Copyright holder: Tyndale University, 3377 Bayview Ave., Toronto, Ontario, Canada M2M 3S4 Att.: Library Director, J. William Horsey Library Copyright: This Work has been made available by the authority of the copyright owner solely for the purpose of private study and research and may not be copied or reproduced except as permitted by the copyright laws of Canada without the written authority from the copyright owner. Copyright license: Attribution-NonCommercial-NoDerivatives 4.0 International License Citation: Davis, Richard Brian and David S. Brown. “A Puzzle for Particulars?” Axiomathes 18, no. 1 (2008): 49-65. DOI: 10.1007/s10516-007-9018-8 Reprinted with permission from Springer (https://www.springer.com/) ***** Begin Content ****** TYNDALE UNIVERSITY 3377 Bayview Avenue Toronto, ON M2M 3S4 TEL: 416.226.6620 www.tyndale.ca Note: This Work has been made available by the authority of the copyright owner solely for the purpose of private study and research and may not be copied or reproduced except as permitted by the copyright laws of Canada without the written authority from the copyright owner. Davis, Richard Brian and David S. Brown. “A Puzzle for Particulars?” Axiomathes 18, no. 1 (2008): 49-65. DOI: 10.1007/s10516-007-9018-8 Reprinted with permission from Springer (https://www.springer.com/) Axiomathes (2008) 18:49-65 DOI 10.1007/s10516-007-9018-8 ORIGINAL PAPER A Puzzle for Particulars? Richard Brian Davis • David S. Brown Received: 24 June 2007 / Accepted: 18 September 2007 / Published online: 21 November 2007 © Springer Science+Business Media B.V. 2007 Abstract In this paper we examine a puzzle recently posed by Aaron Preston for the traditional realist assay of property (quality) instances. Consider Socrates (a red round spot) and red1—Socrates’ redness. For the traditional realist, both of these entities are concrete particulars. Further, both involve redness being ‘tied to’ the same bare individuator. But then it appears that red1 is duplicated in its ‘thicker’ particular (Socrates), so that it can’t be predicated of Socrates without redundancy. According to Preston, this suggests that a concrete particular and its property instances aren’t genuinely related. We argue that Preston’s proffered solution here—to treat property instances as ‘‘mental constructs’’—is fraught with difficulty. We then go on to show how, by fine-tuning the nature of bare particulars, treating them as abstract modes of things rather than concrete particulars, the traditional realist can neatly evade Preston’s puzzle. Keywords Property instances ■ Realism ■ Bare particulars ■ Individuation It is something of a commonplace in contemporary ontology to distinguish between universals, on the one hand, and their instances, on the other.1 Consider two red, round spots: Socrates and Plato, as we might call them. According to a realist view of things, we must distinguish between the property of redness or being red and Socrates’ redness, his instantiation of this property. The former is a universal. It is ________________ 1 See Bradley (1946) for a classic discussion of the relation between a particular and its properties. _________________ R. B. Davis (email logo) Philosophy, Tyndale University College, 25 Ballyconnor Court, Toronto, ON, Canada M2M 4B3 e-mail: rdavis@tyndale.ca D. S. Brown Philosophy, Lindenwood University, St. Charles, MO 63301, USA e-mail: dbrown@lindenwood.edu 50 universal and thus repeatable in that each individual instance of redness shares this "common, qualitative content" (Mertz 1996, 11). The redness of Socrates, by contrast, is neither universal nor repeatable; it is a particular quality instance, individuated specifically to Socrates. Following current convention here, let’s stipulate that ‘red1’ denotes Socrates’ redness, ‘red2’ that of Plato, while ‘wisdom’ (without the subscript) stands for the property or universal common to both. Now it has been said that a "proper realist" (Moreland 1991, 97) will not only acknowledge the existence of property (or quality) instances, she will assay them in a very specific sort of way. A property instance such as red1 is a complex entity, consisting of "the universal redness, the nexus of exemplification, and an individuator...a bare particular" (Moreland 2001, 15).2 Suppose we agree. According to Aaron Preston, assaying a property instance in this way presents the realist with a certain puzzle. For red1’s associated ‘thick’ particular, Socrates, the particular to which red1 belongs, also seems to share the same basic ontological structure: "both are property-individuator complexes bound by predication" (2005, 273). But then in order to "set the[se] terms in relation to one another" (279), red1 must be connected somehow to this thicker particular—a particular in which it seems to be already present. What this suggests, says Preston, is that a concrete particular and its quality instances aren’t genuinely related. Now in this paper we mean to dispute this claim. There are indeed serious and debilitating difficulties associated with proper realism—at least the sort Preston has in view.3 Moreover, Preston’s proffered solution here—to treat quality instances as "mental constructs" (285)—is itself fraught with difficulty. We argue (contra Preston) that his is no realist view of property instances. We then attempt to show how a careful recasting of the traditional conception of bare particulars can provide the basis for our being proper realists about property instances while neatly evading Preston’s puzzle.4 1 ‘Proper Realism’ Presented So what precisely is it to be a "proper realist"? As the term suggests, a proper realist is a particular kind of realist: one who commits herself to the real existence of objects of a certain sort—numbers, perhaps, or sets, or propositions, or what have you. A realist with respect to property instances, for example, believes that expressions such as ‘the redness of Socrates’ or ‘Wiles’ wisdom’ are no mere empty terms; unlike ‘Pegasus’ or ‘the Golden mountain’, they refer to things (more specifically, concrete ______________________ 2 Since this paper is set within the context of proper realism, we will not enter into the debate over the ontological status of universals. For purposes of argument, we shall simply stipulate their existence, taking no position on the question of whether they are imminent or transcendent in nature. 3 The central difficulty, of course, is its commitment to bare particulars. For critical assessment, see Mertz (1996, 72-73). See also Mertz (2001, 2002). For recent attempts to rehabilitate bare particulars by investing them with at least some properties, see Moreland (1998), Moreland and Pickavance (2003). For replies to Moreland’s refurbished theory, see Mertz (2003) and Davis (2003, 2004a). 