causally relevant?” Both Davidson and Robb dismiss the question from the very beginning. Davidson has a purely extensionalist view of causation, and Robb seems to take the strictly singular individuation condition for tropes as the key to solve the issue: A causally relevant property F simply does not have various aspects such that one can legitimately ask whether some but not others are responsible for F’s being causally relevant.

Robb 1997, 191. Noordhof has attacked Robb on this ground. He says: Did the glass shatter as a result of the soprano’s singing a note in virtue of its pitch or its meaning? We want the answer that it is the pitch … how does the trope theorist get this answer? What stops someone from saying that the meaning of the note is causally relevant because the meaning trope is identical to the pitch trope?

Noordhof 1998, 225. The problem, according to Noordhof, lies in the identity conditions for tropes, which are at this point taken to involve a supervenience relation between the types to which the soprano note belongs. This is no solution, though, because we are back to the original problem of mental causation, thus pushed back to our starting point.12 Robb replies that the task of determining the individuation conditions for tropes is a red herring when it comes to establishing the identity of mental and physical tropes, because such an identity is secured by the fact that tropes reconcile the three principles of Distinctness, Closure and, in particular, Relevance.13 So, the task of detailing the individuating conditions for tropes can be pursued after trope monism has been granted

12

Sidney Shoemaker has raised a similar worry (2001, 433-4). Orilia (personal communication) has suggested that it should be the other way around: assuming trope identity allows one to salvage the three principles. However, I am interpreting what Robb has to say on this, more than affirming it myself.

13

151 (cf. Robb 1998, 94)14. However, it is far from clear whether trope monism has been so secured, for the trading in the property “ambiguity” is precisely what raises the problem. In fact, trope monism cannot be secured by endorsing Distinctness, because the very notion of property used in that principle serves the purpose of differentiating between mental and physical types of tropes. Thus, the only option left, the one compatible with Closure and Relevance, is space-time localization. However, the possibility of affirming the co-localization of two tropes is not enough for establishing their identity: the rotation of the Earth and its cooling down are two tropes occurring at the same time and in the same place, they coincide as to their four-dimensional world-line, but are two different tropes, individuated through two different causal powers, or aspects or at any rate the two tropes are simply prima facie quite different. We cannot say that they are one and the same simple trope, because they do not have the same causal powers, the same conceptual role or whatever preliminary individuation conditions for tropes you like. Co-localization is not enough for identity (cf. Casati and Varzi 1996). It is even possible to affirm, with Lowe 1994, 533, that “abstract objects, both universals and particulars, have timeless identity-conditions”, and are not spatial either. The same applies to mental and physical tropes: even if my belief that p and my brain state b were occurring in the very same portion of space-time, this would not secure their identity. This can be shown by reductio ad absurdum as follows. Take my supposedly enduring belief that I am Simone. I retained it since time t and up to time t΄. Suppose that the relevant neurons that were active at t when I held that belief have died and that I hold such a belief at t΄ in virtue of other neurons. Since nothing relevant is physically the same, it is not possible to argue that I retained the same belief, because all the spatio-temporal conditions have changed. Hence, persistent beliefs are not possible. If these are not possible, it is not even possible to change one’s own mind, this being the result of transforming one of one’s own enduring beliefs, but this is absurd. (I have used a line of reasoning quite familiar in the semantic debate concerning holism.) The upshot of this discussion, I think, is that the red herring lies in fixing the individuation condition for tropes prima facie in their spatio14

Moreover, Robb thinks that another advantage in introducing tropes instead of events as the properties of causation is that these are not the relata of causal relations (contrast with Ehring 1997 on this point), rather they are the properties that determine such relations.

152 temporal locations. Such condition is the result of trading in the property “ambiguity”, because the difference between mental and physical properties in Distinctness is established for types (or resemblance classes) while their identity is established in the trope reading of Closure and Relevance, where the only condition for setting the identity in such cases is spatio-temporal co-localization. If the co-localization is a red herring for the individuation conditions, what is left for individuation is the very property itself. This is not surprising after all: when two tropes are placed in the same resemblance class, or are judged to be precisely similar, they are so not by virtue of co-localization but simply by virtue of being the very properties that they are. Such an option, though, is not open to Robb and Davidson, because they want to maintain Distinctness. This unsurprising result, then, has serious consequences for the attempt to rescue a Davidsonian strategy in the mental causation debate. It should be noticed that not even an appeal to the modal status of tropes could be of some help. Just as the loud shot could not possibly be the silent shot, the same event could not be mental and physical by fiat. To say otherwise would be committing oneself to some petitio principii. So, if spatio-temporal co-localization is not enough to secure the identity needed to maintain Closure and Relevance in spite of Distinctness, what is left to this end? Nothing else, I think. Tropes, then, reveal themselves as useless for solving the problem of mental causation. One may wonder whether this conclusion is limited to the problem of mental causation or can be applied to causation in general. It seems to me that the difficult issue in the case of causation is raised by Relevance, where different levels of description are at stake. In this respect, tropes do not seem to provide a substantial help, being tailored to solve the metaphysical scandal of having entities, as universals could be taken to be, scattered in space and time. However, if this scandal is of some help in solving a conundrum, as causation is, I prefer to live with it rather than adopt entities with unstable identity(conditions). REFERENCES Armstrong, D. (1989). Universals. An Opinionated Introduction. Boulder: Westview Press. Casati, R. Varzi, A. (1996). The Structure of Spatial Localization. Philosophical Studies, 85, 205-39.

153 Campbell, K. (1981). Abstract Particulars. Midwest Studies in Philosophy. Reprinted in Properties, eds. D. H. Mellor and A. Olivier, Oxford: Oxford University Press, 1997, 125-39. Campbell, K. (1990). Abstract Particulars. Oxford: Blackwell. Crane, T. (1995). The Mental Causation Debate. Proceedings of the Aristotelian Society, 69, 211-36. Davidson, D. (1969). The Individuation of Events. Essays in Honour of Carl G. Hempel, ed. N. Rescher, Dordrecht: Reidel, 216-34. Reprinted in D. Davidson, 1980, 163-180. Davidson, D. (1970). Mental Events. Experience and Theory, eds. L. Foster and J. Swanson, London: Duckworth. Reprinted in Davidson, 1980, 207-25. Davidson, D. (1980). Essays on Actions and Events. Oxford: Clarendon Press. Davidson, D. (1993). Thinking Causes. Heil and Mele 1993, 3-17. Ehring, D. (1997). Causation and Persistence. Oxford: Oxford University Press. Ehring, D. (1999). Tropeless in Seattle: the Cure for Insomnia. Analysis, 59, 19-24. Heil, J. and Mele, A., eds. (1993). Mental Causation. Cambridge: Clarendon Press. Heil, J. (2003). The Nature of True Minds. Cambridge: Cambridge University Press. Hochberg, H. (2004). Relations, Properties and Particulars. Relations and Predicates, eds. H. Hochberg and K. Mulligan, Frankfurt: Ontos Verlag, 17-54. Kim, J. (1976). Events as Property Exemplification. Action Theory, eds. M. Brand and D. Walton, Dordrecht: Kluwer, 159-77. Kim, J. (1989). The Myth of Nonreductive Materialism. Proceedings and Addresses of the American Philosophical Association, 63, 31-47. Kim, J. (1993). Can Supervenience and “Non-Strict Laws” Save Anomalous Monism?. Heil and Mele 1993, 19-26. Lowe, E. J. (1994). Primitive Substances. Phenomenological Research, 54, 531-52.

Philosophy

and

154 Lowe, E. J. (2008). Tropes and Perception. This volume. MacDonald, C. and MacDonald, G. (1986). Mental Causes and the Explanation of Action. Philosophical Quarterly, 36, 145-158. MacDonald, C. and MacDonald, G. (2006). The Metaphysics of Mental Causation. The Journal of Philosophy, 103, 539-76. Maurin, A.-S. (2002). If Tropes. Dordrecht: Kluwer. Noordhof, P. (1998). Do Tropes Resolve the Problem of Mental Causation?. The Philosophical Quarterly, 48, 221-6. Orilia, F. (2008). Universals, Tropes and the Philosophy of Mind. This volume. Robb, D. (1997). The Properties of Mental Causation. The Philosophical Quarterly, 47, 178-94. Robb, D. (1998). Reply to Noordhof. The Philosophical Quarterly, 48, 904. Robb, D. (2005). Qualitative Unity and the Bundle Theory. The Monist, 88, 466-92. Ryle, G. (1949). The Concept of Mind. London: Hutchinson. Shoemaker, S. (2001). Realization and Mental Causation. Physicalism and Its Discontent, eds. C. Gillett and B. Loewer, Cambridge: Cambridge University Press. Reprinted in S. Shoemaker Identity, Cause and Mind, Oxford: Oxford University Press, 2003, 427-51. Simons, P. (1994). Particulars in Particular Clothing: Three Trope Theories of Substance. Philosophy and Phenomenological Research, 54, 553-75. Sosa, E. (1984). Mind-body Interaction and Supervenient Causation. Midwest Studies, 9, eds. P. French, T. Uehling H. and H. Wettstein, 27181. Stout, F. (1921). The Nature of Universals and Propositions. Oxford: Oxford University Press. Yablo, S. (1992). Mental Causation. Philosophical Review, 101, 245-80. Williams, D.C. (1953). The Elements of Being I & II. Review of Metaphysics, 7, 3-18; 71-92. First part reprinted in Properties, eds. D. H. Mellor and A. Olivier, Oxford: Oxford University Press, 1997, 112-24.

Zombies from Below DAVID ROBB Davidson College [email protected] ABSTRACT. A zombie is a creature just like a conscious being in certain respects, but wholly lacking in consciousness. In this paper, I look at zombies from the perspective of basic ontology (“from below”), taking as my starting point a trope ontology I have defended elsewhere. The consequences of this ontology for zombies are mixed. Viewed from below, one sort of zombie—the exact dispositional zombie—is impossible. A similar argument can be wielded against another sort—the exact physical zombie—but here supplementary principles are needed to get to the impossibility result. Finally, at least two sorts of zombie—the behavioural and functional zombies—escape these arguments from below.

1.

Introduction1

A zombie is a creature just like a conscious being in certain respects, but wholly lacking in consciousness. Zombies have played a central role in recent theorizing about consciousness.2 They often appear in thoughtexperiments designed to elicit anti-materialist intuitions. Consider, for example, the exact physical zombie. This is a being just like me in all physical respects—let me be the conscious being in question—but not conscious. If such a being is so much as possible, then, it would seem, any theory on which consciousness is reducible to the physical is false. After all, my zombie twin is not conscious, yet shares with me any physical feature a materialist may claim as a reductive base of consciousness.

