Axiomathes (2008) 18:273–288 DOI 10.1007/s10516-008-9048-x INVITED PAPER

New Directions in Metaphysics and Ontology E. J. Lowe

Received: 6 August 2008 / Accepted: 7 August 2008 / Published online: 30 August 2008
Springer Science+Business Media B.V. 2008

Abstract A personal view is presented of how metaphysics and ontology stand at the beginning of the twenty-first century, in the light of developments during the twentieth. It is argued that realist metaphysics, with serious ontology at its heart, has a promising future, provided that its adherents devote some time and effort to countering the influences of both its critics and its false friends. Keywords Metaphysics
Ontology
Scepticism
Nominalism
Essence
Modality
Categories
Conceptualism
Pseudo-ontology

The first decade of a new century is a good time for reflection, for anyone engaged in an intellectual discipline whose history spans many hundreds of years and whose character, methods, and justification have always been a matter for deep and sometimes bitter controversy. That metaphysics is such a discipline no one will dispute. So where does metaphysics stand at the beginning of the twenty-first century? In what direction or directions might it be heading? In what direction should it be heading? And what dangers beset its progress in the right direction—if indeed it is appropriate to speak of right and wrong directions in this context? In order to answer these and related questions, I shall have to spend some time looking back to see how metaphysics has arrived at its present state. I shall also have to say something about what I take metaphysics to be, which is a highly contentious matter in itself. In order to avoid dealing purely in generalities, however, I shall try to illustrate my answers to these questions by mentioning some specific developments in metaphysical thinking that have occurred fairly recently. Inevitably, given the limited space available, I shall have to be highly selective in my choice of examples and—perhaps unsurprisingly—my choice of them will be dictated in large measure E. J. Lowe (&) Durham University, Durham, UK e-mail: [email protected]

123

274

Axiomathes (2008) 18:273–288

by my own primary interests in the field of metaphysics. I do not pretend to be offering a perfectly dispassionate and unbiased account of the issues that I am about to address. However, I think that anyone who did claim to be doing that should be viewed with suspicion: for no one can really be a metaphysician without feeling strongly about why metaphysics is important and how it should be pursued.

1 Why Metaphysics? A perfectly understandable query to raise at this point would be this: Why should metaphysics, in particular, of all branches of philosophy, be selected for the treatment that I am about to give it? Couldn’t the general questions that I have just posed be asked, with equal or even greater legitimacy, about any branch of philosophy, or indeed about philosophy itself as a whole? My first response to this query is a purely personal one: I select metaphysics for this treatment because it is a special interest of mine and—in the current jargon—constitutes my main ‘area of expertise’. This is not to say that my philosophical interests are purely metaphysical, which would be a distinctly unhealthy intellectual condition in which to find oneself. I regard the increasing specialization of academic philosophy as an extremely bad and dangerous thing, largely impelled by external factors, such as the growing pressure on academics in all disciplines to produce more publications and to rigidify teaching syllabuses in order to satisfy politically driven managerial demands and bureaucratic assessment criteria. In my view, no one can call him or herself a genuine philosopher who isn’t interested in the whole of philosophy. I see a serious threat to philosophy in the increasingly unquestioned assumption that academic philosophers should conform to a scientific model of research. Indeed, the very idea of conducting ‘research’ in philosophy—other than, perhaps, into its history—is more than a little absurd, because engaging in philosophy is not a matter of producing evidence-based information about some subject matter. Rather, it is a matter of achieving insight and understanding about matters with which, in some sense, all of us are already perfectly familiar. However, my second and more fundamental reason for selecting metaphysics for the special treatment that I am about to give it is that—in my view—metaphysics lies at the heart of all philosophy, whatever its detractors might say. It really is ‘first philosophy’, as it was traditionally called. In my opinion, we cannot pursue studies in ethics, or philosophical logic, or epistemology, or the philosophy of mind, or the philosophy of language, or the philosophy of science, without relying on certain distinctly metaphysical presuppositions—presuppositions which, at some stage, we ought to reflect upon and try to justify. Ethics, for example, unavoidably involves fundamental metaphysical considerations concerning the nature of persons as moral agents and the nature of actions as morally evaluatable events. Again, the philosophy of logic and language requires us to take a view about the ontological status of sentences or propositions as the bearers of truth and falsehood—and indeed about the nature of truth itself. In fact, for reasons that I shall come to later, I consider that metaphysics is presupposed not only by all other branches of philosophy, but by all other intellectual disciplines whatsoever, including not only

123

Axiomathes (2008) 18:273–288

275

the sciences, both natural and social, but also the humanities, such as history and literary studies—in short, it is presupposed by any human intellectual enterprise that includes amongst its aims the pursuit of truth.