4 This is not a paper in the history of philosophy. Readers interested in the history of the debate over how universals, particulars, and property instances are related should consult Mertz (1996) for a helpful overview and discussion. 51 particulars) that really do exist. To join the ranks of "proper" realists, however, we must go a step further. We must distinguish between two modes of predication: the ‘rooted in’ and ‘tied to’ modes. Consider first the proposition (1) Red1 is red where ‘red1’ denotes, as before, the redness of Socrates. Now according to proper realism’s most able exponent, J. P. Moreland, what (1) tells us is that redness is rooted in a concrete particular, red1. Thus Moreland invites us to read (1) as (1*) Red1 has the universal redness as a constituent. Here we see clearly that employing the ‘is’ of rooted in predication has the effect of introducing an internal complexity into an ontic (or extra-grammatical) subject. Now if this were the only mode of predication at our disposal, the constituent(s) of a complex particular would simply ‘float free’, which is presumably undesirable. Moreland therefore invokes what he calls ‘tied to’ predication, whereby properties may receive a proper ontic grounding. Thus (1) is said to be ‘‘grounded in’’ (2001, 146) (2) This bare particular is red which is just to say that while (1) roots the property of being red in red1, (2) ontologically grounds it, tying it down (as it were) to an ontic subject—in this case a bare particular.5 And perhaps this is as it should be; for otherwise a complex particular like red1 would amount to little more than a compresent bundle of properties. Here it is important to recognize that the bare particulars Moreland has in mind are not in fact bare at all. For they have properties externally connected or tied to them, and thus qualify as at least ‘partially clad’. Permitting such ontic connections enables Moreland to deftly side step the oft-repeated charge that bare particulars are incoherent, since nothing can exist without exemplifying some properties. And indeed, if there really are bare particulars, they should exemplify (have connected to them) a host of trivially essential properties: being self-identical, for example, or being colored if red, as these sorts of qualities are metaphysically fastened to every object. In short, ‘‘Bare particulars cannot exist without properties’’ (Moreland 2001, 157); if they exist at all, they must be partially clad. Still, there is a sense in which partially clad bare particulars (PCBPs) are bare. They are bare in the sense that they are wholly without internal constituents. A PCBP may have a property tied to it but never rooted in it. A moment’s reflection reveals why this must be the case; for if a PCBP had property-constituents, then they too would have to be tied to a bare particular; in which case we would be faced with a vicious regress of PCBPs—each containing a lower-level PCBP of its own as a constituent. But in fact the regress here need never begin. Moreland’s account of predication allows the proper realist to treat the PCBP as a brute, internally simple, property anchor, handily terminating any alleged regress before it even begins. _______________________ 5 A reviewer for this journal objects here that the move from (1) to (2) "reduces ontology to the demands of linguistic form." But this assumes an account of predication, s/he says, that is "too naive." This may be true, but it isn’t presently relevant. For we are not defending this account; we are simply reporting on Moreland’s defense of it. 52 Now this, we are told, is the story that all proper realists embrace. Proper realism is therefore not to be taken lightly; it exacts a substantial ontological commitment to an elaborate constituent-whole ontology, invoking the real existence of universals, property instances, and partially clad bare particulars, intricately held together by a highly novel account of predication. One might expect, therefore, that any problems associated with proper realism will involve what seems to be an unusual, bloated ontology. 2 Preston’s Puzzle and Proposal For Preston’s part, however, the chief difficulty with proper realism lies elsewhere. Its vulnerability lies not, as one might have expected, in its commitment to bare particulars (even if partially clad). Their coherence is (we think mistakenly) taken for granted. Instead, Preston is puzzled by how a property instance would relate to the thicker particular instantiating it. The puzzle emerges when we compare how proper realists assay thick particulars over and against their property instances. Some simple diagrams will easily illustrate the point. Let a line segment stand for an ontic predicate (a universal), and let a solid dot attached to such a line stand for a single ontic subject: a bare particular (hereafter, BP). Next suppose, for ease of expression, that Socrates has only two properties: redness and roundness. We can then represent the concrete particular that is Socrates as follows: In this complex—the Socratean complex, as we might call it—the horizontal and vertical line segments signify the universals redness and roundness respectively, and are labeled by means of the arrows. The dot or node labeled ‘S(b)’ stands for Socrates’ BP. Here it is evident that Ured and Uround, while rooted in Complex S, aren’t properly tied to Socrates at all. On the contrary, Socrates is a complex whole ‘built up’ so to speak from these universals being connected to (or predicated of) S(b). So far so good. The problem, if there is one, comes when we turn to the assay of Socrates’ redness or red1. This assay, as we said earlier, essentially involves redness being tied to a BP—a state of affairs, presumably, that can be represented in this way: [Please contact repository@tyndale.ca for diagram information titled: Complex R1 ] And now the question becomes: to which BP have we hereby attached redness? The answer, as Preston notes, can only be Socrates’ BP—that is, S(b). For how otherwise would Complex R1 be a depiction of the redness of Socrates? If the ‘dot’ stood for the PCBP of any other ‘thick’ particular, then surely the resulting diagram 53 would represent the redness of that particular and not Socrates. Thus Preston contends that "the individuating element of a concrete particular [i.e., its bare particular] must also serve as its unifying element" (2005, 267). We may safely assume, therefore, that in both complexes (i.e., S and R1) redness is predicated of one and the same bare particular, S(b). But then, for Preston, red1 is just a "little" particular, a "thinner" version of Socrates (273). Simply mentally ‘delete’ Uround from Complex S, and you’re left with red1. So how should we understand the relation between Socrates and Socrates’ redness, red1? Can we say, e.g., that (3) Socrates is red1 ? If so, what could we possibly mean? How can we