1

This paper expands on the brief discussion of zombies in Heil and Robb 2003, 189. An earlier and shorter version was presented at the 2007 Central Division Meeting of the American Philosophical Association in reply to Kevin Sharpe’s “Tropes and the Zombie Argument”. I thank Sharpe for his paper and the subsequent discussion, which prompted me to write the present article. Thanks also to Francesco Orilia and Chris Shields for comments on an earlier draft. 2 The sort of consciousness in play here is phenomenal consciousness, that is, consciousness of the “what it’s like” variety: see Nagel 1974, Block 1995, Chalmers 1996. An overview of the zombie literature is Kirk 2006; see also the Journal of Consciousness Studies 2:4 (1995).

156 But is the zombie—whether it be an exact physical zombie or some other sort—even possible? If not, anti-materialist arguments making essential use of zombies would not get off the ground. These creatures have accordingly been scrutinized from several viewpoints to ascertain whether, and in what sense3, they are possible: (1) Conceivability: Are zombies conceivable, and if so, is conceivability a reliable guide to possibility? (2) Knowledge: Is the possibility of zombies compatible with what we claim to know about consciousness? In particular, if my zombie twin is possible, do I have any way of knowing about my own conscious experiences? (3) Meaning: What do zombies mean with their psychological vocabulary, with terms such as “understands” and “consciousness”? And do such semantic facts reveal an incoherence in the notion of a zombie? (4) Evolution: If zombies are possible, does that mean consciousness is “invisible” to natural selection, and would this preclude an evolutionary explanation of the emergence of consciousness? (5) Causation: Is the possibility of zombies compatible with the causal role that consciousness seems to play with respect to our behavioural and cognitive lives?4 Each of these lines of inquiry is promising, and a problem as difficult as the nature of consciousness deserves the resources of as many areas of philosophy and science as we can bring to bear on it. In this spirit, I propose in this paper to look at zombies from yet another perspective: ontology. I’ll call this the view “from below”, since it starts with the most basic features of our world. In particular, I’ll look at zombies from the perspective of a trope ontology I have defended, albeit piecemeal, in other places.5 In section 2, I’ll outline this ontology, focusing on those elements most relevant to the possibility of zombies. The consequences of this ontology for the zombie will be mixed. Viewed from below, one sort of zombie, the exact dispositional zombie, looks impossible (section 3). A similar argument can be used against the exact physical zombie, but here 3

Philosophers sometimes distinguish varieties or grades of possibility, such as nomological, metaphysical, and logical possibility. In the discussion of zombies to follow, I’ll always have in mind metaphysical possibility, which is perhaps more accurately called intrinsic possibility, or even just possibility simpliciter: cf. Van Inwagen 1998. 4 Examples from the literature include the following, though many of these could fall into more than one category, as there are important connections among the approaches. Conceivability: Levine 2001, ch. 2, Marcus 2004; Knowledge: Block 1980, Shoemaker 1981; Meaning: Moody 1994, Thomas 1998; Evolution: Flanagan and Polger 1995; Causation: Chalmers 1996, chs. 4-5, Bailey 2006. 5 Primarily in Heil and Robb 2003, but also in Robb 1997, 2001, 2005.

157 supplementary principles are needed to get to the impossibility result (section 4). Finally, at least two sorts of zombie, the behavioural and functional zombies, escape these arguments from below (section 5). 2.

Sketch of a Trope Ontology

I take as my starting point the ontology of properties presented in Heil and Robb 2003. Although we do not use the term “trope” in that paper,6 what we defend is recognizably a trope ontology, for it has the following core features: (1) Properties are what characterize objects; they are “ways objects are”. (2) Properties are in re, in the objects they characterize; they are not, say, predicates, concepts, or Platonic forms. (3) Properties are particulars; they are thus distinguished from types, whether types are predicates, universals, classes, or some other sort of entity. (4) Properties are sparse: there need not be a unique property, or even a unique kind of property, corresponding to every meaningful predicate. (5) Properties can resemble one another, either exactly or partially, and such resemblance grounds sameness of type; that is, when two objects belong to the same type, it is in virtue of their instantiating resembling properties. (6) Properties are not reducible to items in other ontological categories, such as objects or universals. Although trope theorists do not speak with one voice, these six features are common to many trope ontologies.7 More important for present purposes, however, are two further principles from the Heil-Robb paper: Identity: Every property is both qualitative and dispositional. Simplicity: The qualitativity and dispositionality of a property are not aspects or properties of it: they are rather the property itself, differently considered.8

6

We use “property”, as I will do for the most part in this paper, occasionally switching to “trope”. I consider the terms to be interchangeable. 7 Useful overviews of trope theory can be found in Armstrong 1989, ch. 6, Campbell 1981, 1990, and Maurin 2002. 8 These principles can be found in Heil and Robb 2003, 184-5. I’ve reworded both slightly and given them new names.

158 My primary aim here is not to argue for these principles—let alone to defend a trope ontology more generally—but rather to look at their consequences for the zombie. Nevertheless, a few remarks about each are in order, both to motivate the principles within the context of a trope ontology and to forestall a few potential objections to the discussion of zombies that follows. With Identity, Heil and I reject the traditional distinction between qualitative (or categorical) properties and dispositional properties. All properties are both. Trope theory may not make Identity inevitable, but it does make the principle more plausible than it would be otherwise. Start with the qualitativity of every property. It’s not easy to say what it is for a property to be qualitative, but when a property is physical, part of its being qualitative is its filling in space.9 Qualities “colour” our world in a way that bare (non-qualitative) dispositions would not, if there were such. Tropes, at least the physical ones, fit the bill: a trope is in the object that has it, and this at least involves being located where the object is. A physical trope that doesn’t fill in space is difficult to imagine. Granted, it may be that some tropes are not physical. But the most plausible candidates for such properties, if there are such, are phenomenal properties, and these are qualitative if any are. As for the dispositionality of every property, this too is made more plausible by the doctrine that tropes are in re characterizers of their objects. While predicates or abstract forms may not bestow causal powers on the objects that have them, a property that’s in its object, and indeed is just a way that object is, is almost certainly in the causal mix. Alexander’s Dictum—“To be real is to have causal powers”—may be controversial as usually formulated.10 But restricted to what’s in re, the dictum enjoys a much stronger intuitive appeal. Trope theory is friendly to Identity in another way as well: an ontology of tropes undermines two influential arguments against the principle. First, the sparseness of tropes blocks any simple semantic argument for distinguishing qualitative and dispositional properties: because there is no one-to-one correspondence between predicates and tropes, there’s no reason to think that the undeniable semantic differences between qualitative and dispositional predicates is mirrored by an ontological distinction between qualitative and dispositional properties. 9

Cf. Blackburn 1990, Martin 1997, 222. See Kim 1993, 348 ff.

10

159 There is no easy inference, for example, from the non-synonymy of “is made of salt” and “is water-soluble” to the conclusion that this particular cube’s being made of salt (that qualitative trope) isn’t identical to its being water-soluble (that dispositional trope). Second, and for similar reasons, the familiar argument from multiple realizability11 establishes at most a distinction between dispositional and qualitative types. Perhaps any dispositional type—universal, class, concept, or predicate—is realizable in a variety of distinct categorical types, thereby blocking Identity were it formulated as a thesis about types. But if Identity is, as intended, a thesis about properties (tropes), it’s unaffected by this argument.12 I should note, by the way, that even if every property is dispositional, this doesn’t mean that the powers a property bestows must be powers to affect physical objects. That is, it’s compatible with the dispositionality of a property that it be “epiphenomenal” with respect to a given family of properties, such as the physical properties. I mention this now to forestall a potential objection to what follows. One may worry that by insisting on the dispositionality of every property, and thus of phenomenal properties, I am begging an important question against those who think zombies are possible. For it is natural to think the possibility of zombies entails that consciousness is epiphenomenal with respect to the physical, and in particular, with respect to behaviour.13 But at least at this point in the discussion, such epiphenomenalism is an open question. Identity requires that all properties, and so all phenomenal properties, bestow causal powers on the objects that have them, but it does not require that these be powers to affect the physical. Turn now to Simplicity, so-called because it follows from the more general thesis that properties are “attributively simple”: properties do not themselves have other properties or attributes. This is not to say that properties are simple in every way. In particular, properties can be mereologically complex: they can have other properties as parts. For 11

See Prior et al. 1982. See Heil 1999. In Robb 1997, I make the similar point that the multiple realizability of the mental in the physical shows at most that mental types are not physical; cf. Kim 1993, ch. 16. 13 One route from the possibility of zombies to epiphenomenalism goes through the causal closure of the physical: see Stoljar 2001b and, for remarks on closure, Kim 2005, chs. 1-2. Chalmers (1996, 150ff.) explores but does not endorse the zombieepiphenomenalism connection. In any case, I suspect this connection is most often asserted by opponents of zombies: see Perry 2001, ch. 4, Bailey 2006. 12

160 example, the colour of a sphere has, as parts, the colour of its left half and the colour of its right half. The length of a stick has as parts the lengths of the stick’s two halves (and, at a lower mereological level, the lengths of the stick’s four quarters, and so on).14 I’ll return to such complexity later, but for now, the point is just that even if some properties are mereologically complex, all of them are attributively simple. Why should a trope theorist believe in Simplicity? For one thing, because tropes are sparse, there’s no straightforward semantic argument for thinking that properties themselves have properties. It’s certainly true that various predicates apply to any given property. A given red trope, for example, falls under the predicates “is a property”, “is bright”, “is a colour”, “is qualitative”, “is dispositional”, and so on. But given that there’s no one-to-one match between predicates and properties, there’s no easy route from these predications to higher-order tropes. The truthmaker for “This property is bright” and the like may be just the property itself. So trope theory at least weakens one motivation for believing in properties of properties. And indeed, once an ontology of tropes is in place, there’s little motivation from any other source to believe in higher-order properties. Consider, for example, the resemblance among tropes, something that initially may appear to call for higher-order properties. Two colours, for example, may resemble each other, perhaps exactly. But what progress is made by explaining such resemblance in terms of the higher-order tropes they instantiate? Presumably the higher-order tropes can’t explain the resemblance of the colours unless the higher-order tropes themselves resemble, and we are at the start of a vicious regress: the fact of resemblance is passed to (explained in terms of) resemblance at a higher level, and so on. To stop the regress, we must eventually reach tropes at some level that resemble just in virtue of themselves. But if attributively simple tropes are unavoidable, there seems to be little reason to deny this status to the first-order tropes and be done with it. In this case, at the first order, resemblance among properties is primitive: it cannot be explained in more fundamental terms.15 14

Cf. Armstrong’s (1978, 68-71) structural universals. For more on the mereological complexity of some properties, see Robb 2005. 15 When the resembling properties are mereologically complex, their resemblance can be explained in terms of the resemblance among their parts. But this buck cannot be passed indefinitely: at some point, resemblance will have to “ground out” in the primitive resemblance among mereologically simple properties.