2 Metaphysics Now and in the Twentieth Century How, then, do I see the current state of metaphysics? Metaphysics, I am inclined to say, is in better shape now than it has been in for a very long time: better, possibly, than it has been in ever since the seventeenth century, the golden—or more accurately the silver—age of metaphysics. That, of course, was the era of the great rationalist metaphysicians: Descartes, Spinoza, and Leibniz. It was a time at which there was harmony between developments in metaphysics and progress in the natural sciences, with no fundamental barrier between these two enterprises, many of whose most illustrious practitioners were engaged in both. And if I say that it was more accurately only a silver age, this is not to detract anything from its glory, but just to acknowledge that the higher accolade should surely go to the era of Plato and Aristotle, the founders of western metaphysics. But why do I say that the good times have returned for metaphysics? In a nutshell, I say so because—in some quarters, at any rate—metaphysics is, once again, not being done surreptitiously, under the guise of something else. What other guises do I have in mind here? Well, in the middle decades of the twentieth century, at least, metaphysical inquiry was conducted, for the most part, under the guises of epistemology, the philosophy of language, and the philosophy of logic—as indeed it is still is in rather too many places. I shall try, in a moment, to explain why this happened. But it is not difficult to illustrate my thesis. One has only to look, for example, at the principal works of A. J. Ayer, Michael Dummett, and W. V. Quine, to take just three of the best-known analytic philosophers of the last century. Ayer’s most significant forays into metaphysics were largely driven by epistemological concerns, especially problems arising in the philosophy of perception.1 As for Dummett, he has consistently maintained that metaphysics should be pursued through the medium of questions in the philosophy of language and, more particularly, the theory of meaning.2 So too have many other analytic philosophers who, like Dummett, have spoken approvingly of the so-called ‘linguistic turn’ in philosophy, which they locate at some point in the early-to-mid twentieth century. Then again, consider Quine’s characterization of what he calls the ‘ontological question’ and its supposed answer in ‘On What There Is’—surely one of the most frequently reprinted papers of the last century.3 The question, according to Quine, is ‘What is there?’—and its one-word answer is ‘Everything’. This only semifacetious remark is, of course, directly connected with Quine’s criterion of 1

See, for example, Ayer (1940). I am setting aside here Ayer’s youthful outburst against metaphysics in his (1936) because, despite the anti-metaphysical rhetoric of that book, it is clear that in his later work Ayer did address distinctively metaphysical questions, even if only in a somewhat oblique way, by framing them in an epistemological context.

2

For one of his more recent expressions of this view, see his (1991).

3

See Quine (1961).

123

276

Axiomathes (2008) 18:273–288

ontological commitment, encapsulated in his famous slogan, ‘To be is to be the value of a variable’.4 What clearer evidence could there be of metaphysics being pursued via the philosophy of logic and language? Incidentally, allow me to sound here a brief note of protest against those present-day philosophers who—quite mistakenly, in my view—give Quine the credit for putting ontology back on the agenda of analytic philosophy. For reasons that I shall try to make plain later, I believe that Quinean ‘ontology’ is a travesty of the real thing and more of a threat to genuine ontology than anything issuing from the pens of self-professed antirealists and relativists.

3 Metaphysical Scepticism and its Antidote If you grant me the thesis that metaphysics during much of the twentieth century was typically pursued under various other guises, you will want me to address the following further questions. Why was it done under these guises? And how else might it be done? What would it be to do metaphysics ‘directly’, as it were, rather than obliquely, under the guise of something else—and how could we do that? A brief answer to the first of these questions—why was metaphysics done under other guises?—is that this happened as a reaction to the by then almost universal belief in the impossibility of metaphysics, as this branch of philosophy was traditionally conceived. That metaphysics so conceived is simply impossible was urged by antirealists and empiricists of various colours, such as the logical positivists and their latterday heirs—and here I include once more such philosophers as Ayer, Dummett, and Quine. The seeds of this hostility to traditional metaphysics were, however, originally sown long ago in the eighteenth century, first by David Hume and later by Immanuel Kant. These seeds were then nurtured throughout the nineteenth century by Kant’s idealist successors, such as Fichte, Hegel, and Schopenhauer, amongst whom we may also include, in the English-speaking world, the late Hegelian idealists Bradley and McTaggart.5 But Kant it was who did the greatest damage: for how could traditional metaphysics easily have survived attack by the greatest philosopher since Aristotle? In the Critique of Pure Reason, Kant claimed, in effect, that we cannot know reality ‘as it is in itself’, only how we conceive of it. But how could we know even that, if he was right? This, in my view, is the fundamental contradiction at the heart of Kant’s system. How we conceive of reality—that is, the structure and content of our thought about reality—is itself just a part of reality, not something that could intelligibly be set in opposition to reality as a possible object of our knowledge, in the way that the rest of reality is allegedly not. We find exactly the same contradiction in the work of latter-day antirealists and relativists, who try to set up a similar opposition between ‘language’ and ‘the world’, maintaining that our attempts to characterize the latter—extra-linguistic reality—are merely unconscious 4

See, for example, Quine (1969).

5

This may be rather unfair to Hegel, at least as his views on metaphysics are interpreted in some quarters: see, in particular, Stern (2008, forthcoming).