join red1 to something in which it is "inseparably present" (279) as a constituent? It seems that we cannot. And thus Preston promptly concludes: "The fact that there is redundancy involved suggests that this is not a case of genuine relation" (ibid). A couple of initial comments. Preston’s duplication or redundancy problem seems to get its legs by taking the ‘is’ in (3) as expressing the tied to relation. But as Preston well knows, for Moreland, nothing is ever tied directly to a complex ordinary particular, especially not a quality instance which is, after all, just another particular and therefore not the sort of thing that could even serve as an ontic predicate. At the end of the day, the tied-to relation only holds between universals and the BPs that ground them. This is precisely why there is no point in retreating to the position that what (3) involves instead is rooted-in predication. For on Moreland’s view, apart from BPs themselves, anything rooted in a complex particular must be tied to a unique BP. But of course property instances aren’t predicative; they aren’t tied to anything. And even if they were, just how could something like red1 tie itself to S(b), its bare particular, when it already includes it? Can a BP be tied to itself? Surely not. But what follows? That a complex particular and its property instances aren’t genuinely related? Preston moves too quickly here. He first contends that there is neither a real nor a formal distinction to be made between Socrates and red1. They aren’t really distinct, he says, since neither is ‘‘capable of independent existence’’ (284). But they aren’t formally distinguishable either because there is no distinction between them that ‘‘exists in reality prior to having any mental activity focused upon it’’ (ibid). Indeed, the inseparability of Socrates and red1 is not limited to the realm of ‘real existence’...but is just as unavoidable in the realm of cognition. Whether we recognize it or not, the bare particular that we entertain in a thought of, say, red1, is by definition and in essence, the bare particular of Socrates. Thus even in our most isolated thought of red1, there is still an implicit reference to Socrates (285). We must therefore conclude, says Preston, that property instances are "no more than mental fabrications" (286). By way of our cognitive activities, we human knowers actually "construct" (285) them; and without our contribution, they presumably wouldn’t exist at all. This is apt to strike one as flagrantly anti-realist; but Preston insists that a property instance so conceived is not "a complete and utter 54 fabrication" (286). For mentally manufacturing such an instance might well occur as we mentally reflect on the constituents of a property-individuator complex— constituents "all of which are distinct in the real order" (ibid). In this way, Preston assures us, property instance constructivism can still be considered a species of traditional realism. 3 Puzzles for the Proposal 3.1 "Fabricated" Realism? But is this really true? We don’t think so. Note, first, that on Preston’s view it is we human beings who create property instances (hereafter, PINs). Somehow or another, perhaps by way of our powers of abstraction, we bring it about that a given PIN exists. But why believe a thing like that? If we cannot solve the redundancy problem, says Preston, the prudent thing to do is to "treat [PINs] as mental constructs" (285) and as "no more than mental fabrications" (286). Now this way of speaking certainly leaves the impression that PINs enjoy no extra-mental existence. Naturally enough, of course, they are dependent: Socrates’ redness won’t exist unless Socrates does. That seems obvious enough. But Preston goes a step further. On his view, the redness of Socrates depends not only on Socrates’ existence, but also on our mental activities—our grasping the real distinction between ‘‘the individuator, the universal quality, and their binding relation’’ (286). Apart from this activity, presumably, Socrates’ redness wouldn’t exist. Now this is not altogether easy to accept. Does anyone really believe that such things as the earth’s roundness or the cleverness of Quine—both PINs in their own right—owe their existence and fundamental nature to our cognitive contributions? And even if so, why should we think this counts as property instance realism? Indeed, one suspects that Preston is no more a realist about PINs than Locke is about the existence of secondary qualities. Sure, for Locke, there are colors, tastes, and sounds. There’s just one catch: they don’t exist outside the mind. What we have here is anti-realism at its finest. 3.2 The ‘Round Square’ Problem Still, you might think that Preston’s brand of realism has a bit more going for it than that. For Preston’s PIN-constructs are erected upon an extra-mental foundation. On this point, we may all agree. Even so, it must be said that whatever reality these fabricated PINs enjoy is at best derivative; it results from mental activity being directed at a specific arrangement of constituents in a concrete particular. Now for the proper realist, the critical element in this mix is the thin or bare particular, which serves as both individuator and unifier of its thicker cousin. Without a viable "pincushion into which universals may be poked" (Sider 2006, 387), there just is no concrete particular, and thus no ontic springboard for mental constructions of any sort. 55 Here we come face to face with the Achilles heal of PIN constructivism. For it seems that bare particulars lack the necessary ontological endowments to perform their dual functions of individuation and unification. The fundamental problem concerns their nature—or lack thereof. According to D. W. Mertz, for example, there are serious problems associated with tying properties to bare particulars. Where ‘pa’ denotes a bare particular, Mertz points out that nothing is in relatum pa (being devoid of all content) to be the source or cause of the Tied-to relation linking it to any universal as the other relatum. The Tied-to relation is completely external in regard to relatum pa. It makes no difference to the nature of relatum pa what properties are tied-to it, and so pa exists independently of such relatedness. (Mertz 2001, 51) He goes on to say that if the relatum pa has no "controlling content," no intrinsic nature, then "there is no reason why both Round and Square could not be tied to pa" (ibid). In other words, bare particulars are impossible objects. So the question is: do thin particulars have an intrinsic nature? Mertz thinks not. Sider, on the other hand, answers "in a flat-footed spirit; yes they do." Thus Sider: "The intrinsic nature of a particular is given by the monadic universals it instantiates. Since thin particulars instantiate monadic universals, they have intrinsic