What Can Global Pragmatists Say about Ordinary Objects?

What can Global Pragmatists Say about Ordinary Objects? Amie L. Thomasson Forthcoming in Javier Cumpa and Bill Brewer, eds. The Nature of Ordinary Objects (Cambridge: Cambridge University Press ) Abstract: Neo-pragmatist approaches have been making a comeback lately in various local debates. Expressivist approaches to moral discourse have drawn increasing attention, and allied non-representational views of modal, logical, and even epistemic discourse have also been developed. Huw Price has argued for a form of global pragmatism, applied to all areas of discourse. But other defenders of local pragmatisms (such as Simon Blackburn) have denied that pragmatism can be extended globally, in part because it cannot be extended to everyday talk about ordinary objects. Here I examine the question: can a broadly pragmatist approach be extended to cover talk about ordinary objects? If so, what would this form of pragmatism look like, and what would the consequences be for ontology and metaontology? Interest in neo-pragmatist approaches has been on the rise. The form of pragmatism I will discuss here is not the classical pragmatism of such figures as Peirce and Dewey, but rather the kind of neo-pragmatism that has been more recently developed by such figures as Huw Price, Robert Brandom, and Michael Williams. As Michael Williams (2011, 318) puts it, following Huw Price, this form of pragmatism is distinguished by two features. First, a commitment to linguistic priority that says, when investigating classic philosophical problems about an area like the moral, don’t begin by asking metaphysical questions about what moral properties or facts are, or even if there are any. Instead, begin by asking about the discourse: what does moral discourse do for us? Why would we want to have moral terms in our vocabulary? What are we doing when we moralize? (Williams 2013, 128). The second distinguishing feature of a pragmatist approach is what Williams and Price call ‘anti-representationalism’. This involves the idea that we should drop the assumption that the meanings of the terms in question must be given in terms of the items they refer to (or, more broadly, the idea that we must explain meaning in terms of semantic world-word relations (Price 2011, 233 and Williams 2013, 128)). It’s easy to see the relation here—for if we begin by asking about the function of a certain area of discourse (rather than asking metaphysical questions about the things allegedly referred to by it), we may find at least in certain cases that the discourse functions otherwise than to track or represent a certain range of entities. This then gives us reason to drop the assumption that the meanings of the relevant terms (say, moral terms) must be given via the moral properties or facts they refer to, and to look for other accounts of meaning—say, in terms of other functions the terms serve and the rules they follow that enable them to fulfill those functions. Perhaps the best-known and best-developed pragmatist positions are those in the expressivist tradition in ethics, defended and well developed by figures like Simon Blackburn (1993) and Alan Gibbard (1990). Others have recently defended forms of pragmatism about modal discourse (Brandom 2008, Thomasson 2007 and forthcoming, following Sellars 1958 and Ryle 1949) or about logical notions like 1 logical consequence (Restall 2012), truth (Horwich 1998) or epistemic notions (Chrisman 2007). Local forms of pragmatism are attractive since they promise to help us avoid long-standing ontological and epistemic problems. On the ontological side are what Price calls ‘placement problems’ about how such odd entities could find their place among the natural furniture of the world. As I have tried to make clear elsewhere (2015, 136-8) the expressivist needn’t deny that there are moral properties or modal facts, in the only sense these terms have. For instead she may begin by arguing that modal talk serves a fundamentally non-descriptive function (such as conveying constitutive semantic rules in particularly useful ways), and yet go on to show how the rules governing modal talk license us to make hypostatizing moves introducing noun terms that then entitle us to speak of modal facts and properties. What we can get out of these hypostatizing moves (I have argued (2015)) is simply realism about the relevant entities—not any kind of reduced or quasi-realism. Nonetheless, the modal properties we can speak of are not ‘posited’ as ‘explanatory’ analogs of scientific properties, and so we don’t need to explain their relationship to the natural properties of the world. Nor do we need to appeal to them in order to explain our modal talk. We also get new hope of avoiding the traditional epistemic problems about how we could come to know what the moral or modal truths are. For we don’t have to think of our modal claims as aiming to describe properties to which we lack empirical access. Instead, we can give a different account of our ability to know modal facts (perhaps, an account that goes via extrapolation from our conceptual competence). For the remainder of this paper, I’ll leave aside questions about the development or plausibility of these local forms of pragmatism. For I’m interested instead in the prospects of a more global form of pragmatism, and in particular whether discourse about ordinary objects forms a barrier to it. Some, prominently including Price himself (2011, 2013), have defended a form of global pragmatism. Yet even some prominent defenders of local forms of pragmatism, such as Simon Blackburn (2013), have raised grave doubts about whether one can and should accept a form of global pragmatism.1 For even if we can develop plausible forms of local pragmatism to address moral or modal discourse, the sticking place seems to be everyday discourse about ordinary objects—the very sort of discourse early expressivists sought to contrast with moral discourse. As Ted Sider suggests in