123

Axiomathes (2008) 18:273–288

277

reflections of distinctions built into the former, ‘our language’ or ‘our ways of describing the world’. But again I protest that ‘our language’ and ‘our descriptions’ are just more bits of the world, and not particularly interesting or important bits at that, unless one happens to be a linguist or a lexicographer. My contention, then, is that because metaphysics is unavoidable for any inquiring mind, those philosophers who feared ridicule at the hands of sceptics and antirealists for raising metaphysical issues were compelled to address them in an indirect fashion—for example, as questions about ‘our concepts’ or ‘our language’ of causation, time, personal identity, or what-not. And those who peddled the sceptical doctrine connived at this, relishing the demise of traditional metaphysics that they imagined they had brought about. As for the second question that I raised a moment ago—what, then, would it be to do metaphysics ‘directly’, and how could we do it that way?—this has, in my view, a very practical answer. The answer is to be found by taking courage and simply trying it for yourself. Don’t ask, for instance: ‘How should we analyse our concept of causation?’, or ‘What is the meaning of the word ‘‘cause’’?’—just ask yourself: what could causation possibly be? You will, in all probability, soon find yourself beginning to think of some possible answers to this question and beginning to find arguments for or against various of those answers. My advice is: Just pursue these arguments and see where they lead you. There is absolutely no guarantee that you will be led to an indisputable final answer to your question, but even so you will learn much during the quest. We should not expect metaphysics to be able to produce such final solutions, any more than we should expect them in mathematics or any other intellectual discipline. It was, indeed, Kant’s unreasonable expectation that we should be able to arrive at certainty in metaphysics that led him to distort it into an examination of the structure and content of our thought about reality rather than of the structure and content of reality itself.

4 What is Metaphysics? At this point you would be fully entitled to ask me: What, then, is metaphysics, shorn of all its false guises? Well, there are three complementary ways in which I like to characterize it, all of them needing a good deal of unpacking, some of which I shall try to do as I proceed. These three characterizations—I won’t say definitions, since I rather dislike definitions, finding them, for the most part, either tendentious, vacuous, or useless—are the following.6 First, metaphysics is the study of the most fundamental structure of reality as a whole. Second, metaphysics is the systematic exploration of the bounds of possibility. And third, metaphysics is the science of essence. The last of these characterizations will seem particularly obscure and gnomic, perhaps, but I think that, ultimately, it is the most important of all. Now I want to say something briefly about each of these three characterizations of metaphysics and why I see them as apt, with each of them capturing an important 6 My dislike, I should emphasize, is confined to merely verbal or semantic definitions, as opposed to socalled real definitions, the latter being understood as expressing essences.

123

278

Axiomathes (2008) 18:273–288

dimension of the discipline. And let me emphasize that I see the three characterizations as being entirely complementary and, indeed, as implying each other. Firstly, I said, metaphysics is the study of the most fundamental structure of reality as a whole. As I see it, metaphysics is needed in this role because truth is single and indivisible, but other disciplines pursue it only within delimited domains.7 Because other disciplines pursue truth only within delimited domains, no one of these other disciplines is equipped to adjudicate between the potentially conflicting truth-claims of their various practitioners. To give an obvious example: theological truth-claims may threaten to conflict with the truth-claims of physical scientists. Since truth is single and indivisible, a resolution must be sought, but neither physical scientists nor theologians as such are equipped to seek, much less to find, it. If a resolution can be found at all, it will be found in metaphysics. Secondly, I said, metaphysics is the systematic exploration of the bounds of possibility. The point here is that we need a discipline whose task it is to chart the bounds of possibility because what is actual is delimited by what is possible.8 Let me spell out this idea a little more fully. Most intellectual disciplines are concerned with the discovery of truths about actuality—about whether, for instance, electrons are really fundamental particles, or whether such-and-such a drug is protective against some form of disease. These disciplines must, of course, appeal to empirical evidence to support their findings concerning such questions. But empirical evidence can only support a conclusion whose truth is at least possible. No amount of evidence can, for example, support the conclusion that the path of a comet is both elliptical and parabolic, or that a chemical substance is both acidic and alkaline. In these cases, the impossibility is manifest, but in many other cases it is not. A classic example serves to illustrate the point: is time-travel into the past possible or not? It is simply not obvious that it is, nor that it is not. Settling the question is ultimately a matter for metaphysics. And do not protest here that this is usurping the rightful role of physical science. Physicists offer us their various theories of time and space, but the question of what, if anything, time is is a metaphysical one par excellence. The mere fact that a mathematical theory of physics employs a variable ‘t’ whose numerical values are somehow correlated with clock-readings is no guarantee whatever that what those values really signify are times. There is a prior question to be addressed as to the ultimate nature of temporality and whether, indeed, there are even such entities as ‘times’. Thirdly and lastly, I said, rather darkly, that metaphysics is the science of essence. All I shall say about this for the moment is that we need knowledge of essence because essence is the ultimate ground of all possibility. But more of this anon. However, before proceeding, let me emphasize that by ‘science’ in this context I emphatically do not mean empirical science and by ‘essence’ I do not mean some sort of secret ingredient hidden away inside things—I am certainly not proposing that metaphysics is some kind of latter-day alchemy! In this respect, mathematics provides a much closer model to what I have in mind. By a ‘science’, 7

For a defence of the thesis of the unity of truth, see my (2006, chap. 11).

8

I defend this thesis in my (1998, chap. 1).

123

Axiomathes (2008) 18:273–288

279

in this context, I just mean a systematic and rationally constrained study of some subject-matter, issuing in objective knowledge about it.