natures" (2006, 389). Period. This breezy remark strikes us as missing the mark. Since BPs are bare—i.e., without internal constituents—their natures (if any) will be tied to them externally. In and of themselves, however, BPs are natureless. Consequently, they should be wholly indifferent to their ontic attachments; for any thin particular t and property P, it will be contingent that t has P. In which case, since natures are essential to whatever they are natures of, it follows that thin particulars lack external natures as well. Any conjunction of properties (including contraries) may therefore be predicated of them. So Mertz’s charge of incoherence remains in force. Can anything be said by way of reply? Well, consider Socrates once again: our red, round spot. And let’s suppose that S(b)—Socrates’ BP—is entirely without a nature. We can certainly agree that S(b) is possibly round (since it actually is), but also possibly non-round (since it could have been more square-like in shape). We might still wonder how S(b) could be simultaneously round and non-round. This isn’t, after all, just obvious. For perhaps it is part of the nature of these properties that they can’t be jointly exemplified—the nature of whatever co-exemplifies them having nothing to do with it. Mertz is unmoved by this consideration: Consider the shared ‘subject’ of Triangle in the facts corresponding to ‘Round is contrary to Triangle’ and ‘Square is contrary to Triangle’. Here there is nothing in the natures of Round and Square as contraries that prevents them from being related by the relation Is-contrary-to to the shared relatum Triangle. Indeed, it is the very fact that they are contraries that allows them to enter into this particular relation with the intension [i.e., property] Triangle. Hence, it is possible that contrary properties can be related to the same subject. (2001, 51) This is an interesting little argument. Taking ‘is contrary to’ to mean ‘is jointly unexemplifiable with’, Mertz is apparently reasoning as follows. Since both 56 (4) Roundness is jointly unexemplifiable with triangularity and (5) Squareness is jointly unexemplifiable with triangularity are true, it follows that (6There is something to which roundness and squareness are both related is possible. Fair enough. But what we really need is a sufficient reason for believing (7) There is a PCBP to which roundness and squarenessare both ‘tied’ could be true; and (6)—whether actually or just possibly true—doesn’t provide it. From the fact that roundness and squareness, individually, can stand to some thing in some relation, it hardly follows that they can collectively stand to this thing (i.e., bare particular) in this very relation (i.e., being tied to). The inference is obviously invalid. A thin particular’s not having a nature is indeed a crippling defect; it’s just not clear that any combination of properties may therefore be predicated of it. 3.3 The ‘Insertion’ Problem This brings us to another less tractable problem. A constructed PIN, on Preston’s view, presupposes the reality and unity of the metaphysical elements composing a concrete individual. As Preston himself notes, "the individuating element of a concrete particular [i.e., its bare particular] must also serve as its unifying element" (2005, 267). Otherwise, Socrates’ various properties (e.g., redness, roundness) remain ungrounded and don’t belong to one and the same thing: Socrates. But how is Socrates’ BP supposed to do a thing like that? How is it supposed to unify anything, if it has no intrinsic nature? We need something to which redness and roundness can be connected or joined. We need a ‘‘whole consisting of spokes [universals] united to a hub [a bare particular]’’ (Mertz 2003, 15). We need a ‘‘pincushion into which universals may be poked’’ (Sider 2006, 388). Agreed. But is that all there is to it? We simply declare that a bare particular adequately serves this purpose? Surely this is but a first step. After all, when spokes fit into a hub, it’s obviously because the hub is pre-fitted for their insertion. It comes, so to speak, with slots or insertion points of a very specific sort. Not just any shape- or-sized spoke will do. That is to say, the hub must possess a particular nature ontologically prior to spoke insertion. This is why the spokes ‘connect’ or ‘inhere’ in their hub, and why the hub is thereby suited to the task of uniting them. The important thing to see, here, is that BPs do not possess this sort of nature. And thus we have no explanation for how a property ‘spoke’ could possibly fit into one of these featureless ontic ‘hubs’. Even if we hold, following Sider, that the intrinsic nature of a thing is a function of the (essential) properties it instantiates, the question still arises: how did those properties manage to get themselves inserted into an ontic hub whose nature isn’t suited to them (since in itself it has no nature)? 57 Sider’s reply is unavailing: "a thing could have a nature simply by failing to instantiate monadic universals" (2006, 388). Well, perhaps so. If having a nature is simply being a certain way, then perhaps a thin particular (in itself) does have a nature: that of being without property attachments. Does it follow, however, that it could sustain such attachments? It doesn’t. By way of analogy: suppose you work on the assembly line of a factory that makes bicycle wheels. Your job is to take wheel hubs and get them ready for spoke attachment. The hubs passed to you are hollow, symmetrical cylinders with raised flanges (ribs) at either end. They have no other distinguishing features. Now of course when you first receive a hub, it has the property having no spokes. But that fact by itself doesn’t guarantee that it could have spokes. For suppose you do nothing at all to the hub, and simply pass it down the line for its spokes to be attached. How is that supposed to happen? You haven’t done your job; you were supposed to drill holes in the flanges, so that the spokes could be attached. As it is, however, you didn’t and so they can’t. In fact, there are no doubt many (prior) essential features that a hub must possess before spoke attachment is even possible: being cylindrical, having flanges, having spoke holes, and the like. This is the sort of nature a spoke-friendly hub cannot be without. Now if you go on to insist that we must connect spokes to a thing that not only has no spoke holes, but is also defined by the total absence of properties—that is, it’s not a hub, not cylindrical, has no flanges, etc.