161 Simplicity is just a particular instance of this more general thesis. The qualitativity and dispositionality of a trope aren’t properties of the trope. If they were, then given Identity, they themselves would need to be qualitative and dispositional, and a regress threatens again. But even setting aside Identity, there is little explanatory value a trope theorist has to gain from saying that qualitativity and dispositionality are higher-order tropes. Consider, for example, the qualitativity Q of a scarlet trope S, and suppose Q were a higher-order trope of S. If Q has the same qualitative nature as S, then clearly no progress has been made in explaining S’s qualitativity. But if Q has some other qualitative nature or, worse, no qualitative nature at all, how could it have anything to do with S’s qualitativity? Similar remarks could be made about S’s dispositionality. Much more needs to be said about both principles,16 but with this sketch of the ontology in place, I’ll now move on to my main topic in this paper. What do zombies look like given the ontology? Viewed from below, from the perspective of this trope ontology, do zombies look possible? 3.

The Exact Dispositional Zombie

The answer to this question depends on what a zombie is stipulated to be. Earlier, I described a zombie as a creature just like a conscious being in certain respects, but wholly lacking in consciousness. But the “respects” in question can vary.17 For example, are they limited to intrinsic features, or do they also include extrinsic or “wide” features as well? Are they dispositional, physical, or both? And is the resemblance between my twin and me supposed to be exact or merely partial, limited to certain salient respects? The answers to these questions generate several sorts of zombie, far more than I can look at here. To simplify matters, I will take the respects of resemblance always to be intrinsic.18 And I will begin with just one sort of zombie, a non-conscious being who exactly resembles me 16

For a partial defence, see Heil and Robb 2003, and for more a more detailed defence, Heil 2003, 2005. See also Martin 1997 and Martin’s contribution to Armstrong et al. 1996. 17 Cf. Güzeldere 1995. 18 This is a significant simplification on some views. For example, Dretske’s (1995) brand of representationalism about consciousness allows a non-conscious creature just like me in every intrinsic physical respect, but it prohibits a non-conscious creature just like me in every intrinsic and extrinsic (in particular, historical) respect.

162 dispositionally: this is the exact dispositional zombie. Starting with a description of my twin as dispositionally just like me, I will try to show, using the ontology of properties just sketched, that such a being must be conscious, and so not a zombie. I’ll present the argument as a series of numbered steps. (1) My twin and I fall under all of the same dispositional types. This much can be stipulated, so long as we are careful to remain neutral on the ontology of types. (So, for example, (1) might be read as the relatively innocent claim that my twin and I fall under all of the same dispositional predicates.) Since the resemblance here is exact, my twin behaves as I do, and in addition shares my non-behavioural dispositions; for example, we both tend to turn red when exposed to the sun for too long. And note that we must also share our unobservable dispositions, down to the finest detail. If I am, say, disposed to undergo a very slight rearrangement of unobservable parts given the small gravitational attraction of a passing car, then my twin is disposed to undergo an exactly similar rearrangement in the same circumstances. And so on.19 (2) Therefore, for every dispositional property I have, my twin has one dispositionally just like it. This inference from (1) to a “trope isomorphism” between me and my twin would be unwarranted if the two of us were merely partially similar dispositionally—that is, if we fell under merely some of the same dispositional types. (I will return to this point in the final section.) However, if we are dispositionally indiscernible, if we fall under all of the same dispositional types, then trope isomorphism is inevitable, for anything less than such an isomorphism would have to be reflected in a dispositional difference somewhere, even if at some very fine level of detail.

19

I here consider only forward-looking dispositional types, but my twin and I also fall under the same backward-looking dispositional types. On this distinction, see Shoemaker 2003, 412.

163 (3) My phenomenal properties—that is, those properties in virtue of which I am conscious—are dispositional. This follows immediately from Identity: Every property is dispositional; a fortiori, my phenomenal properties are. I assume here that there are phenomenal properties, a controversial assumption even within the confines of the ontology. I cannot take on eliminativism about phenomenal properties here, except to say that the primary evidence for their existence is that we are—or at least seem to be—directly acquainted with them in conscious experience. There are some who would dispute the evidential value of such seeming,20 but I cannot do justice to their objections here, and so will just take for granted the existence of phenomenal properties. (4) Therefore, my twin has properties dispositionally just like my phenomenal properties. This follows from (2) and (3). (5) Properties that are dispositionally indiscernible are qualitatively indiscernible. That is, if two properties exactly resemble dispositionally, then they exactly resemble qualitatively. This crucial step is supported by Simplicity, on which the dispositional resemblance of two properties is not a matter of their instantiating resembling properties, i.e. higher-order tropes. In general, when properties resemble, it’s not in virtue of their properties (they have none), but just in virtue of themselves. So two properties cannot be exactly alike in one “way” but not exactly alike in another. There just are no ways in a trope’s nature to allow for such variation. And so if two properties are dispositionally just the same, they are qualitatively just the same as well. Here’s another way to put the point: If two properties “share” their dispositional natures (i.e., exactly resemble each other dispositionally) then they must share their qualitative natures, for on Simplicity, a property’s dispositional nature is its qualitative nature, each being just the property itself.

20

See e.g. Rey 1997, 305-7.

164 This point remains whether the resembling properties are mereologically simple or complex, though the reasoning is a bit more complicated in the latter case. Let T1 and T2 be dispositionally indiscernible properties, and suppose these tropes are mereologically complex, so that their dispositional resemblance is grounded in the dispositional resemblance of their parts. Exact dispositional resemblance here requires exact dispositional resemblance at every mereological level, so for every part of T1, there is a part of T2 dispositionally indiscernible from its counterpart in T1. The reasoning in the previous paragraph will then show that these parts must be qualitatively just the same. (If these parts themselves have parts, then repeat the reasoning until mereologically simple properties are reached.) And so however the qualitative parts of T1 combine to yield the qualitative nature of T1, the qualitatively identical parts of T2 will yield a qualitatively identical property.21 (6) Therefore, my twin has properties qualitatively just like my phenomenal properties. This follows from (4) and (5). (7) Therefore, my twin has phenomenal properties. This follows from (6): Phenomenal properties are paradigmatically qualitative, and so anything qualitatively just like a phenomenal trope is itself phenomenal. There are, granted, some views on which the phenomenal nature of a property is not intrinsic to it, so that a property can be qualitatively just like a phenomenal property, yet fail to be phenomenal. This is the case on, for example, “higher-order thought” theories of consciousness.22 On such views, roughly, for a property to be phenomenal, it must be the subject of a thought about that very property. If what makes a given phenomenal property of mine phenomenal is that it’s the subject of such a thought, then it’s possible that there be a property qualitatively just like my own, yet not phenomenal, since this qualitative duplicate may not itself be the subject of higher-order thinking. But whatever the merits of 21

When T1 is a phenomenal property, these qualitative parts may be what Chalmers 1996, 127 calls “protophenomenal properties”, properties that, while not themselves phenomenal, combine to produce phenomenal properties. 22 See e.g. Rosenthal 1997.

165 the higher-order-thought theory of consciousness are, I don’t think it’s going to have much effect on the present argument. Any difference in higher-order thinking between me and my twin would have to be a dispositional difference. After all, it would consist in a difference of properties instantiated, all of which are, by Identity, dispositional. But my twin and I are dispositionally indiscernible, so if my phenomenal properties are the subject of higher-order thoughts, so are those of my twin. The upshot is the same: since my twin has properties qualitatively just like my phenomenal tropes and (let us suppose) all are the subject of higherorder thinking, my twin’s tropes must be phenomenal. (8) Therefore, my twin is conscious (and so not a zombie). This is a consequence of (7), for to be conscious just is to instantiate phenomenal properties. (Note that a conclusion even stronger than (8) is warranted by the foregoing: Not only is my twin conscious, his phenomenal properties exactly match my own. That is, he is not even phenomenally “inverted” with respect to me. What’s said about zombies in this paper, then, is also relevant to the possibility of phenomenal inversion, but I will not explore that topic here.) 4.

The Exact Physical Zombie

Even if the above argument is sound, it rules out only one sort of zombie. Consider now a creature physically just like me, but, as before, wholly lacking in consciousness. In this case, my zombie twin and I fall under the same (intrinsic) physical types: this is the exact physical zombie. Such a zombie is arguably the sort most often discussed in the contemporary literature. The exact physical zombie and the exact dispositional zombie are easily conflated, but strictly speaking they are distinct. There is nothing in the notion of a disposition—whether it be type or trope—requiring that the physical and the dispositional coincide. Note, for example, that it’s compatible with the conclusions of the previous section that all of my phenomenal properties, in spite of being dispositional, are non-physical. In this case, the exact dispositional zombie would remain impossible, but the exact physical zombie would be possible. Ruling out the latter requires more work than what’s been done so far.

166 What, then, is needed to supplement the ontology in order to rule out the exact physical zombie? The thesis that every phenomenal property is physical would help. But while I do endorse such a psychophysical “trope monism”,23 a direct appeal to it is not needed here, for in fact something weaker will do. Note first that each of my physical properties is dispositional: this is a consequence of Identity. Now suppose that each dispositional property is paired with a qualitative “ground”. On Identity, a dispositional property is its ground, but set that aside for a moment and consider just the following: Grounding: My phenomenal properties are among the qualities grounding my physical dispositions. This principle falls short of identifying physical properties with their phenomenal grounds. How short it falls depends on the nature of the grounding relation, here left unspecified; perhaps it is a nomological relation, or, more broadly, a species of supervenience.24 In any case, however the grounding relation is spelled out, its initial characterization will need to be strong enough so that once combined with the trope ontology—and in particular, with Identity—Grounding collapses a physical disposition into its qualitative ground. This forms the basis of an argument against the exact physical zombie, as follows. The argument can begin along the lines of the previous section: My twin, by stipulation, falls under all of the same physical types I do, and since the resemblance is exact down to the finest detail, an inference to a trope isomorphism is permitted as before. In this case, however, exact physical resemblance is in play: for every physical property I have, my twin has one physically just like it, a trope of the same physical type. Now bring in Grounding. My phenomenal properties ground some of my physical dispositions, and Identity collapses the qualitative “ground” of a dispositional property into the property itself. So with these principles we 23

Robb 1997, Heil and Robb 2003. Whether this sort of trope monism should count as physicalism is a question I will not take on here. Note in any case that even if trope monism were assumed, ruling out the exact physical zombie would not be trivial, for we would still need (P1) or (P2) (see below). 24 Such relations are usually defined in terms of types (or families of types), but they have trope versions as well. On the varieties of supervenience, see Kim 1993, ch. 4. Chalmers 1996, ch. 4, endorses a nomological version of Grounding, albeit with respect to physical and phenomenal types, not tropes.