5 The False Dawn of Modern Metaphysics Now let me turn again to some fairly recent history. In my opinion, there was a false dawn for metaphysics in the 1970s and 80s, with the rise of possible-worlds semantics for modal discourse, going hand-in-hand with the growing deployment of set-theory and mereology—the formal theory of part–whole relations—in metaphysical theorizing, which thereby became increasingly technical in character. This, it seems to me, gave rise to a kind pseudo-ontology, in which substitutes for various parts of reality are posited in place of the realities themselves. David Lewis, incomparably brilliant philosopher though he was, encouraged many budding metaphysicians to take wrong directions in this manner, or so I believe. Lewis had, it seems, scant regard for the ‘robust sense of reality’ that Bertrand Russell urged us to retain even in the most abstract of studies. On the contrary, one suspects that he even rather relished the fact that his key metaphysical doctrine—his so-called modal realism—was commonly received with what he called ‘the incredulous stare’.9 Lewis advertized his domain of realistically conceived possible worlds as a veritable metaphysician’s paradise. Recall that, for Lewis, every possible way the world could be is a way that some world is—in short, is a world, just as solidly real as we take the actual world to be. For Lewis, the actual world is just ‘me and my surroundings’, as far as they stretch out from me in space and time. And other possible worlds are just other very big, concrete things of the same kind, differing from ours only in this or that little or large respect. You may say: surely this is taking ontology seriously, if anything is! Well, I don’t think so. I think that Lewis’s modal realism is just the product of taking too seriously Saul Kripke’s possibleworlds semantics for modal logics and Quine’s criterion of ontological commitment, along with a severely pragmatic cost–benefit approach to the evaluation of competing metaphysical schemes. What I mean by this is the following—expressed very simply and crudely, no doubt. Kripke showed that we could, for logical purposes, usefully interpret ‘It is possibly the case that p’ as meaning ‘There is some possible world, w, such that p is true in w’. Applying Quine’s criterion of ontological commitment, this translation commits us to ‘worlds’ as the possible values of variables such as ‘w’ here. So what is a ‘world’? Lewis argued that, all things considered, the most parsimonious theory of worlds that could most adequately accommodate our modal intuitions was his full-bloodedly realist one, as opposed to the various ‘ersatzist’ rival accounts— accounts according to which worlds are maximal consistent sets of propositions or other such abstract entities. Using the paraphernalia of possible worlds, we are then supposedly equipped with ways of reducing all sorts of entities to various settheoretical constructions. For example, we can take propositions to be sets of worlds—namely, the sets of worlds in which they are true. And we can take 9

See Lewis (1986a, p. 133).

123

280

Axiomathes (2008) 18:273–288

properties to be functions from worlds to sets of objects existing in those worlds. As for objects themselves, if these persist through time we can treat them as mereological sums or fusions of their supposed temporal parts. The underlying axiom of Lewisian metaphysics is the doctrine of Humean supervenience: fundamentally all that exists are the worlds, each of them a patchwork of physical qualities distributed across space and time—all else ‘supervenes’ on this.10 The only principles of combination that are recognized are set-theoretical and mereological in character. That is to say, all ‘complex’ entities are either set-theoretical constructions or mereological fusions. And, true to the spirit of Hume himself, there are no ‘necessary connections’ between ‘distinct existences’. The result is a marvellously comprehensive system, put together as only a genius like Lewis could do it. But is this really what metaphysics and ontology should be about? I myself don’t think so. For this approach isn’t really concerned with understanding the nature of things, because it substitutes mere surrogates for the things themselves and rests content with understanding them—to the extent, indeed, that we do really understand these surrogates, such as the ‘worlds’. Whatever a property is, for example, it is surely not a function from worlds to sets of objects. It may be thought that my criticisms should not be extended to Kripke himself—for did he not do valiant service in defence of metaphysics by revitalizing the distinction between essence and accident and in explaining how knowledge of necessary truths might be accommodated with empiricism?11 But Kripke, too, led philosophers in the wrong direction regarding precisely these matters, or so I believe. For it was he who led philosophers to think that modal notions—notions of necessity and possibility—are prior to the notion of essence, whereas I think that the very reverse is the case. But more of this later.

6 The Revival of Real Ontology So where did the true dawn of modern metaphysics lie, if not in development of possible-worlds semantics for modal discourse? In my view, it lay in the work of philosophers such as David Armstrong, who amongst other things revived interest in the theory of universals.12 This, at last, was genuine ontology. And what, exactly, is that? Well, ontology is the science of being qua being—to borrow from Aristotle— with the theory of ontological categories at its heart. Again, our greatest debt is to Aristotle, since his early work the Categories is the founding text of this branch of philosophy. Before Armstrong, nominalism was the unspoken assumption of most analytic philosophers, under the pervasive influence of Quine. By ‘nominalism’ here, however, I mean a view which is, I think, best characterized by its implicit denial of the reality of ontological categories. I should emphasize that this is not what ‘nominalism’ was quite properly taken to mean in earlier times, when it signified a denial of the existence of universals in favour of the existence of 10

See, especially, the Introduction to Lewis (1986b).