—then there isn’t the slightest hope we’ll ever assemble a proper bicycle wheel. The upshot is that a thin particular t can only unify the property constituents of a thing if its ontic attachments (properties) are grounded in the intrinsic nature of that thing. This nature can’t be external. For the properties composing it would then be externally tied to t, so that t would already have to possess an attachment grounding nature. This nature can’t be internal either, since then t would be a complex entity and not a bare (internally simple) particular. We are therefore left with no explanation for how concrete particulars ‘hold together’, as it were: how properties manage to insert and stay connected to natureless ontic hubs. It is difficult to see, in any of this, just where the stable (extra-mental) foundation is on which Preston invites us to construct property instances. 3.4 The ‘Swapping’ Problem As we learn from Gustav Bergmann, "Any two [bare particulars] are not intrinsically but only numerically different. That is their bareness" (1967, 24). Moreland puts it this way: bare particulars "simply come individuated" (2001, 155). And that’s all. But how can this be? According to Sider, for example, Truly bare particulars do have natures...Indeed they all have the same nature, and that nature is exhausted by the fact that they instantiate no monadic universals. That is the way that they are. ‘‘What is a truly bare particular like?’’ Answer: "It is not charged. It does not have any mass. It does not have any spin. And so on." (2006, 392) 58 Unlike Moreland, Sider holds that bare particulars can exist apart from any property attachments; they "can stand on their own" since they are "a fundamental ontological category" (ibid). But here we strike a problem. For if all BPs "have the same nature," then there doesn’t appear to be anything about this or that BP that would preclude it from being swapped with one of its fellows. But if not, does it really make sense to speak of Socrates’ BP? What accounts for its being the BP of Socrates? Nothing in its nature; it has none. Nor can this very BP—S(b), as we dubbed it earlier—be Socrates’ in virtue of the properties tied to it essentially; for these are presumably shared with everything else of the same kind—that is, every other spot. What about Socrates’ contingent property ties? Do they account for S(b) belonging to Socrates? Again, it would seem not. For let P stand for the conjunction of all Socrates’ non-essential properties: being red, being round, and the like. Then consider any bare particular distinct from S(b). Couldn’t P have been tied to that BP instead of S(b)? This certainly seems like a possibility.6 But then if S(b) belongs to Socrates just because P is one of its attachments, being tied to P is what individuates S(b). It is therefore necessary that any bare particular tied to P belong to Socrates. And since all BPs ‘‘have the same nature,’’ they all could have been tied to P, in which case each and every BP is such that it is Socrates’ BP. Accordingly, if S(b) really does individuate Socrates, the world consists of a single red, round, spot! You might object that our conclusion is somewhat overdrawn: not all BPs could have been tied to P. What if a BP’s essential ties preclude certain contingent attachments? For example, if Socrates’ BP has spothood tied to it essentially, then obviously it can’t take being a prime minister as an attachment—for the simple reason, of course, that no spot could be a prime minister. This is true but presently irrelevant. Our initial question was not whether qualified BPs could be swapped with one another salva distinguo, that is, without altering the distinction between the particulars containing them. The question, rather, is whether BPs simpliciter can be intersubstituted in this way. Since they ‘‘all have the same nature,’’ there is every reason to believe they can, in which case they are impotent individuators if they are individuators at all. Moreover, even if we follow Moreland in denying of BPs such allegedly essential properties as simplicity and particularity,7 it is arguable that each BP will still have at least one positive essential property—a kind defining property, let’s say (e.g., being a spot). Surely this is reasonable. Surely all spot-BPs will be necessarily connected to spothood. You’ll never find one without the other: nor could you. But if that is the case, then wouldn’t all spot-BPs have precisely the same essential properties qua spot-BPs? And of course their underlying nature would be precisely the same, namely, being per se without property attachments. So really, on this way of thinking, there should only be one spot, one dog, one person, one book, and so on. ________________________ 6 Even more so, if Moreland is right and all ‘ties’ to BPs are contingent. On his view, e.g., "bare particulars actually have no necessary properties...the admission that bare particulars have some properties necessarily is mistaken" (Moreland and Pickavance 2003, 8). 7 Thus Moreland and Pickavance: "We believe that the properties said to be necessary for bare particulars are not genuine properties; these include simplicity, particularity, unrepeatability, and those of the three categories of transcendental, disjunctive, and negative properties" (Ibid, 10). 59 Perhaps what the objector is really trying to say, here, is only that being tied to P isn’t sufficient, doesn’t guarantee, that a given BP belongs to Socrates. For the purposes of individuation, it is enough, rather, that any world in which S(b) exists is one in which it is tied to a unique but contingent array of properties. But the thing to see is that this array needn’t be the same from world to world. So perhaps in the actual world, it is being tied to P that individuates Socrates’ BP; in another world w, however, a wholly different set of property ties does the job. Still, as long as that tie is distinct from each of its fellows in w, it looks as though there should be no problem in saying that, in w, S(b) belongs to Socrates. We think this reply leaves a good deal to be desired. In essence the objector is saying that individuation of BPs is not only derivative but also a world-relative affair: having P simpliciter does not individuate S(b), but having P in a does (where ‘a’ rigidly designates the actual world). Socrates’ BP isn’t Socrates’ because it is attached to P, but rather because it is connected to the world-indexed, one-owner property P-in-. To borrow Plantinga’s language, P-in-a is an essence of Socrates;8 for in every world in which Socrates’ BP exists, it is tied to this property. Furthermore, it isn’t so much as possible for any other BP to have this property as an attachment. But then it immediately follows that having P-in- individuates S(b); being attached to this property is a unique mark of distinction. In short, we’re looking not to BPs to do our individuative work but to WOPs (world-indexed, one- owner properties).9 What place then for the bare particular, ‘‘the grossest of metaphysical errors’’ (Sider 2006, 392)? None so far as we can see. 4 Simple Aspected Particulars 4.1 How to