167 can conclude that my phenomenal properties are physical. Since my twin physically duplicates all of my physical properties, and since these now include my phenomenal properties, it follows that for every phenomenal property I have, my twin has one physically just like it. But are these properties in my twin qualitatively just like my phenomenal properties, and thus phenomenal? We are still not to this conclusion, for it could turn out that physical indiscernibility is—like the behavioural and functional indiscernibility to be considered in the next section—a case of partial resemblance. The notion of a physical type (or for that matter, a physical property) used so far is too under-described to rule this out. There are at least two ways of “thickening” the notion that would complete the argument. Both attempt to fill out part of what it is to be a physical type, though neither attempts a definition.25 The first is: (P1) Properties of the same physical type exactly resemble. This is the trope theorist’s version of the claim that physical types are (or stand in for) genuine “universals”. Given (P1), those properties in my twin that are of the same physical type as my phenomenal properties will exactly resemble them. By Identity, exact resemblance must include qualitative resemblance, so my twin’s tropes are phenomenal as well. This route to the conclusion, it should be noted, does not depend on Simplicity, since Identity, Grounding, and (P1) suffice. There’s a different completion of the argument, however, that is structurally more like the argument against the exact dispositional zombie, for it ultimately depends on Simplicity. This appeals to a thesis slightly weaker than (P1): (P2) Properties of dispositionally.

the

same

physical

type

exactly

resemble

As I mentioned earlier, the physical and the dispositional needn’t coincide. After all, if there are non-physical properties, they are, by Identity, dispositional. But (P2) says merely that exact dispositional resemblance is part of what it is to be of the same physical type. Combining this principle with Simplicity will yield the conclusion. Since the relevant properties in 25

For a useful recent discussion of the nature of the physical, see Stoljar 2001a, and for scepticism about finding an adequate definition, Daly 1998.

168 my twin are of the same physical type as my phenomenal properties, they are, by (P2), dispositionally indiscernible. But by Simplicity, as in the previous section, exact dispositional resemblance entails exact qualitative resemblance. My twin, that is, has phenomenal properties. Clearly the route to impossibility here is less direct than it was in the previous section, for it goes through Grounding and either (P1) or (P2). And the argument is distinctive in another way: Grounding is a contingent thesis. After all, it would be false if Cartesian dualism were true, and even materialists are usually willing to concede that such dualism is at least possible. So even if the ontology is necessarily true, it cannot be used in all possible circumstances to rule out the exact physical zombie. And this is as it should be, for if Cartesian dualism were true, the exact physical zombie clearly would be possible. So the conclusion that exact physical zombies are “impossible” must be limited to those zombies duplicating conscious beings in whom Grounding is true.26 By contrast, the last section’s argument against the exact dispositional zombie is not similarly limited, for it appeals to Identity and Simplicity without needing Grounding as a supplement, and the former two principles, if true at all, are necessarily true. 5.

Other Zombies and Partial Similarity

The arguments of the previous two sections target two sorts of zombie, but there are other sorts whose possibility is compatible with the ontology, even when supplemented by some suitably modified version of Grounding. Consider first the behavioural zombie. This is a being whose behavioural dispositions are just like mine, but who is, as before, wholly lacking in consciousness. If behaviour is here restricted to overt bodily behaviour, then my twin is partially similar to me dispositionally.27 That is, 26

A related point, presented primarily in semantic terms, is in Braddon-Mitchell 2003. I’ll take partial similarity here to be mere partial similarity, similarity falling short of indiscernibility. And I’ll think of a behavioural zombie as a mere behavioural zombie. That is, I will not count an exact dispositional zombie as a behavioural zombie, even though the former shares all of my behavioural dispositions. I don’t think much turns on this decision. If instead the exact dispositional zombie is included as a version of the behavioural zombie, the main point can just be rephrased: In the ontology, there’s nothing in the notion of a behavioural zombie as such requiring that it be conscious, even if some “overqualified” versions of it—namely, the exact dispositional 27

169 my twin and I fall under some of the same dispositional types, namely the behavioural ones, but not all of them. And because the dispositional resemblance is partial, an argument of the form used against the exact dispositional zombie will be of no use here. Such an argument would, in fact, be blocked at the first inference. My twin and I fall under all of the same behavioural-dispositional types, but the trope isomorphism of section 3 won’t follow from this stipulation. Indeed, one can’t even infer here that my twin exactly duplicates my behavioural-dispositional properties. Consider, for example, the property I have in virtue of which I tend to wince when my skin is burned. This is, no doubt, a mereologically complex property, composed all of sorts of properties in my nervous system and elsewhere, but it is a property all the same. Now since my behavioural twin also tends to wince when his skin is burned, he has a property that, to this extent, is dispositionally similar to my own “wincing” property. But this is just a case of partial similarity. Properties often bestow some of the same powers without being dispositionally indiscernible. The molecular structures (complex properties) of two objects may dispose each to dissolve in water, yet may make one digestible to humans and the other not. Examples could be multiplied.28 Note, then, that even given some behavioural version of Grounding, we still will not have the resources to show that my behaviouraldispositional twin must be conscious. Suppose, that is, my phenomenal qualities are among those grounding my behavioural dispositions. Even so, since my twin’s behavioural-dispositional properties are partially like my own dispositionally, they will be partially like my own qualitatively. And the qualitative differences between my twin and me may be exactly what makes my twin phenomenally “dark” inside. The upshot is that for all the ontology tells us, even when supplemented by an appropriate Grounding thesis, the behavioural zombie is possible.29

versions—must be conscious. (Similar remarks apply to the functional zombie considered next.) 28 For further discussion of partial similarity, see Heil 2003, ch. 14. 29 (P1) or (P2) would be of use here if it were stipulated that my twin’s behaviouraldispositional properties are of the same physical type as my own. But then my twin would be a different sort of zombie: he includes more in his description than the standard behavioural zombie.

170 Similar points show that the functional zombie survives the previous arguments.30 A functional zombie falls under all of the same psychologically relevant dispositional types that I do. That is, my functional twin and I are dispositionally just the same in any way that would be of interest to the science of psychology, whether it be current psychology or some idealized version. Viewed from below, such zombies do not look impossible. (I am here neutral on whether they are in fact possible.) Like my behavioural twin, my functional twin is partially similar to me dispositionally. And so he will be partially similar to me qualitatively. As a consequence, the ontology leaves open whether my twin has phenomenal properties. And this remains so even given the appropriate form of Grounding. That is, even if my phenomenal qualities are among those grounding my psychological dispositions, my functional twin won’t have properties duplicating these qualities. Partial qualitative resemblance is all he enjoys, since only partial dispositional resemblance is stipulated to hold. And for all the trope ontology tells us, phenomenally “dark” qualities can resemble my phenomenal properties to the appropriate degree. My functional twin, that is, need not be conscious. It may be wondered at this point, however, whether the ontology really does warrant treating the behavioural and functional zombies so differently from their more exact cousins. The exact dispositional zombie, for example, is supposed to look impossible from below due to (among other things) Simplicity. In particular, I said that a property’s dispositional nature is not an aspect of the property, but just the property itself. There’s thus no room for two properties to be just alike dispositionally without being just alike qualitatively. But then how can two properties resemble behaviourally or psychologically without being exactly alike, both qualitatively and dispositionally? After all, the objection goes, by Simplicity, the behavioural-dispositional or psychological-dispositional “nature” of a property is just the property itself. So bestowing the same behavioural or psychological powers should result in dispositional (and thus qualitative) indiscernibility. Bring in the appropriate forms of Grounding, and the behavioural and psychological zombies should, like their more exact cousins, look impossible. The worry here, in short, is that Simplicity takes what initially looked like partial resemblance and collapses it into exact resemblance—rendering all of these zombies impossible. 30

Cf. Heil 2003, 247-8.

171 But Simplicity is in fact tolerant of partial similarity. This is clearest if the resembling properties are mereologically complex: partial similarity could just amount to similarity of some (but not all) parts. And in particular, the partial dispositional similarity of two complex properties could just amount to the dispositional similarity of some of their parts. This is compatible with Simplicity: no appeal to dispositionality as a “higherorder trope” is needed. Simplicity also permits the partial similarity of mereologically simple properties, though in this case an explanation is harder to come by, since, as I mentioned earlier, the similarity of simple properties is primitive. Still, I can say this much: Since partial resemblance among simple properties cannot be analyzed, it cannot (a fortiori) be analyzed as exact resemblance in some particular way or other. When simple properties partially resemble, then, there’s no sense in which they “share” a nature: they just partially resemble, and that’s the end of it. This is why Simplicity doesn’t threaten to collapse all resemblance into exact resemblance, and it’s why in the end the ontology permits the behavioural and functional zombies. REFERENCES Armstrong, D. M. (1978). A Theory of Universals: Universals and Scientific Realism, Volume II. Cambridge: Cambridge University Press. Armstrong, D. M. (1989). Universals: An Opinionated Introduction. Boulder: Westview Press. Armstrong, D. M., Martin, C. B. and Place, U. T. (1996). Dispositions: A Debate, T. Crane ed., London: Routledge. Bailey, A. (2006). Zombies, Epiphenomenalism, and Physicalist Theories of Consciousness. Canadian Journal of Philosophy, 36, 481-510. Blackburn, S. (1990). Filling In Space. Analysis, 50, 62-5. Block, N. (1980). Are Absent Qualia Impossible?. The Philosophical Review, 89, 257-74. Block, N. (1995). On a Confusion About a Function of Consciousness. Behavioral and Brain Sciences, 18, 227-87. Braddon-Mitchell, D. (2003). Qualia and Analytical Conditionals. Journal of Philosophy, 100, 111-35. Campbell, K. (1981). The Metaphysic of Abstract Particulars. Midwest

172 Studies in Philosophy, 6, 477-88. Campbell, K. (1990). Abstract Particulars. Oxford: Blackwell. Chalmers, D. (1996). The Conscious Mind. Oxford: Oxford University Press. Daly, C. (1998). What Are Physical Properties?. Pacific Philosophical Quarterly, 79, 196-217. Dretske, F. (1995). Naturalizing the Mind. Cambridge, MA: MIT Press. Flanagan, O. and Polger, T. (1995). Zombies and the Function of Consciousness. Journal of Consciousness Studies, 2, 313-21. Güzeldere, G. (1995). Varieties of Zombiehood. Journal of Consciousness Studies, 2, 326-33. Heil, J. (1999). Multiple Realizability. American Philosophical Quarterly, 36, 189-208. Heil, J. (2003). From an Ontological Point of View. Oxford: Clarendon Press. Heil, J. (2005). Dispositions. Synthese, 144, 343-56. Heil, J. and Robb, D. (2003). Mental Properties. American Philosophical Quarterly, 40, 175-96. Kim, J. (1993). Supervenience and Mind. Cambridge: Cambridge University Press. Kim, J. (2005). Physicalism, or Something Near Enough. Princeton: Princeton University Press. Kirk, R. (2006). Zombies. The Stanford Encyclopedia of Philosophy, ed., E. Zalta, URL: . Levine, J. (2001). Purple Haze: The Puzzle of Consciousness. Oxford: Oxford University Press. Marcus, E. (2004). Why Zombies Are Inconceivable. Australasian Journal of Philosophy, 82, 477-90. Martin, C. B. (1997). On the Need for Properties: The Road to Pythagoreanism and Back. Synthese, 112, 193-231. Maurin, A.-S. (2002). If Tropes. Dordrecht: Kluwer. Moody, T. (1994). Conversations With Zombies. Journal of Consciousness