11

See Kripke (1980).

12

See Armstrong (1978).

123

Axiomathes (2008) 18:273–288

281

particulars alone. Genuine nominalists believe in ontological categories—it’s just that they think that all existing things belong to the one category of particulars. That, for example, is what John Locke thought. What happened under the influence of Quine is that ‘things’ came to be thought of as nothing more than the values of variables—items over which we take ourselves to be quantifying when we talk about this or that subject-matter. Another way to put this is to say that Quinean ‘ontology’—or, more accurately, pseudo-ontology—is a no-category ontology, in contrast with the classical nominalist’s one-category ontology of particulars. I speak here of the classical nominalist, because in more recent times another kind of nominalism has developed, which is again a one-category ontology of particulars, but this time of particular properties rather than particular objects. These ‘particular properties’ are nowadays popularly known as tropes.13 Of course, the object–property distinction itself is another categorial distinction of traditional metaphysics, one which, in fact, cuts across the particular–universal distinction. Aristotle himself, it seems, favoured a four-category system of basic ontology, the categories being, in effect, the following: (1) particular properties, (2) universal properties, (3) particular objects, and (4) universal objects—or, in more familiar and traditional terms, individual accidents or modes, attributes, individual substances, and substantial kinds or species. It is a version of this system that I favour myself.14 However, there is, obviously, plenty of room for healthy dispute here between genuine ontologists. Armstrong, for instance, espouses a two-category ontology of particular objects and universal properties (and relations), which he takes to be the basic constituents of all states of affairs.15 But at least these genuine ontologists are trying to work out what sorts of basic entities the world really does contain, rather than talking in terms of set-theoretical constructions which are at best mere surrogates for those entities. The development of set theory in the late nineteenth century was undoubtedly a great boon to mathematics: but it was, in my view, misappropriated by logically-minded analytic philosophers in the following century for entirely misguided purposes. So what, exactly, are ontological categories and why do we need a theory of them? To start with, it is important to appreciate that ontological categories, while they are categories of actual and possible entities, are not themselves entities. That is to say, they should not themselves be included in a complete inventory of ‘what there is’. You may protest: but how can you even talk about ‘them’ if ‘they’ don’t exist? Well, if you say this, then you are probably under the spell of Quine and his doctrine that ‘to be is to be the value of a variable’. I have said that I favour an ontological system according to which there are four basic ontological categories. Translate this claim into Quinese and we get an existentially quantified statement beginning ‘(Aw)(Ax)(Ay)(Az)(w is an ontological category & x is an ontological category & …’. Now applying Quine’s criterion of ontological commitment, it seems that we can immediately conclude that I am ontologically committed to the existence of ontological categories—at least four of them. But this is ludicrous. If I 13

See Campbell (1990).

14

See again my The Four-Category Ontology.

15

See Armstrong (1997).

123

282

Axiomathes (2008) 18:273–288

thought that ontological categories were entities at all, I would probably have to hold that they are fundamental entities belonging to a basic category of their own— the category of ontological category. That would make my system a five-category ontology. But it isn’t! Anyway, I’m going to go on talking about ontological categories without presuming that ‘they’—as opposed to the entities belonging to them—exist, in defiance of Quine’s criterion. So—to repeat—what ‘are’ ontological categories, if I may be excused for framing the question in this way? Well, very roughly, they are the basic varieties or types of entities, a catalogue of which is indispensable for the purpose of characterizing the fundamental structure of reality as a whole. What ultimately distinguishes them from one another, in my view, are the distinctive existence and identity conditions of the entities belonging to them and the relations of ontological dependence that characteristically hold between entities belonging to the various categories. By way of example, I can do no better than to cite again some instances mentioned earlier: particular objects, or individual substances—things like this glass or that table—belong to one ontological category, while particular properties, or tropes, such as the roundness and transparency of this glass or the rectangularity and brownness of that table, belong to another. The existence and identity conditions of substances are distinctly different from those of tropes. Moreover, substances and tropes stand in certain characteristic relations of ontological dependence to one another—for example, each trope, in my view, depends for its existence and identity upon one particular substance, but the reverse is not the case. Here it may be asked: But aren’t these so-called ontological categories really entirely ‘language-relative’, merely reflecting the grammatical idiosyncrasies of the languages that we happen to use to describe the world? For example, isn’t the category of individual substance just the ontological projection of the grammatical category of substantive noun? Not in my view. It shouldn’t surprise us if there is some kind of match between grammatical distinctions and ontological distinctions, because language is, after all, designed—or, at least, has evolved—to enable us to talk informatively about the world. But let us not be duped, simply on this account, into accepting the extravagant views of Benjamin Lee Whorf and other such extreme linguistic and cultural relativists.16 On closer examination, grammar is not, in fact, a very good guide to ontology—nor should we expect it to be, since our palaeolithic ancestors probably had little time for the luxury of metaphysical speculation when the human language-faculty was in the process of evolving. Let us not lose sight of the fact that there can be major disputes amongst genuine ontologists who speak the same language—something that could hardly be the case if categorial systems were fixed in our brains as a product of language. Even if reflection about ontological categories begins with acceptance of a common-sense scheme implicit in everyday language, it certainly needn’t—and, in my view, shouldn’t—end there. In the terminology made famous by P. F. Strawson, we shouldn’t rest content with merely descriptive metaphysics: genuine metaphysics is nothing if not revisionary in spirit.17 Though that, of course, was not Strawson’s own view. 16

See Whorf (1956).