Dress Up a Bare Particular Taken as either individuators or unifiers, then, bare particulars are beset by serious difficulties. Does it follow, however, that we should abandon substratum theory altogether and opt for some bundle theory of concrete particulars? Not necessarily. What’s needed here is a more robust and stable ontic subject: something whose nature is neither tied to it externally, nor included among its internal constituents. Thus Wolterstorff: And now for the question: why should there not be a certain entity which, like everything else, just is a certain nature, but which, unlike most or all other entities, is nothing more than that—is not a composite?...It will be just a certain something-as-such, a certain what-it-is-as-such. That would be an extraordinary entity. We would know next to nothing about what it would be like to be such an entity. But there seem to be no ontological difficulties in the proposal that there is such an entity. (1991, 543) _______________________________________ 8 See Plantinga (1974, 70). 9 See Davis (2002, 2004b) for brief discussions of WOPs and their role as individuators. 60 Borrowing a chapter from Mertz, let’s say that a simple entity is one that has no internal constituents at all. Well, why couldn’t there be individuating unifiers—not particulars with natures—but things that just are natures: simple, concrete particulars. Wolterstorff claims that "next to nothing" would be known about such a thing. But perhaps not. In order for them to be ‘natured’, we might say that they possess discernible (metaphysical) aspects. How something could possess multiple aspects and yet remain internally simple could then be explained as follows. Consider an ordinary circle . Although this figure is undeniably simple in that it is continuous and unbroken, it appears that we can discriminate between at least two of its aspects (curves): ⁀ and ‿, let’s say. These aspects are "‘in’ the circle" (Mertz 1996, 75), but are separable only by an act of mental abstraction, that is, by forming a mental image of the circle’s being divided in half. But surely this in no way implies that in reality the circle is a complex whole, consisting of two connected halves. The separability of the halves here is purely epistemic. Now consider our friend Socrates once more. On the proper realist view of things, the proposition (8) Socrates is a spot actually entails (8*) Socrates’ BP is (essentially) tied to the property being a spot which has the effect of ‘stripping down’ the ontic subject to dangerously minimal proportions (as we have already seen). On our proposal, this is wrong-headed; what (8) entails is not (8*) but rather (8**) Socrates has being a spot as an ontically inseparable aspect where ‘Socrates’ denotes Socrates qua simple aspected particular (hereafter, SAP). Socrates isn’t a thing with a nature; it is a concrete nature.10 The property-universal being a spot is in Socrates as an ontologically inseparable but epistemically discernible aspect. Thus (8) invokes neither ‘rooted in’ nor ‘tied to’ predication. The ‘is’ at stake in (8) is rather one of aspectival ontic inseparability, which nicely captures the intuitive notion behind what we normally think of as essential property possession. What advantages are there to recognizing SAPs in one’s ontology? Well for one thing, a SAP clearly has regulatory functions; it places restrictions on the nature of its ontic attachments. Since it is a nature, it will preclude arbitrary attachments (e.g., being round and square). For instance, consider _________________________ 10 According to one of the reviewers, we must provide "empirical examples" of SAPs in order to justify including them in our ontology. But this strikes us as little more than an a priori empiricist stipulation, which we are entirely free to reject. You might as well argue contra Aristotle that substances aren’t form- matter complexes on the grounds that he doesn’t give us any "empirical examples" of forms. But this misses the point. Aristotle’s distinction between form and matter isn’t empirical; rather, it is the product of ontological assay. To be sure, what we empirically detect are substances qua wholes; it hardly follows, though, that each ontological constituent of a substance must be separately detectible. For more on the deficiencies of empiricist constraints in analytic ontology, see Moreland (2001, 150-151). 61 (9) Socrates is red. According to the SAP ontologist, (9) tells us that (9*) Socrates is (contingently) tied to the property redness. The property of being red isn’t an ontologically inseparable aspect of Socrates; it’s something it could perfectly well do without—and still go on existing. (And so things will go with all of Socrates’ contingent properties.) Notice, however, that because Socrates is a multi-aspected concrete nature, it qualifies as a suitable ontic ‘hub’ for property ‘spoke’ insertion. Since (8**) is true, e.g., Socrates can accommodate redness or roundness as attachments, but not of course being prime or being a principal. Thus we deftly circumvent Mertz’s complaint that if the individuating unifier of a thing is bare, then it is bereft of essential properties, so that it hasn’t a nature and any property you please may be tied to it. This is not so. It is quite true that only contingent properties can be tied to Socrates; it doesn’t follow that all Socrates’ properties are contingent—not, at any rate, if there are SAPs. Secondly, does our proposal help with the ‘Swapping’ Problem? Well, it depends on what we allow to count as aspects. Obviously, not every essential aspect of a SAP will serve to distinguish it from its fellows. For example, every SAP has such trivial essential aspects as being colored if blue, being even if identical with the number 6, and perhaps even existing.11 But these aspects hardly distinguish one SAP from the next, if they are had by all (and had essentially). Moreover, even the aspects essential to being a SAP—e.g., being simple, having aspects, and being a particular—will be of no use here, since they too are held in common with all such particulars. Thus if we have only trivially essential and SAP essential aspects at our disposal, we will be at an utter loss to generate any real distinctions between simple aspected particulars. So this is not the way of true individuation. Fortunately, there is a way forward. The SAP ontologist, we suggest, should consider adding to her ontological arsenal a special sort of aspect—one that is essentially unique to a SAP, that is, one such that without it this very SAP could not exist, but also such that no other SAP could possibly have it. What she requires, in other words, is one or more ‘haecceitistic’ aspects: aspects connected with a thing’s self-identity (e.g., being identical with Socrates) or perhaps even its world-indexed one-owner properties (e.g., being Gustav’s favorite spot in a).12 As long as all SAPs have aspects of this sort, they will count as wholly distinct concrete thisnesses, in which case the very idea of ‘swapping’ SAPs becomes unthinkable. 