another context, “an expressivist semantics for discourse about bachelors, cups, and the rest is clearly a nonstarter” (2011, 46). The first reason Blackburn gives for resisting global pragmatism is what (following Robert Kraut) he calls the ‘no exit’ problem: that the pragmatist’s genealogical stories about how troublesome areas of discourse are introduced apparently rely on an unquestioned bedrock of common sense ordinary discourse— 1 There may be room for independent arguments that those who are pragmatist with respect to the modal should also be pragmatists with respect to ordinary objects—as the differences between ontological views that do and do not ‘posit’ ordinary objects are plausibly modal differences—differences in views about what modal profiles are instantiated, what identity and persistence conditions things have, etc. But I will leave that to the side here. 2 which then seems insusceptible to a similar treatment (2013, 78). In fact, it might seem then that the only explanation to be given of why we go in for ordinary object talk is because there are these ordinary objects surrounding us in the environment. Second, there are (as Blackburn points out) ‘huge asymmetries’ between our talk of common sense objects and talk of the moral, modal, or mathematical (2013, 82). And this, in turn, might give the expressivist reason to accept a bifurcation between common sense talk and talk in these other areas—leaving us again with merely localized forms of expressivism. So the question to be addressed here is: what can pragmatists say about discourse about ordinary objects? Is there a plausible story to be told, which can avoid the ‘no exit’ problem and preserve the apparent differences between talk of ordinary objects and talk of moral facts that originally motivated local expressivists? I will begin by outlining what a global pragmatism would be—how we should understand it. I will then turn to outline Blackburn’s reservations about adopting a global form of pragmatism before showing how we can address them and defend a form of global pragmatism that addresses these reservations and respects the crucial differences across different areas of discourse. In closing I will discuss what follows from this sort of global pragmatism, and why applying a broadly pragmatist view even to discourse about ordinary objects might be attractive. 1. What would a global pragmatism be? To those exposed only to local debates, say, about expressivist versus descriptivist approaches in ethics, the very idea of a global pragmatist approach might be hard to fathom. For local versions of the approach are often introduced by contrasting supposedly expressive (say, moral) discourse—that (say) serves to express the speaker’s attitudes rather than to describe the world, with the apparently descriptive discourse used in ordinary or scientific discourse—say, when we say how many cups are on the table, or how many electrons are in a carbon atom. So what would pragmatism, writ large, even mean? Simon Blackburn, one of the original and chief proponents of an expressivist approach to the moral, characterizes pragmatism as follows: You will be a pragmatist about an area of discourse if you pose a Carnapian external question: how does it come about that we go in for this kind of discourse and thought? What is the explanation of this bit of our language game? And then you offer an account of what we are up to in going in for this discourse, and the account eschews any use of the referring expressions of the discourse… or any semantic or ontological attempt to ‘interpret’ the discourse in a domain, to find referents for its terms, or truth-makers for its sentences (2013, 75) For if we hope to get a different line of explanation that avoids some of the traditional metaphysician’s problems, the explanation of the relevant sort of talk must say why employing a language or conceptual scheme like this is useful for us, and yet do so without appealing to the supposed objects we are talking about. The pragmatist approach has Carnapian roots. From an ontological point of view, I take the basic insight of a pragmatist approach to be Carnap’s: that we don’t 3 require an ontological justification for introducing a new form of speech. Carnap describes the problem with external questions as follows: Many philosophers regard a question of this kind [a philosophical question about the existence or reality of a system of entities] as an ontological question which must be raised and answered before the introduction of the new language forms. The latter introduction, they believe, is legitimate only if it can be justified by an ontological insight supplying an affirmative answer to the question of reality. In contrast to this view, we take the position that the introduction of the new ways of speaking does not need any theoretical justification because it does not imply any assertion of reality (1950, 214). I’d actually like to revise the last part a little. The point I want to preserve is the idea that the introduction of a language form does not require ontological justification, or involve ontological presuppositions. So, for example, we do not need to first do a metaphysical check, to see if there really are numbers, in order to be justified in introducing nominative number talk. Instead (in this case), we may introduce the talk of numbers in a way that guarantees that our new noun terms for numbers refer, and so that we are entitled to say there are such things. (But notice this is emphatically not to say that numbers—rather than our talk of them and commitments to them—are created by or dependent on our language or thought, nor is it to deny that there may be empirical presuppositions behind introducing some sorts of terminology. More on this below). If, for a given group of terms, the only explanation we could give of why we make use of them, and why these particular rules govern their use, is to say: We use terms for Xs because there are Xs, then it seems we must begin from