17

See Strawson (1959).

123

Axiomathes (2008) 18:273–288

283

7 Of Entity and Essence Now, to fulfil an earlier promise, I want to say something concerning entity and essence. I choose the phrase deliberately, to echo of the title of Thomas Aquinas’s famous treatise, De Ente et Essentia—not that I shall have anything to say specifically about Aquinas’s own views on this subject. It is natural at this point to turn to the notion of essence, since an entity’s ontological category is, par excellence, part of its essence—part of what it is. As Locke so perceptively remarked, ‘essence’ in the ‘proper original signification’ of the word denotes ‘the very being of any thing, whereby it is, what it is’.18 If we are to think about anything, we must at least know what it is that we are thinking of—in other words, we must grasp its essence. And something’s essence determines what is possible for it. For example, it is part of the essence of any individual substance that it is an individual substance, as a consequence of which it could not have been something belonging to another ontological category. Thus, this table, although it might conceivably have been round instead of rectangular, could not itself have been a particular property, such as roundness or rectangularity. Being an individual substance, it is the type of thing that necessarily possesses properties such as these, but couldn’t itself have been one of these properties, any more than the number 7 could have been a fried egg. And here is another example of how an entity’s essence grounds what is possible for it: it is part of the essence of any living creature that it interacts with its environment through processes of nutrition, respiration and the like, all of which involve exchanges of matter between the creature and other material things. This, partly, is what it is for such a creature to be alive. Consequently, it must be possible for a living creature to undergo a change of its material parts over time. This is not the case with, for instance, a mere aggregate of material particles, such as a lump of rock. To put it another way: living creatures and mere lumps of matter have different identity conditions and consequently different persistence conditions. And this reflects the fact that living creatures and mere lumps of matter, while they both belong to the ontological category of particular objects, or individual substances, nonetheless belong to two different subcategories of that category. Ontological categories can be represented as organized in a tree-like hierarchical structure, with the topmost or most general category of all being simply that of entities—something like this: entities

universals

attributes

substantial kinds

particulars

tropes

individual substances

living creatures

18

lumps of matter

See Locke (1975), III, III, 15.

123

284

Axiomathes (2008) 18:273–288

Of course, this is only a very partial diagram, ignoring many further subcategories additional to those of living creatures and lumps of matter. The important point to emphasize is that an entity’s ontological category is part of its essence— part of what it is—and that its essence grounds what is possible for that entity. However, it is vital here not to fall into the trap of supposing that an entity’s essence is some further entity, specially related to it in some peculiarly intimate way. It could not be, not least because we would then be faced with an infinite regress of essences, since each essence—being, supposedly, a further entity—would have to have its own essence, and so on ad infinitum. But in that case only an infinite mind, at best, could grasp the essence of anything. The implication would be that we with finite minds could really not think about anything—for remember that to think about something one must understand what it is that one is thinking about: in other words, one must grasp its essence. However, it is simply incoherent for us to deny that we can think about anything, for the very denial involves just such a thought. This, if you like, is a ‘transcendental proof’ of our ability to grasp the essences of at least some things and shows that essences themselves are not entities of any sort, any more than ontological categories are. Furthermore this, at bottom, explains how modal knowledge—knowledge of possibility—is itself possible for beings like us. For it is part of our essence, as thinking beings, that we can grasp the essences of at least some things and thereby understand at least in part what is possible for those things, since what is possible for them is grounded in their essences. If essences were entities of a special sort, then indeed our grasp of them would be difficult—or worse, impossible—to explain. We certainly couldn’t know them by some peculiar kind of perception or observation, since any such mental act with intentional content requires a grasp or understanding of that content by its subject and thereby a grasp of the essences of its intentional objects. It is no use, then, for those who regard essences as entities to suppose that we could be acquainted with them by means of some special perceptual faculty. So what kind of ‘faculty’ is it that we possess, that enables us to grasp the essences of at least some things? The answer is simple: it is the faculty of understanding, which we exercise whenever we engage in processes of reasoning or reflection. It may seem surprising, in the light of what I have just said, that any philosopher should have been trapped into supposing that essences could be entities—but many good ones have been, from Plato onwards. Locke himself provides an example, for he supposed the ‘real essences’ of material substances to be their ‘hidden inner constitutions’, by which he meant—roughly speaking—their atomic composition and structure. At this point let me try to quash one possible line of objection to what I have just been saying concerning essence. Some philosophers, who are in still in thrall to the idea that metaphysics has to be done via the philosophy of mind and language, will urge that all talk about ‘essences’ can at best be construed as talk about our concepts. Thus, it may be said, it is just our concept of a living creature that implies that such a being can undergo exchanges of matter with its environment and, indeed, just our concept of a natural number which precludes us from identifying the number 7 with a fried egg. But it should be quite manifest that this view is fraught with difficulty. For it seems clear that these philosophers, implicitly at least, take