4.2 A Better Way: SAPs and Simples So what we have so far is that simple aspected particulars can be seen to succeed where bare particulars fail. SAPs meet the grade as proper ontic ‘hubs’ and can be ______________________ 11 See Plantinga (1974, 60-62). 12 This is not the place for an extended defense of haecceities. For detailed discussion, see Rosenkrantz (1993). 62 outfitted to perform as individuators. To our mind, we have arrived at a far more adequate foundation for constructing Preston’s property instances (PINs) than that provided by those embarassing bare particulars. But what about his original worry: that a PIN such as red1—Socrates’ redness—can’t be "genuinely related" to Socrates, since it includes something already present in Socrates: one and the same bare particular? Does our proposal help allay this concern? Not exactly. But then Preston’s concern strikes us as misplaced to begin with. We can see this if we ask ourselves why Preston believes the distinction between a concrete particular and its property instances isn’t in fact real but only mental in nature. His answer goes like this: if there were a real distinction between Socrates and red1—that is, if they were distinct but related (extra-mentally real) entities—then they would be "capable of independent existence" (Preston 2005, 284). But they’re not. Hence the distinction isn’t real, and one of these entities is nothing but a "mental construct." We think this is at best a weak argument for PIN constructivism. For example, what reason is there to think that Socrates and red1 aren’t capable of independent existence? Preston’s reply is that even in our thinking "there is a necessary overlap between a quality (property) instance and its thicker particular for the simple reason that they share a part—namely, the same bare particular" (285). But that doesn’t prove a symmetrical existential dependence. Think of some particular cell that is part of your appendix. That cell is also part of you. So you and your appendix share a part. Does that mean that you can’t exist without your appendix? Obviously not. But then sharing a part doesn’t prove that two things depend on each other for their existence. Nor can we see how the "inseparability" of Socrates and red1 is "just as unavoidable in the realm of cognitio"’ (ibid). Each of us is perfectly capable of thinking of Socrates without thinking of Socrates’ redness. All that’s needed is some modal imagination. Just think of a possible world in which Socrates isn’t red at all, but rather blue. Then you will be thinking of a state of affairs in which Socrates exists but red1 does not. In other words, you can avoid thinking of the one, while thinking of the other. That means that Socrates is both existentially and epistemically independent of red1.13 Preston’s basic reason for thinking that "it seems most correct to say that a mere mental distinction obtains between a quality instance and its thicker particular" (ibid) is therefore flawed. The argument is unsound. What has gone wrong in all this? The root problem, we believe, is that we have fallen into the grips of a (false) dilemma: believe that red1 is a constituent of Socrates, or believe that red1 is a mental construction upon Socrates. The first alternative leads to the redundancy problem; the second tilts toward PIN anti- realism. The thing to do, of course, is to split the horns of the dilemma. But how so? The key lies in getting clear about our ontological categories. Just what is red1 supposed to be? We have been referring to it as a ‘property instance’—what Williams calls a ‘trope’ (1953) and E. J. Lowe a ‘mode’ (1998). Now most _________________________ 13 Naturally enough, however, Socrates’ property instances will depend on Socrates in both of these ways. Still, this isn’t sufficient to prove ‘‘inseparability.’’ For that it must be shown that neither can do without the other. 63 philosophers consider tropes or modes to be concrete particulars. Thus Lowe says, ‘‘Modes of concrete entities are themselves concrete entities, existing in space and time’’ (1998, 78). But is this really so? Our ground-level intuition about modes is that they are ways something is.14 Now some modes, of course—Wiles’ being a mathematician, for example—are actual; they obtain. Others such as Wiles’ being a jazz musician are non-actual and fail to obtain. And yet surely Wiles could have been that way; he could have ignored the joys of solving Fermat’s Theorem to become a premier smooth jazz guitarist. This isn’t impossible. So in addition to his actual modes, we should no doubt recognize his other possible modes of existence as well. To be sure, these modes will be non-actual; but that doesn’t mean they won’t exist, as they will all possess the modal property of being possibly actual. We can now put our point as follows. Whatever else modes may be, they’re not concrete objects; they’re more like abstract states of things. A way Wiles could have been (but isn’t) is not a concrete, physical thing at all. To insist otherwise reveals a failure to distinguish between ‘ways things are/could have been’ and the ‘things that are/could have been that way’.15 The latter are physical entities (numbers, sets, and the like excepted); the former are not. But if this is so, then modes such as Socrates’ being red are not necessarily subject to the proper realist’s tri-partite assay: universals bound to bare individuators by the tie of exemplification. We can treat them instead as simple states of affairs, neatly avoiding both the redundancy problem (since they don’t contain BPs) and constructivism (since there will be too many modes to be explained by human cognitive activity). On this view, it is relatively easy to state how a mode is related to the particular thing of which it is a mode. We can say that a mode such as Socrates’ being red obtains (/is actual) just in case Socrates includes a SAP that exemplifies the universal redness. Here someone will object as follows. ‘‘You say that Socrates’ being red is a mode of Socrates, one of his tropes. But in fact it is a states of affairs; it’s not a property instance at all. Talk about Socrates’ red trope is talk about Socrates’ redness not his being red. These are quite different things.’’16 Here our reply is twofold. First, there are some trope theorists who do allow a thing’s having a property (e.g., Socrates’ being red) to count as a trope.17 Secondly, if tropes are ways things are or could be, as Armstrong and Lowe suggest, then Socrates’ redness isn’t a mode of Socrates; like Socrates, it is a thing itself: something that is or could be one particular way or another. What we’re contemplating here, presumably, is a spatially located, physical instance of a property. There isn’t space here to go into the lengthy details; however, we might note that even if we could somehow manage ___________________ 14 According to David Armstrong, "there is nothing to bar, and much to recommend, treating property tropes and relation tropes as ways" (1997, 30, emphasis added). Compare also Lowe: "A mode is a particular way something is" (1998, 78). 