173 Studies, 1, 196-200. Nagel, T. (1974). What Is It Like to Be a Bat? The Philosophical Review, 83, 435-50. Perry, J. (2001). Knowledge, Possibility, and Consciousness. Cambridge, MA: MIT Press. Prior, E., Pargetter, R. and Jackson, F. (1982). Three Theses about Dispositions. American Philosophical Quarterly, 19, 251-7. Rey, G. (1997). Contemporary Philosophy of Mind. Oxford: Blackwell. Robb, D. (1997). The Properties of Mental Causation. The Philosophical Quarterly, 47, 178-94. Robb, D. (2001). Reply to Noordhof on Mental Causation. The Philosophical Quarterly, 51, 90-4. Robb, D. (2005). Qualitative Unity and the Bundle Theory. The Monist, 88, 466-92. Rosenthal, D. M. (1997). A Theory of Consciousness. The Nature of Consciousness, eds. N. Block et al., Cambridge, MA: MIT Press, 72953. Shoemaker, S. (1981). Absent Qualia are Impossible—a Reply to Block. The Philosophical Review, 4, 581-99. Shoemaker, S. (2003). Identity, Cause, and Mind, Expanded Edition. Oxford: Clarendon Press. Stoljar, D. (2001a). Two Conceptions of the Physical. Philosophy and Phenomenal Research, 62, 253-81. Stoljar, D. (2001b). The Conceivability Argument and Two Conceptions of the Physical. Philosophical Perspectives, 15, 393-413. Thomas, N. (1998). Zombie Killer. Toward a Science of Consciousness II, eds. S. Hameroff et al., Cambridge, MA: MIT Press, 171-7. Van Inwagen, P. (1998). Modal Epistemology. Philosophical Studies, 92, 67-84.

Tropes and Perception E. JONATHAN LOWE Durham University [email protected] Abstract. The place of tropes in a neo-Aristotelian four-category ontology is explained and an acceptance of them is defended against some recent objections, by Jerrold Levinson, to the very idea of “particularized properties”. Following this, an argument from perception for the existence of tropes is developed and defended against some criticisms that Levinson has recently directed at an earlier version of it.

1.

Introduction

My objective in this paper is twofold. First, I want to explain the place of tropes—or, as I prefer to call them, modes—in a neo-Aristotelian fourcategory ontology, defending this conception against certain objections raised recently by Jerrold Levinson to the very idea of “particularized properties”. Second, I want to present in greater detail than I have done elsewhere an argument from perception for the existence of tropes, defending this too against Levinson’s recent criticisms. 2.

The place of tropes in a neo-Aristotelian four-category ontology

Amongst contemporary metaphysicians who believe in the existence of properties, there is an ongoing dispute as to whether properties should be regarded as universals or particulars. Those who favour the latter position commonly call themselves “trope theorists”, exploiting for this purpose D. C. Williams’ far from perspicuous use of the word “trope” to denote a property conceived as a particular. My own position, I should at once declare, is that I do not think that we have to decide between regarding properties as universals and regarding them as particulars, because I think that we ought to include both categories of entity as fundamental ones in our ontology. In deference to an older tradition, I use the term “attribute” to denote properties conceived as universals and the term “mode” to denote properties conceived as particulars. I maintain that modes are simply the particular instances of attributes, in exactly the same way that individual

176 objects or “substances” are the particular instances of substantial kinds or species—the latter being what Aristotle, in the Categories, called “secondary substances”. Indeed, my more general aim is to resuscitate, or at least reinvigorate, the four-category ontology that Aristotle adumbrates in Chapter 2 of that work.1 This system, as characterized by Aristotle himself, is perspicuously represented by a version of the so-called “Ontological Square”: see figure 1 below.2 Substantial Kinds

Individual Substances

Attributes

Said of but not in a subject

Both said of and in a subject

Neither said of nor in a subject

Not said of but in a subject

Modes

Figure 1: The Aristotelian Ontological Square

As will be seen from this diagram, Aristotle considers that substantial kinds or species are “said of but not in a subject”, attributes are “both said of and in a subject”, modes are “not said of but in a subject”, and individual substances are “neither said of nor in a subject”. I confess that I am not myself entirely happy with this way of explicating the relevant differences between entities belonging to the four different categories, partly because “said of” ostensibly expresses a linguistic rather than a metaphysical relation and partly because the meaning of “in a subject” is somewhat obscure, being suggestive of a spatial relation which seems inappropriate in at least some cases. This, however, is not the place for me to attempt to engage in Aristotelian exegesis—a task for which I am, in any case, inadequately 1

See my 2006. I refer the reader to that book for more details about the version of the four-category ontology that I wish to defend. 2 Compare Angelelli 1967, 12-15.

177 qualified. Let me instead, then, simply present my own preferred version of the Ontological Square, in the form of figure 2 below. Kinds

instantiated by

Objects

characterized by

exemplified by

characterized by

Attributes

instantiated by

Modes

Figure 2: The Ontological Square, Version II

It will be observed that I have abbreviated “substantial kinds” to “kinds” and replaced the somewhat archaic “individual substances” by “objects”. These are merely terminological niceties and nothing much hinges on the choice of labels for the four corners of the Square. What is more significant is that the key relationships between entities belonging to the different categories are differently expressed in my version of the Square—and are expressed there purely in terms of two metaphysical relations, instantiation and characterization. Kinds are characterized by attributes and instantiated by objects, attributes characterize kinds and are instantiated by modes, modes characterize objects and instantiate attributes, and objects are characterized by modes and instantiate kinds. It will be noticed that my version of the Square also includes a “diagonal” relationship between objects and attributes: the former, I say, exemplify the latter. However, I do not regard exemplification as being a fundamental or primitive metaphysical relation, like instantiation or characterization, since I regard it as coming in two different varieties—“dispositional” and “occurrent”— each of which is a different “resultant” of instantiation and characterization, the difference consisting in the order in which it is “composed” out of these two relations. To be more explicit: an object O exemplifies an attribute A dispositionally when O instantiates some kind, K, that is characterized by A; and an object O exemplifies an attribute A occurrently when O is characterized by some mode, M, that instantiates A. This distinction can be illustrated graphically by yet another version of the Ontological Square, as shown in figure 3 below. The two types of exemplification are there represented as two different “routes” that can be taken around the Square from the bottom left-hand corner to the top right-

178 hand corner. I shall say more about my conception of the dispositional/occurrent distinction shortly. Kinds

Objects occurrent exemplification

dispositional exemplification Attributes

Modes

Figure 3: The Ontological Square, Version III

It is not my objective in this paper to attempt to explain and justify my own version of the four-category ontology in all its aspects, as I have tried to do that elsewhere. Rather, I want to focus on just one reason for believing that we must include modes or tropes in our ontology, that is, properties conceived as particulars. This is that we cannot, in my view, adequately account for the nature of our perception of objects without doing this. I have argued this point before, but perhaps not in sufficient detail. Moreover, I need to engage more fully with certain important objections that have been raised against me in this regard by Jerrold Levinson in a recently published paper.3 Levinson has long been interested in, and sceptical about, trope theory, but the paper in question expresses his doubts most forcefully and cogently. So, what I want to do in what follows is, on the one hand, to respond to Levinson’s general doubts about the coherence of trope theory and, on the other, to reply to his more specific doubts about my own appeal to perceptual considerations in support of an ontology including modes or tropes.

3

See Levinson 2006. An earlier version in French appeared as “Pourquoi il n’y a pas de tropes”, in La structure du monde: objets, propriétés, états de choses, ed. JeanMaurice Monnoyer, Paris: J. Vrin, 2004. I should emphasize that everything substantive that I have to say about Levinson in what follows relates to this recent paper of his, not to his earlier writings on the subject.

179 3.

Levinson’s general objections to trope theory

Levinson was probably the first contemporary philosopher to speak explicitly of properties as ways of being—a locution that has since become extremely popular.4 His intention in using the expression was to wean us away from “thingifying” properties—that is, treating them as if they were objects of a queer sort. In this regard, I wholly applaud his aim. However, Levinson has never been happy with philosophers who speak of there being particular or “particularized” ways of being and who maintain that this is precisely what tropes or modes are.5 He thinks that ways of being are, by their very nature, shareable, although he concedes that there may be a logical reason why a way of being—such as being the first dog to reach the North Pole—cannot actually be shared by two or more objects. Even so, a different object could have had that way of being, so that it is, as it were, shareable “across possible worlds”. Now, a very important factor in Levinson’s approach to these matters is the distinction that he purports to find between properties and qualities. He suggests that there are two very different ways of nominalizing adjectives, such as “red” or “heavy”. One is to do so by means of gerundive expressions, such as “being red” or “being heavy”—and these, he says, denote properties. The other is to deploy abstract general nouns, such as “redness” or “heaviness”—and these, he says, denote (or purport to denote) qualities. The grammatical behaviour of the two types of expression is significantly different, he points out. In particular, we can quantify qualities—speaking, for instance, of “a little redness” or “more redness”—whereas we cannot do this with properties: “a little being red” and “more being red” don’t make sense. Now, Levinson’s contention is that it only makes any sense at all to particularize qualities, not properties. But, moreover, he contends that the only way in which it makes sense to particularize qualities is to think of qualities as abstract stuffs and of particular qualities as bits of that stuff—quality-bits. And, while he thinks that this makes sense, he regards it as being metaphysically utterly extravagant. One reason for this is that he thinks that talk in terms of qualities can, in principle, be paraphrased perfectly satisfactorily by talk in terms of properties. Thus, for example, instead of speaking of “more redness”, he suggests that we can speak of “being red in a higher degree”. 4 5

See Levinson 1978. See Levinson 1980.