123

Axiomathes (2008) 18:273–288

285

concepts themselves to be entities of some kind—entities which somehow populate our minds and which we are either acquainted with in some special way, or else somehow learn to master through linguistic practice. In point of fact, such philosophers are often rather reticent—and perhaps advisedly so—about what, exactly, they do take ‘concepts’ to be. For whatever a concept is, if indeed it is an entity of some sort, our knowledge of it would require our grasping its essence— knowing what it is. So, I conclude, either talk of ‘concepts’ is just another way of talking about essences in my sense—in which case, of course, I have no quarrel with it—or else it is purportedly talk about mental or, perhaps, abstract entities of some special kind. But in the latter case the same applies to concepts as applies to all other entities: namely, that to think about them we must grasp their essences and that their essences ground what is possible concerning them. Consequently, it seems to me, conceptualism, as we might call it—the notion that all possibility is grounded in ‘our concepts’—either simply reduces to genuine essentialism in my sense or else presupposes it and thereby undermines itself.

8 The Persistence of Pseudo-Ontology I spoke earlier of what I called surrogate or pseudo-ontology. I fear that this spurious conception of ‘ontology’ still characterizes too much work that is being done in the name of metaphysics. We find it, for instance, in a great deal of the current literature on the philosophy of time and persistence. Allow me to cite two of the more blatant examples. First, there is the appeal by so-called ‘perdurantists’— philosophers who maintain that persisting objects are temporally extended—to the notion of temporal parts or stages. This appeal and its rejection by the rival school of ‘endurantists’ have given rise to the increasingly convoluted and seemingly interminable dispute between ‘three-dimensionalism’ and ‘four-dimensionalism’.19 But what, really, is the difference between saying that objects persist by having different ‘temporal parts’ existing at successive times during their careers and saying, instead, that they persist by being ‘wholly present’ at each successive time? The key expressions in play here—‘temporal part’ and ‘wholly present’—are technical terms of art whose definitions are themselves no less contentious than the issue allegedly in dispute. If this were simply a debate about the nature of time itself—the question as to whether or not time is rightly thought of as being a dimension of reality, in anything like the sense in which the three spatial dimensions are—then, I think, it would be a metaphysically serious one. But in that case, why not focus on the real issue, rather than recasting it in terms of whether or not persisting objects have ‘temporal parts’? I suspect that the underlying answer to this question is just that the problem has the superficial appearance of being more tractable when recast in this way, because the new technical vocabulary that has been developed by the parties to the dispute lends it a spurious air of formal rigour.

19 For the distinction between perdurantism and endurantism, see Lewis, On the Plurality of Worlds, p. 202. Lewis himself, of course, favoured perdurantism. See also Sider (2001).

123

286

Axiomathes (2008) 18:273–288

My second example is the attempt by many self-styled ‘presentists’—philosophers who maintain that only the present moment and presently existing things are real—to exploit an analogy between times and possible worlds.20 Here, too, I think that there is a genuine metaphysical issue at stake, but one that is only obscured by indulging in pseudo-ontology. In this case, the pseudo-ontology involves a construal of ‘times’ as being abstract entities on the model of what Lewis called ersatz possible worlds. On this construal, a past or future time t is an abstract representation—or, more accurately, misrepresentation—of present truth, such as a maximal consistent set of present-tensed propositions, some of which are not presently true. Thus, on this interpretation, a past-tense statement of the form ‘p was true at t’ is interpreted as signifying something like ‘If reality were as t represents it to be, p would now be true’. But why this apparent compulsion to find some sort of thing for t to be, the object of desire being found, in this case, in the form of a special kind of abstract entity? Once again, I suspect we see at work here the malign influence of Quine’s doctrine, ‘to be is to be the value of a variable’. We supposedly find ourselves ‘quantifying over’ non-present times, when we utter such mundane truths as ‘There was a time when Napoleon was alive’—and as a consequence the values of these variables of quantification have, allegedly, to be identified in order to legitimize our temporal talk by Quinean standards. But ‘times’ so conceived have nothing to do with the real nature of time, any more than ‘possible worlds’ have to do with the real nature of possibility. They are mere surrogate objects, dangerously alluring to us because they have the spurious appearance of making the real metaphysical issue at stake easier to focus on. So how should we approach this issue instead—the issue of the nature of temporality? Here is a thought to begin with: just as possibility is grounded in essence, so temporality is grounded in coming to be and passing away. In both cases, the ground lies not in some special class of existing entities. But there the similarities end. Time is not a thing, nor are times. We should resist the temptation to reify it and them. Rather, it is just fundamental to the mode of existence of concrete things that they successively come into and go out of existence—and all of our talk about time and change is predicated upon this basic truth.

9 What Lies Ahead? Some Positive and Negative Developments Let me now draw matters to a close by posing and attempting to answer the following question: What positive developments are there now to be seen in metaphysics—and what negative? Here I shall be brief and highly selective—and, once again, my answers will be openly partisan. Amongst the negative developments I include the growing popularity of deflationism and fictionalism of various kinds—for instance, concerning truth and concerning modality. These, in my view, are strategies of despair, born of an unwarranted loss of philosophical confidence. It goes without saying, perhaps, that I also include scientism, euphemistically described by most of its devotees as ‘naturalism’— by which I mean the attempted 20

See further my (2002, pp. 42–43).