15 This is noted in van Inwagen (2001, 169). 16 According to William Mann, for example, "all states of affairs exist, but only some of them obtain or are actual. This feature does not hold for property instances. In order for a property instance to exist, it must be actual: some existing thing must either exemplify it or be it" (1982, 457). 17 See Bacon (1995, 1, 4). Bacon permits both ‘x’s having F-ness’ and ‘the P of x’ to make successful reference to tropes. 64 to make sense of this proposal for (monadic) property instances, it’s extraordinarily difficult to see how it could work for relation instances.18 Take, for example, the specific instance of a’s being between b and c. Following standard convention, let ‘Betweeni (a, b, c)’ stand for that instance.19 Now on the present objection, this instance, like Socrates’ redness, is a concrete thing with a spatial location. But it is extremely difficult to see where this might be. It cannot be located in the mereological structure a-b-c (or its parts), since that structure (and its parts) are spread out and have spatial dimensions.20 But it scarcely makes sense to say that the betweeness of a with respect to b and c has a specific height, width, and depth. Nor can we locate this instance throughout the region of space between b and c. For how could Betweeni (a, b, c) possibly be between b and c when, quite obviously, it encompasses those relata? You might as well argue that a football field (which includes both end zones) is located between its end zones. The fact of the matter is: a thing can be between other things; relations and relation instances cannot. The wisest course of action is therefore to treat a’s being between b and c as a simple, abstract way these particulars can be related.21 We must resist reifying this state of affairs into a concrete, physical thing that (like Socrates) can exist in various sorts of ways. There is obviously much more to be said on this score. But for now, perhaps, we have said enough to shed at least some doubt on the dogma that PINs are spatially localized properties. In conclusion, then: there are strong theoretical advantages for replacing bar particulars with SAPs. Bare particulars are poor ‘‘pincushions’’ and impotent individuators. Not so for simple aspected particulars. Moreover, if we eschew Preston’s constructivism about PINs, taking them to be simple, non-physical mode instead, we avoid the redundancy problem while side-stepping the perils of PIN anti-realism. So there appear to be good reasons for rejecting Preston’s view an none for accepting it. Why, then, should anyone want to be a property instance constructivist? References Armstrong DM (1997) A world of states of affairs. Cambridge University Press, Cambridge Bacon J (1995) Universals and property instances: the alphabet of being. Blackwell, Oxford Bergmann G (1967) Realism. University of Wisconsin Press, Madison ______________________________ 18 There is a serious but, we think, unsuccessful attempt to defend a one-category ontology of relation instances in Mertz (1996, 2001, 2002). For extended critique of Mertz’s program, see Davis (2006). 19 The subscript ‘i’ indicates that we are considering an instance of being between. 20 See Grossmann (1992) for an expanded discussion of this point. 21 This should make it clear that we are not advancing a species of trope theory. For on all accounts, tropes are concrete (particularized) instances of properties. On our view, however, modes are abstract particulars. They are abstract in the sense that they are non-spatio-temporal objects. But they are also particulars given that a mode such as Socrates’ being red doesn’t obtain (/fail to obtain) with respect to multiple particulars, but only Socrates. Keith Campbell (1990) has also advocated what he calls ‘abstract particulars’. However, by ‘abstract’ Campbell simply means ‘grasped by an act of mental abstraction’. For Campbell, abstract particulars are spatio-temporal concrete tropes, and thus not to be confused with modes qua abstract particulars. 65 Bradley FH (1946) Appearance and reality: a metaphysical essay. Clarendon Press, Oxford Campbell K (1990) Abstract particulars. Basil Blackwell, Oxford Davis RB (2006) Is Socrates a predicate? Metaphysica 7:5-18 Davis RB (2004a) The brave new bare particularism. The Modern Schoolman 81:267-273 Davis RB (2004b) On individuators: a reply to Timothy Pickavance. Philosophia Christi 6:111-120 Davis RB (2003) ‘Partially Clad’ bare particulars exposed. Australas J Phil 81:534-548 Davis RB (2002) Haecceities, Individuation, and the Trinity: A Reply to Keith Yandell. Relig Stud 38:201-213 Grossmann R (1992) The existence of the world. Routledge, London and New York Lowe EJ (1998) The possibility of metaphysics: substance, identity, and time. Clarendon Press, Oxford Mann W (1982) Divine simplicity. Relig Stud 18:451-457 Mertz DW (2003) Against bare particulars: a response to moreland and pickavance. Australas J Phil 81:14-20 Mertz DW (2002) Combinatorial predication and the ontology of unit attributes. The Modern Schoolman 79:163-97 Mertz DW (2001) Individuation and instance ontology. Austral J Phil 79:45-61 Mertz DW (1996) Moderate realism and its logic. Yale University Press, New Haven Moreland JP (2001) Universals. McGill-Queen’s University Press, Montreal & Kingston Moreland JP (1998) Theories of individuation: a reconsideration of bare particulars. Pacific Phil Quart 79:251-63 Moreland JP (1991) How to be a nominalist in realist clothing. Grazer Philosophische Studien 39:75-101 Moreland JP, Pickavance T (2003) Bare particulars and individuation: reply to Mertz. Australas J Phil 81:1-13 Plantinga A (1974) The nature of necessity. Clarendon Press, Oxford Preston A (2005) Quality instances and the structure of the concrete particular. Axiomathes 15:267-292 Rosenkrantz (1993) Haecceity: an ontological essay. Kluwer, Dordrecht Sider T (2006) Bare particulars. Phil Persp 20:387-397 van Inwagen P (2001) Ontology, identity, and modality: essays in metaphysics. Cambridge University Press, Cambridge Williams DC (1953) The elements of being. Rev Metaphys 7(3-18):171-192 Wolterstorff N (1991) Divine simplicity. Phil Persp 5:531-552 ***** This is the end of the e-text. This e-text was brought to you by Tyndale University, J. William Horsey Library - Tyndale Digital Collections *****