180 As for properties, he maintains that in speaking of these we are speaking of conditions that objects can be in and that, as I’ve just said, it makes no sense to think of such conditions as being “particulars”. Levinson himself uses the generic expression “attribute” to cover both properties and qualities, but we may sum up his preferred position as being that all real attributes are properties and these are universals rather than particulars. As such, he seems to favour a position close to that of David Armstrong—one that acknowledges the existence of particular objects and universal properties, which come together in states of affairs when the former exemplify the latter. Armstrong himself, of course, regards states of affairs containing particulars as themselves being particulars—this is his doctrine of “the victory of particularity”6—but it would seem that Levinson is happy to go along with this too. What he wants to deny is that properties, although they are constituents of particular states of affairs, are themselves anything other than universals. All of this is very interesting, but also very contestable, in my view. Let me explain how I would like to reinterpret the grammatical data that Levinson appeals to in support of his own position. Let us begin with the gerundive expressions of the form “being A”, such as “being red” and “being heavy”. These expressions are formed from predicates of the form “— is A”, such as “— is red” and “— is heavy”, by turning the “is” of predication into the gerund “being”. Now, according to my own version of the four-category ontology and its attendant logical grammar, predication comes in two different varieties, corresponding to the two different varieties of exemplification—dispositional and occurrent. Thus, when we say of an object, such as a certain piece of cloth, that it “is red”, we may intend this in either of two different senses. If we know that the cloth is in a darkened room, or is illuminated by blue light, what we mean is that it is disposed to display or manifest redness in normal lighting conditions. On the other hand, if we know that it is in normal lighting conditions, what we mean is that it is actually (or “occurrently”) displaying or manifesting redness. This distinction applies to all predicates, but only sometimes is it explicitly marked in English. English does not explicitly mark the two different sense of “is red”, but with a verb, such as “dissolve”, it does so by means of a difference in what grammarians call “aspect”. Thus, we may say of a grain of salt either that it “dissolves in water” (dispositional) or that it “is dissolving in water” (occurrent). If English had used verbs 6

See Armstrong 1997, 126-7.

181 instead of adjectives to describe colour, as some other languages do, we might have been able to say of our piece of cloth either that it “reds” (dispositional) or that it “is redding” (occurrent). Similarly, we might have been able to say of an elastic piece of rubber either that it “squares” (dispositional) or that it “is squaring” (occurrent), but as it is English limits us to saying that it “is square”, thus disguising the dispositional/occurrent distinction that applies even in the case of shape attributes. Now, I do not consider that the dispositional/occurrent distinction is one that serves to distinguish between two different types of attribute. An attribute, as such, is neither dispositional nor occurrent in nature. Rather, the dispositional/occurrent distinction concerns how attributes are exemplified by objects. Consequently, because gerundive expressions of the form “being A” include not only mention of an attribute, A, but also of its exemplification—via the gerund “being” which nominalizes the “is” of predication—they do not purely serve to pick out attributes. In short, it is an error to suppose that “being A” simply denotes a certain attribute or property, A, even if we often do, misleadingly, speak of “the property of being A”. In both the predicate “— is red” and the gerundive expression “being red”, “red” denotes a certain attribute, namely, redness. But the copula “is” and its corresponding gerund, “being”, express the relation of exemplification, which, as we have seen, comes in two different varieties. Thus, expressions of the form “being A”, which Levinson regards as denoting properties, are in fact hybrid expressions, being composed of a categorematic expression “A” together with a syncategorematic expression, “being”. What Levinson calls quality terms, such as “redness” and “heaviness”, do, by contrast, simply denote attributes. That is why “This apple is red” is logically equivalent to “This apple exemplifies redness”, but not to “This apple exemplifies being red”, which is a syntactic monstrosity. The lesson of all this is that Levinson attempts to eliminate the wrong type of expression when he seeks to re-express all talk of “qualities” in terms of (what he calls) “properties”. There are no “properties” in his sense, since his notion of a property illicitly mixes together the notion of a universal with that of its exemplification. Thus far, what we may conclude is that, if we are going to admit genuine universals into our ontology, we need to admit what Levinson calls “qualities”, such as redness and heaviness—quite contrary to his own preference for “properties” (as he conceives of them). But what, then, of Levinson’s objections to the idea of particularized qualities? Recall that, according to Levinson, it’s not that this idea makes no sense, in the way

182 that, according to him, the idea of particularized properties makes no sense (and here I agree with him, but only because I think that his conception of “properties” makes no sense). Rather, it’s that he thinks that it is metaphysically extravagant to suppose that there are “abstract stuffs” such as redness, which have “bits”, in the way that concrete stuffs such as water and gold do. However, I entirely repudiate Levinson’s conception of what a particularized quality would have to be. A particularized quality is not a quality-bit but a quality instance. For, quite generally, the relationship of particulars to universals is that of instantiation, not parthood. Thus, just as an individual apple (an “object”, in my terminology) is an instance of the kind apple, so too this apple’s individual redness is an instance of the attribute redness. Levinson misapplies the language of mass terms when he talks of “quality-bits”. The sense in which a bit of something is part of that thing applies perfectly well in the case of concrete stuffs: one can quite properly speak, for instance, of a bit of the butter in the dish. Here, the expression “the butter in the dish” denotes a certain individual portion or “parcel” (as John Locke would have called it) of butter and a bit of that portion is just a lesser such portion that is included in the bigger portion as a proper part of it. But the mass term “butter” does not denote a very large and scattered portion of butter—as it were, the biggest such portion in the world, of which all other portions are proper parts. Rather, it denotes a substantial kind, of which all the butter-portions in the world are particular instances. So, by analogy, the only sense in which this apple’s redness is “a bit of redness” is the sense in which it is a particular instance of redness. There is no implication that redness is somehow spread around the world and that this apple’s redness is a part of that large and scattered entity. I conclude that the ontology of tropes or modes is not bizarre and extravagant in the way that Levinson seems to suppose. Now, I said earlier that, in my view, Levinson’s conception of what he calls “properties” makes no sense. But really I should qualify this. What I strictly mean to say is that nothing rightly regarded as a property, conceived as a universal, can intelligibly be thought of as being denoted by an expression of the form “being A”, such as “being red” or “being heavy”. I don’t necessarily mean to deny that something can be denoted by such an expression. In fact, I am inclined to agree with Levinson’s own suggestion that such expressions denote conditions—or, as I would prefer to say, states—of objects. Thus, I am happy enough to say that there can be both dispositional and occurrent states of objects. But a state is not a property: rather, it consists in the exemplifying of a property—and, because a

183 property (conceived as a universal) can be exemplified either dispositionally or occurrently, there are, accordingly, both dispositional and occurrent states of objects. However, states are not fundamental entities according to my version of the four-category ontology, in the way that properties (conceived either as universals or as particulars) are. An object is in a certain dispositional state in virtue of instantiating a kind that is characterized by a certain property-universal, or attribute. An object is in a certain occurrent state in virtue of being characterized by a propertyparticular, or mode, that instantiates a certain attribute. According to this diagnosis, Levinson’s error is to confuse properties with states. But what, it may now be asked, about states of affairs? What are they, according to my version of the four-category ontology? Well, once again, I don’t regard them as being fundamental entities, in the way that Armstrong apparently does—since his view seems to be that objects and properties (the “constituents” of states of affairs) are abstractions from or invariants “across” states of affairs.7 On the other hand, I certainly don’t want to identify them with what I have just been calling states of objects. This is because a state of an object, O, does not include O itself as a “constituent”. An expression of the form “being A” denotes such a state— which, as we have seen, may be either dispositional or occurrent in character, depending on whether “being” is taken to signify dispositional or occurrent exemplification. But a state of affairs, in the most familiar sort of case, is denoted by an expression of the form “O’s being A”, in which explicit mention of an object, O, is made. States of affairs, thus, are standardly denoted or referred to by nominalizations of complete sentences, of the form “O is A” (in this most familiar sort of case), whereas states of objects are standardly denoted or referred to by nominalizations of predicates, of the form “— is A” (at least, in the simplest sort of case). Both types of nominalization differ importantly from pure nominalizations of adjectives, of the form “A-ness”, which genuinely denote properties or qualities. (Having rescued the term “property” from Levinson’s misapplication of it, we can now return to using it as being pretty much interchangeable with the term “quality”.) Thus, we end up with the following view of these matters. There are properties or qualities, conceived as universals—in short, there are attributes. There are also objects. Both objects and attributes are entities of fundamental sorts. Then there are states of objects and there are states of affairs that contain 7

See Armstrong 2004.

184 objects, but neither such states nor such states of affairs are entities of fundamental sorts. Both states of objects and states of affairs containing objects come in two different varieties, dispositional and occurrent, depending on the sort of exemplification that they involve. In addition, there are, of course, states of affairs containing as constituents entities belonging to my other two fundamental categories, kinds and modes. For example, “water’s being transparent” denotes a state of affairs that contains a substantial kind, water, as a constituent, along with an attribute, transparentness. And “This redness’s being brighter than that redness”— where “this redness” denotes, say, the apple’s redness and “that redness” denotes the cloth’s redness—denotes a state of affairs containing two redness modes or tropes. But, once again, such states of affairs are not entities of a fundamental sort according to my four-category ontology, which considers the only fundamental entities to be those that belong to one or other of the four basic categories. It has been necessary for me to spell out in such detail the differences between my ontology and that seemingly favoured by Levinson, not only in order to rebut the general objections that he raises about the very notion of tropes or modes, conceived as particular(ized) properties, but also in order to address his more specific objections to my appeal to perceptual considerations in defence of the inclusion of tropes or modes in my ontology. I hope that I have by now gone some way towards my first objective, of showing that Levinson’s general objections to the very idea of a particular(ized) property are ill-founded. But it is one thing to argue that a certain ontological category—in this case, that of tropes or modes—has been coherently conceived by its advocates, rather than resting on some sort of linguistic or conceptual confusion. It is quite another to justify the inclusion of entities belonging to that putative category in one’s ontology. That is the issue that now concerns me. I have to concede that my own four-category ontology is more complicated than some others that have recently been advanced, including, most obviously, that of the pure trope ontologist, who adopts a one-category ontology. However, this is not the place for me to register in any detail my differences with the pure trope ontologist, since for present purposes I can regard myself as the temporary ally of such an ontologist, in respect of our opposition to Levinson. So now I want to turn to the explanation and defence of my argument from perception for the existence of tropes.

185 4.

The argument from perception for the existence of tropes

It will be helpful, I think, if I begin by simply restating, in its original form, my argument from perception for the existence of tropes or modes, since it is this formulation of the argument that is the explicit target of Levinson’s criticisms. This, then, is what I said:8 When I see a green leaf, do I not see the very greenness of the leaf, rather than just the leaf itself? [The opponents of tropes] may reply that they conceive of universals in an ‘Aristotelian’ fashion, as existing in re or ‘immanently’, so that greenness (the universal) is, as the popular phrase has it, ‘wholly present’ in each individual green thing. So, on this view, I do see the greenness of the leaf, but what I see is not a particular greenness but just the universal, which is ‘wholly present’ in the leaf (but is also ‘wholly present’ in other, spatially separate things). But here I protest for the following reason: when I see the leaf change in colour—perhaps as it is turned brown by a flame—I seem to see something cease to exist in the location of the leaf, namely, its greenness. But it could not be the universal greenness which ceases to exist, at least so long as other green things continue to exist. My opponent must say that really what I see is not something ceasing to exist, but merely the leaf’s ceasing to instantiate greenness, or greenness ceasing to be ‘wholly present’ just here. I can only say that that suggestion strikes me as being quite false to the phenomenology of perception. The objects of perception seem, one and all, to be particulars—and, indeed, a causal theory of perception (which I myself favour) would appear to require this, since particulars alone seem capable of entering into causal relations. (I should add that, if my [earlier] argument against the existence of concrete universals is correct, then the universal greenness cannot literally be located where a green leaf is and so cannot be something seen to be there: in which case, what is seen to be there must instead be the leaf’s particular greenness.)