123

Axiomathes (2008) 18:273–288

287

surrender of the domain of metaphysical inquiry into the hands of empirical scientists. This, I think, is a strategy of ignorance, because it fails to recognize the unavoidable dependence of all empirical science upon metaphysical presuppositions. I also include all those ingenious philosophical strategies designed to give us a priori ‘armchair’ metaphysics on the cheap, such as those invoking the currently fashionable two-dimensional modal semantics—an approach to the interpretation of modal language which, in my view, simply compounds the errors of possible-worlds talk and dresses up conceptualism in a fancy new garb.21 These, to my way of thinking, are strategies of self-deception, motivated by a desire to find something useful for philosophy to do, on the false assumption that it cannot provide an independent source of objective knowledge. Amongst the positive developments I include the growing allegiance, at least in some quarters, to the truthmaker principle—the principle that, in general, truths need to be made true by the actual existence of mind-independent entities of suitable sorts.22 Of course, I don’t think that this principle applies to truths concerning essence—nor, hence, to modal truths—for reasons that should be plain, in view of points that I have made earlier. One great virtue of the truthmaker principle, methodologically speaking, is that it keeps us ontologically honest, for it requires us to think seriously about the ontological grounds of our assertions, rather than trivializing such issues in the way that Quine’s criterion of ontological commitment encourages us to. Another positive development, in my view—and this again will come as no surprise—is the dawning realization, in the metaphysics of modality, that the notion of essence must be regarded as prior to that of either necessity or possibility.23 In due course, we may hope to see entirely reversed the contrary assumption, which has dominated discussion in this area for more than three decades. But as the final and most important example of a positive development I would cite what I am inclined to call the new rationalism. This is the conviction that there is a source of a priori metaphysical knowledge that is available to us purely in our capacity as rational beings able to engage in reasoned thought. This kind of knowledge, I believe, transcends and undergirds any kind of merely empirical knowledge that can be gleaned from our sensory encounters with the world, whether they be direct or passed on to us through testimony. For understanding the world requires precisely that—understanding. Mere information is no substitute for understanding, however immediate, numerous, influential, or authoritative its sources. And understanding requires reason and reflection. During much of the twentieth century, philosophers lost the confidence to believe that we can acquire genuine knowledge in this way—and indeed that such knowledge is not only worthwhile but absolutely indispensable for all of our other intellectual endeavours. Now some of us are regaining that confidence. Philosophy, with serious metaphysics at its heart, really does have a future!

21 Although I disagree fundamentally with its approach to metaphysics, one of the best exemplars of the genre is Jackson’s (1998). 22

See Armstrong (2004).

23

For this we owe much to the work of Fine (1994).

123

288

Axiomathes (2008) 18:273–288

Acknowledgements I am grateful to comments received when earlier versions of this paper were delivered to audiences in Durham, Nottingham, and Buffalo.

References Armstrong DM (1978) Universals and scientific realism. Cambridge University Press, Cambridge Armstrong DM (1997) A world of states of affairs. Cambridge University Press, Cambridge Armstrong DM (2004) Truth and truthmakers. Cambridge University Press, Cambridge Ayer J (1936) Language, truth and logic. Victor Gollancz, London Ayer J (1940) The foundations of empirical knowledge. Macmillan, London Campbell K (1990) Abstract particulars. Blackwell, Oxford Dummett M (1991) The logical basis of metaphysics. Duckworth, London Fine K (1994) Essence and modality. In: Tomberlin JE (ed) Philosophical perspectives, 8: logic and language. Ridgeview, Atascadero Jackson F (1998) From metaphysics to ethics: a defence of conceptual analysis. Clarendon Press, Oxford Kripke SA (1980) Naming and necessity. Blackwell, Oxford Lewis D (1986a) On the plurality of worlds. Blackwell, Oxford, p 133 Lewis D (1986b) Philosophical papers, vol II. Oxford University Press, New York Locke J (1975) In: Nidditch PH (ed) An essay concerning human understanding. Clarendon Press, Oxford Lowe EJ (1998) The possibility of metaphysics: substance, identity, and time, chapter 1. Clarendon Press, Oxford Lowe EJ (2002) A survey of metaphysics. Oxford University Press, Oxford, pp 42–43 Lowe EJ (2006) The four-category ontology: a metaphysical foundation for natural science, chapter 11. Clarendon Press, Oxford Quine WV (1961) From a logical point of view, 2nd edn. Harvard University Press, Cambridge Quine WV (1969) Speaking of objects’ in ontological relativity and other essays. Columbia University Press, New York Sider T (2001) Four-dimensionalism: an ontology of persistence and time. Clarendon Press, Oxford Stern R (2008) Hegelian metaphysics. Oxford University Press, Oxford (forthcoming) Strawson PF (1959) Individuals: an essay in descriptive metaphysics. Methuen, London Whorf BL (1956) In: Carroll JB (ed) Language, thought and reality. MIT Press, Cambridge

123