It will be seen from this passage that I make a number of different claims in the course of the foregoing argument, which can be listed as follows.

8

See my 1998, p. 205.

186 (1) When I see a green leaf, I do not only see the leaf: I also see its greenness. (2) When I see the leaf change in colour from green to brown, I see its greenness cease to exist. (3) The universal greenness does not cease to exist when the leaf changes colour from green to brown (assuming that other green things continue to exist). (4) Therefore, from (2) and (3), the greenness that I see when I see the green leaf is its particular greenness, not (or not only) the universal greenness. (5) In any case, only particulars can be seen and hence only the leaf’s particular greenness, because the causal theory of perception is correct and only particulars can enter into causal relations. (6) Furthermore, it is mistaken to suppose that universals, even conceived as “immanent”, are literally located where the objects exemplifying them exist, so that the universal greenness cannot be what I see when I see the leaf’s greenness, since the latter is located where the leaf is.9 This is what Levinson says in response to the foregoing argument.10 [D]o I see something cease to exist when the leaf turns from green to brown? Certainly, but this need not be a greenness trope of the leaf, if I am right that we should not countenance any such 9

Levinson shares my view that universals are not located where the objects exemplifying them exist and are not “wholly present in” those objects: see his 2006, 566, n. 8. Hence, I have no need to defend (6) in the context of a response to Levinson, and I shall say no more about it in what follows. However, I do defend this view elsewhere, in my 1998, 155-6, and in my 2006, 98-9. 10 See Levinson 2006, 576-7.

187 thing. What, then? Two things. First, on a physical plane there is the leaf’s green pigment, which in the process described indeed ceases to exist. Second, on a metaphysical plane, there is the leaf’s exemplifying or exhibiting of the colour green or the colour property being green, which again, ceases to exist or comes to an end. Put otherwise, the leaf changes from being one way, green, to being another way, brown. ... In addition, one surely sees that the leaf is green, and then subsequently, that it is another colour. But none of these latter admissions implicate anything that is both an attribute and a particular.

We can sum up Levinson’s response as follows: (7) I do see something cease to exist when I see the leaf change colour from green to brown—indeed, I see two things cease to exist: the leaf’s green pigment and the state of affairs of the leaf’s being green—but neither of these is a trope. (8) Furthermore, I first see that the leaf is green, and subsequently that it is brown, but this fact, too, by no means implies that tropes exist. I do not find Levinson’s response satisfactory, for a number of reasons. First of all, he says that I see the leaf’s green pigment cease to exist. This green pigment, scientists tell us, is chlorophyll. Chlorophyll is a chemical substance that is indeed green in colour and can be used as a colouring agent. And it is true enough that, when the green leaf is turned brown by the flame, its chlorophyll is destroyed—ceases to exist—and that that is why the leaf no longer appears green in colour. Now, perhaps I do indeed see the leaf’s chlorophyll cease to exist when I see the leaf change colour. But the leaf’s chlorophyll is a quantity of chemical stuff, not a visible property. And yet it is a visible property, not just a quantity of chemical stuff, that visibly seems to cease to exist when I see the leaf change colour from green to brown. Of course, the leaf’s chlorophyll has a visible property—greenness—and its greenness is what makes the leaf appear green in colour. So perhaps it is really the chlorophyll’s greenness that visibly seems to cease to exist when I see the leaf undergo this change: perhaps what I have been calling the leaf’s greenness is really just the chlorophyll’s greenness. No matter: it is still a greenness that visibly seems to cease to exist, not just some green object or quantity of stuff. So it

188 strikes me that Levinson’s proposal that what I see cease to exist is the leaf’s chlorophyll is simply inadequate to the phenomenology of the situation. The foregoing objection to Levinson’s first proposal can be strengthened as follows. His proposal, recall, is that what I see cease to exist in the location of the leaf is the leaf’s green pigment, its chlorophyll. However, seeing a green object’s green pigment cease to exist is neither necessary nor sufficient for seeing that object change in colour from green to another colour. It is not necessary because the green pigment might itself change in colour, without ceasing to exist. And it is not sufficient, because only someone who sees not just the green pigment but also its greenness will notice a colour-change when the pigment ceases to exist—and even then only provided that another green pigment does not come into existence at the same time, replacing the first one. What does appear to be both necessary and sufficient for seeing a green object change in colour is seeing its greenness cease to exist. This reinforces my conclusion that Levinson’s first proposal is unsatisfactory. Now let us turn to Levinson’s second proposal: that what I see cease to exist is the state of affairs of the leaf’s being green. Now, it is open to question whether we ever literally see states of affairs. Certainly, “I saw the leaf’s being green” is a syntactic monstrosity, unlike “I saw the greenness of the leaf”, which is perfectly acceptable. The nearest we come to speaking of seeing states of affairs is when we speak of seeing situations or events. For instance, although it is syntactically improper to say “I saw John’s entering the room”, it is perfectly acceptable to say “I saw John enter the room”. In such a case, the verb “see” is complemented by a socalled naked infinitive construction.11 In this example, it seems that what I am claiming to have seen is a certain event involving John—the event of John entering the room on a certain occasion. So could Levinson’s second proposal be modified—could he claim that what I see cease to exist when I see the leaf change colour from green to brown is a certain event involving the leaf? No, he couldn’t—and certainly not without undermining the very purpose of his proposal. In the first place, events do not begin or cease to exist: they occur. Secondly, although I am perfectly happy to say that I do see an event in the case in question, that event is precisely the colourchange of the leaf—for events, quite generally, are changes. But what does this colour-change of the leaf consist in? Surely, it simply consists in the 11

See further Barwise and Perry 1981 and 1983, ch. 8.

189 leaf’s original greenness ceasing to exist and being replaced by its subsequent brownness. So we are back, once again, to the conclusion that one thing that I see when I see the leaf change colour is its greenness cease to exist. Finally, let me turn to the additional comment that Levinson makes about the example under discussion. This is that “one surely sees that the leaf is green, and then subsequently, that it is another colour”. First of all, this claim may be challenged. Very arguably, one can see a green leaf, and even see its greenness, without seeing that the leaf is green. The point here concerns a distinction, famously made by Fred Dretske, between “epistemic” and “non-epistemic” seeing.12 Someone who can be said to see that the leaf is green can, for that very reason, also be assumed to believe that the leaf is green. But, very plausibly, someone can be said to see a green leaf, and even see its greenness, without any implication that this person necessarily believes that the leaf is green—for the person may be unaware that what he or she is seeing is a leaf and, hence, that the greenness is the leaf’s. For example: suppose that an artist sticks a green leaf on to a white sheet of paper and then paints around the leaf in a shade of green exactly matching the colour of the leaf. Someone looking at this sheet of paper from a suitable distance may simply not notice that the leaf is there, because it blends so well into its background. Nonetheless, it seems correct to say that this person sees the leaf and indeed sees its greenness, because if the leaf had not been there his or her visual experience would have been noticeably different: after all, the person would have noticed the white leaf-shaped patch of paper that would have been exposed in the absence of the leaf. Secondly, I am inclined, in any case, to go even further than Dretske and say that “seeing that” is not really a kind of seeing at all. What we standardly report using the “seeing that” construction is, I suggest, not literally an episode of visual perception as such, but rather a perceptual judgement or belief—such a judgement or belief being one that a percipient makes or forms on the basis of an episode of visual perception, but which should not be confused with that episode itself. For example, I may say to a companion in an art-exhibition, “I wonder who painted that picture in the corner” and my companion may reply, rather condescendingly, “Can’t you see that it’s by Turner?”. My companion and I have very similar visual experiences: we both see the swirling patterns of blue, green, and pink 12

See Dretske1969.

190 paint in exactly the same detail. But, with his superior knowledge of art history, he is able to form a judgement on the basis of his visual experience that the artist in question is Turner, whereas I cannot. We both see a painting by Turner, but only he sees that it is a painting by Turner. This does not reflect any difference between his visual acuity and mine, only a difference between our cognitive capacities—in this case, between our artistic knowledge and powers of artistic judgement. It is my cognitive deficiency that my companion is calling attention to by asking his condescending question, not my capacity for visual discrimination. The case would have been very different if my companion had asked me that same question, believing that I was an expert on Turner: for then the implication would have been that, indeed, something was wrong with my eyesight. In that case, the question would not have been asked in a condescending way, but with an expression of astonishment and concern. How, exactly, do these observations bear upon claim (8) above? Well, in the first place, as I have just indicated, it simply does not follow from the fact that one sees the leaf change colour from green to brown that, as Levinson puts it, “one ... sees that the leaf is green, and then subsequently, that it is another colour”. To be persuaded of this, we simply need to imagine, as before, that the leaf in question is stuck on to a sheet of white paper and surrounded by green paint, in such a fashion that one does not see that the leaf is green before the flame is applied to it. A fortiori, in such a case, one does not see that the leaf is green and subsequently that it is brown—and yet one would still be seeing the leaf change colour from green to brown. Let us suppose, however, that we are dealing with a case in which one does see that the leaf is green and subsequently that it is brown. Then we need to ask (a) whether Levinson is correct to suppose that this fact does not imply the existence of tropes or modes as entities that are seen in such a case and (b) whether, if he is correct about that, he is also correct to suppose that this threatens to undermine the argument from perception for the existence of tropes. Now, with regard to (a), Levinson clearly is correct. For the fact that one sees that the leaf is green and subsequently that it is brown is, as I have just explained, a fact concerning “epistemic” seeing—or more accurately, in my view, a fact concerning perceptual judgement or belief—and not a fact concerning what entities are seen. “S sees that the leaf is green” does not, in itself, tell us what entities S sees, only what S judges or believes on the basis of his or her visual experience. S could see that the leaf is green even without seeing the leaf—if, for example, S saw a colour-photo of the leaf and judged or

191 believed on the basis of that visual experience that the leaf is green. Thus, since the truth of “S sees that the leaf is green and subsequently that it is brown” leaves entirely open the question of what entities S sees, Levinson is clearly right to suppose that its truth does not imply the existence of tropes or modes as entities that are seen by S. With respect to (b), however, Levinson is certainly not correct to suppose that the foregoing fact undermines the argument from perception for the existence of tropes. It does not, for the very simple reason that the argument from perception is not an argument that appeals to considerations concerning “epistemic” seeing or (as I would prefer to put it) concerning perceptual judgment or belief. Of course, if we were entitled to suppose that all seeing is “epistemic” seeing, then indeed Levinson would have helped to undermine the argument from perception for the existence of tropes. But I would vigorously deny that we are entitled to suppose that. In short, claim (8) is simply irrelevant to the argument from perception for the existence of tropes because, although facts about seeing that don’t imply the existence of tropes as entities that are seen, this is only because they leave entirely open questions about what entities are seen and, in any case, the argument from perception for the existence of tropes does not appeal to facts about seeing that but, rather, to facts about the phenomenology of visual perception. 5.

Concluding remarks

Jerrold Levinson has certainly presented what may seem, at first sight, to be devastating objections both to the very idea of tropes or modes as particular(ized) properties or qualities and to my own argument from perception for the existence of such entities. What I hope to have achieved in this paper is, first of all, a convincing defence of trope theory—or, more exactly, of the theory of modes within my own version of a neoAristotelian four-category ontology—against Levinson’s charge that such a theory is at worst incoherent (when it concerns “properties”), being founded on linguistic confusion, and at best metaphysically extravagant (when it concerns “qualities”). Secondly, I hope that I have succeeded in presenting more persuasively the argument from perception for the existence of tropes and in defending it against Levinson’s contention that, although something particular is indeed seen ceasing to exist in the example appealed to by that argument, there is no reason to suppose that

192 any such particular is a trope or mode. Let me finish, however, by saying that, while I have been vigorous in my defence of trope theory and the argument from perception against Levinson’s objections, I have expended so much effort in this defence only because I have the very highest regard for Levinson’s metaphysical insight. It is because trope theory could not have a worthier enemy that a rebuttal of his objections to it is such a pressing task for the friends of tropes. REFERENCES Angelelli, I. (1967). Studies on Frege and Traditional Philosophy. Dordrecht: Kluwer. Armstrong, D. M. (1997). A World of States of Affairs. Cambridge: Cambridge University Press. Armstrong, D. M. (2004). How do Particulars Stand to Universals?. Oxford Studies in Metaphysics, Volume I, ed. D. W. Zimmerman, Oxford: Clarendon Press, 139-154. Barwise J. and Perry J. (1981). Scenes and Other Situations. Journal of Philosophy, 78, 369-97. Barwise J. and Perry J. (1983). Situations and Attitudes. Cambridge MA: MIT Press. Dretske, F. (1969). Seeing and Knowing. London: Routledge and Kegan Paul. Levinson, J. (2006). Why There are No Tropes. Philosophy, 81, 563-79. Levinson, J. (1978). Properties and Related Entities. Philosophy and Phenomenological Research, 39, 1-22. Levinson, J. (1980). The Particularisation of Attributes. Australasian Journal of Philosophy, 58, 102-15. Lowe, E. J. (1998). The Possibility of Metaphysics: Substance, Identity, and Time. Oxford: Clarendon Press. Lowe, E. J. (2006). The Four-Category Ontology: A Metaphysical Foundation for Natural Science. Oxford: Clarendon Press.

About the authors Simone Gozzano is professor of philosophy of science at the university of L’Aquila in Italy. His main interests are in the philosophy of mind, metaphysics and the cognitive sciences. He is the author of three books in Italian, Pensieri materiali (Material Thoughts, Utet Università, Turin, 2007), Storia e teorie dell’intenzionalità (History and Theories of Intentionality, Laterza, Rome, 1997) and Intenzionalità, contenuto e comportamento (Intentionality, Content and Behaviour, Armando, Rome, 1997). Moreover, he has written a number of papers in international and Italian journals. John Heil is professor of philosophy at Monash University and at Washington University in St Louis, U.S.A. His current work concerns issues in metaphysics and the philosophy of mind. His most recent books include Philosophy of Mind: A Contemporary Introduction, 2nd ed. (Routledge, London, 2004) and From an Ontological Point of View (Oxford University Press, Oxford, 2003). E. Jonathan Lowe is Professor of Philosophy at Durham University, UK. He specializes in metaphysics, the philosophy of logic and language, the philosophy of mind and action, and early modern philosophy (especially the philosophy of John Locke). His books include Kinds of Being (Blackwell, London, 1989), Subjects of Experience (Cambridge University Press, Cambridge, 1996), The Possibility of Metaphysics (Oxford University Press, Oxford, 1998), A Survey of Metaphysics (Oxford University Press, Oxford, 2002), Locke (Routledge, London, 2005), The Four-Category Ontology (Oxford University Press, Oxford, 2006), and Personal Agency (Oxford University Press, Oxford, 2008). Anna-Sofia Maurin is assistant professor of theoretical philosophy at Lund University, Sweden. Her research interests include trope theory, truth theory and methodology (the latter mainly in metaphysics and ontology). She is the author of If Tropes (Kluwer Academic Publishers, Dordrecht, 2002), which investigates and evaluates the foundations of trope theory. Ausonio Marras is professor of philosophy at the University of Western Ontario, Canada. He holds a Ph.D. degree from Duke University and has

194 published widely in the philosophy of mind in international journals such as Erkenntnis, Synthese, The British Journal for the Philosophy of Science, Behavioral and Brain Sciences. His current research interests are in mental causation, consciousness, and nonreductive physicalism. Francesco Orilia is professor of philosophy of language at the University of Macerata, Italy. His main research interests are in the theory of reference, logical paradoxes, ontology. He has co-edited (with W. J. Rapaport) Thought, Language and Ontology – Essays in memory of Hector-Neri Castañeda (Kluwer Academic Publishers, Dordrecht, 1998) and is the author of a book in Italian on nonexistent objects Ulisse, il quadrato rotondo e l’attuale re di Francia, (Ulysses, The Round Square and the Present King of France, ETS, Pisa, 2002) as well as of Predication, Analysis and Reference (CLUEB, Bologna, 1999) and La Référence singulière et l’autoréférence (Reference and Self-reference, eum, Macerata, 2006; free download available at http://archiviodigitale.unimc.it). He has also published many papers in international journals such as Journal of Symbolic Logic, Journal of Philosophical Logic, Notre Dame Journal of Formal Logic, Synthese. David Robb is Associate Professor and Chair of Philosophy at Davidson College, U.S.A. He is interested in the philosophy of mind and metaphysics, especially mental causation, consciousness, and the nature of properties. He has co-edited (with Timothy O’Connor), Philosophy of Mind: Contemporary Readings (Routledge, London, 2003) and is author (or co-author) of a number of papers in international journals such as The Monist, American Philosophical Quarterly, Philosophical Review, Philosophical Quarterly. Juhani Yli-Vakkuri holds an M.A. degree in philosophy from the University of Western Ontario and is currently a Ph.D. candidate at McGill University. His research interests lie in the philosophy of language and the philosophy of mind.

PhilosophischeAnalyse PhilosophicalAnalysis 1 Herbert Hochberg Russell, Moore and Wittgenstein The Revival of Realism

8 Rafael Hüntelmann Existenz und Modalität Eine Studie zur Analytischen Modalontologie

2

9 Andreas Bächli / Klaus Petrus Monism

ISBN 3-937202-00-5 334 pp., Hardcover € 94,00

Heinrich Ganthaler Das Recht auf Leben in der Medizin Eine moralphilosophische Untersuchung

ISBN 3-937202-01-3 167 pp., Hardcover € 58,00

3 Ludger Jansen Tun und Können Ein systematischer Kommentar zu Aristoteles’ Theorie der Vermögen im neunten Buch der „Metaphysik“ ISBN 3-937202-02-1 302 pp., Hardcover € 70,00

4 Manuel Bremer Der Sinn des Lebens Ein Beitrag zur Analytischen Religionsphilosophie ISBN 3-937202-03-X 134 pp., Hardcover € 58,00

5 Georg Peter Analytische Ästhetik Eine Untersuchung zu Nelson Goodman und zur literarischen Parodie ISBN 3-937202-04-8, 332 pp. Hardcover € 94,00

6 Wolfram Hinzen / Hans Rott Belief and Meaning Essays at the Interface ISBN 3-937202-05-6 250 pp., Hardcover € 58,00

7 Hans Günther Ruß Empirisches Wissen und Moralkonstruktion Eine Untersuchung zur Möglichkeit von Brückenprinzipien in der Natur- und Bioethik ISBN 3-937202-06-4 208 pp., Hardcover € 58,00

ISBN 3-937202-07-2 189 pp., Hardcover € 58,00

ISBN 3-937202-19-6 340 pp., Hardcover € 70,00

10 Maria Elisabeth Reicher Referenz, Quantifikation und ontologische Festlegung ISBN 3-937202-39-0 ca. 300 pp., Hardcover € 89,00

11 Herbert Hochberg / Kevin Mulligan Relations and Predicates ISBN 3-937202-51-X 250 pp., Hardcover € 74,00

12 L. Nathan Oaklander C. D. Broad's Ontology of Mind ISBN 3-937202-97-8 105 pp., Hardcover € 39,00

13 Uwe Meixner The Theory of Ontic Modalities ISBN 3-938793-11-2 374 pages, Hardcover,€ 79,00

14 Donald W. Mertz Realist Instance Ontology and its Logic ISBN 3-938793-33-3 252 pp., Hardcover, EUR 79,00

15 N. Psarros / K. Schulte-Ostermann (Eds.) Facets of Sociality ISBN 3-938793-39-2 370 pp., Hardcover, EUR 98,00

16 Markus Schrenk The Metaphysics of Ceteris Paribus Laws ISBN 13: 978-3-938793-42-8 192pp, Hardcover, EUR 79,00

EditedBy • HerbertHochberg • RafaelHüntelmann ChristianKanzian • RichardSchantz • ErwinTegtmeier

PhilosophischeAnalyse PhilosophicalAnalysis 17 Nicholas Rescher Interpreting Philosophy The Elements of Philosophical Hermeneutics ISBN 978-3-938793-44-2 190pp., Hardcover € 89,00

18 Jean-Maurice Monnoyer(Ed.) Metaphysics and Truthmakers ISBN 978-3-938793-32-9 337 pp., Hardcover € 98,00

19 Fred Wilson Acquaintance, Ontology, and Knowledge Collected Essays in Ontology ISBN 978-3-938793-58-9 XX, 726., Hardcover, EUR 159,00

20 Laird Addis, Greg Jesson, and Erwin Tegtmeier (Eds.) Ontology and Analysis Essays and Recollections about Gustav Bergmann ISBN 978-3-938793-69-5 312 pp., Hardcover, EUR 98,00

21 Christian Kanzian (Ed.) Persistence ISBN 978-3-938793-74-9 198pp., Hardcover, EUR 79,00

22 Fred Wilson Body, Mind and Self in Hume’s Critical Realism ISBN 978-3-938793-79-4 512pp., Hardcover, EUR 139,00

23 Paul Weingartner Omniscience From a Logical Point of View ISBN 978-3-938793-81-7 188pp., Hardcover, EUR 79,00

24 Simone Gozzano, Francesco Orilia Tropes, Universals and the Philosophy of Mind Essays at the Boundary of Ontology and Philosophical Psychology ISBN 978-3-938793-83-1 194pp., Hardcover, EUR 69,00

EditedBy • HerbertHochberg • RafaelHüntelmann ChristianKanzian • RichardSchantz • ErwinTegtmeier