metaphysics: assuring ourselves that there are Xs to justify our use of X-talk, and understanding the meaning of X-talk in terms of the Xs to which it refers. We would, in short, have to be representationalists about the language involved, and engage in a metaphysics-first approach that failed to pursue the pragmatist’s linguistic priority approach. The pragmatist’s aim, by contrast, is to give a pragmatic explanation for the introduction (or retention) of the relevant discourse without having to cite an ontological justification of the sort that would say: we go in for it because it represents the correct ontology, the objects or properties there really are. So, for example, a pragmatist might (with Horwich (1998)) explain why we go in for truth talk by appealing to the role the truth predicate plays in enabling us to express generalizations; or might (following Yablo’s (2005) analysis, though he himself is no pragmatist) explain the introduction of number talk as a way of enabling us to state certain laws in finite form. The introduction of these useful forms of speech might then entitle us to talk about numbers or truth, and even to say that there are numbers or true propositions; but it doesn’t rely on a prior ‘insight supplying an affirmative answer to the question of reality’. The idea of global pragmatism, then, is to take this approach across the board: denying that we ever need an ontology-first approach to saying what justifies introducing or retaining the relevant linguistic forms (of the form: “we go in for Xtalk because there really are Xs). (This is also a way of avoiding the Representationalism Price rejects). As I understand it, Carnap’s approach was global 4 in this way: the view was that we don’t ever, (not even for the thing language) require ontological justification for introducing a form of speech. 2. The No Exit Problem Simon Blackburn, one of the iconic proponents of forms of local expressivism, has recently raised worries about taking the pragmatist approach globally—worries that, I suspect, echo suspicions felt in the wider philosophical community. Perhaps the biggest source of resistance to adopting a global expressivist approach arises from considering our ordinary, common sense talk about ordinary objects like tables and chairs and the like. The first worry Blackburn raises about embracing a global pragmatism is what he calls the ‘no exit’ problem, that, “…even genealogical and anthropological stories have to start somewhere” (2013, 78). So, for example, genealogical stories about the origin of moral discourse might (with Hume) take off from talk of “natural propensities to pain and pleasure, love and hate, and an ability to take up a common point of view with others”; a Fregean story about mathematical discourse “would start by placing us in a world of kinds of objects with distinct identity conditions, such as tigers and eggs and warriors, and then a capacity to tally them” (2013, 78). So genealogical stories about suspect forms of discourse, like the moral or mathematical, begin by showing how such discourse may be introduced from an unquestioned ‘common-sense background’ of ordinary discourse. But then the question we are left with is whether we have any hope of giving a similar genealogical story about that common-sense discourse itself. Blackburn doubts that can be done: If we insisted… on posing the Carnapian external-sounding question, how come that we go in for descriptions of the world in terms of surrounding middle-sized dry goods?, then the answer is only going to be the flat-footed stutter or self-pat on the back: it is because we are indeed surrounded by middle-sized dry goods. (2013, 78-9) But this involves apparently giving up the pragmatist line for common sense talk: in that case, we must appeal to the ontology itself (‘because we are surrounded by middle-sized dry goods’) to justify or explain why we employ the relevant linguistic forms (our terminology for ordinary objects)—apparently returning to a Representationalist reading of that area of discourse. By contrast, in other cases, “there is every prospect of bracketing the existence [of the relevant entities] and coming to understand why we go in for the mode of thought in question in other terms. In other words, there is every prospect of giving an anthropology or genealogy which is itself free of the commitments in question” (2013, 83). The key challenge Blackburn has presented here is to explain why we go in for descriptions in terms of middle-sized dry goods, without appealing to that very ontology. As has often been argued in the literature on contrastive explanation (Garfinkel 1981, van Fraassen 1980), proper explanations must appeal to a contrast case: e.g. we will get different explanations if we ask why Judy bought the car (rather than the truck) versus why she bought the car (rather than leasing it). In this case, I think we can identify three different questions, or demands for explanation, in the vicinity. When we ask ‘Why do we go in for descriptions of the world in terms 5 of surrounding middle-sized dry goods?’, we could be asking this with at least three different contrasts in mind: 1. Why do we go in for descriptions of middle sized dry goods, as opposed to not employing descriptive modes of speech at all? 2. Why do we go in for descriptions in terms of, say, tables and trees, as opposed to descriptions, say, in terms of dragons and phlogiston? 3. Why do we go in for descriptions in terms of tables and trees, as opposed to in terms of particles arranged table-wise, in terms of its tabling here, or in terms of sequenced temporal parts rather than enduring objects—that is, why do we employ a thing-language instead of employing some ontologically alternative language? Each of these questions would demand a different sort of explanation. But once the question is clarified in this way, I think that whichever way we read the demand for explanation, the pragmatist has an available line of response— one that does not simply appeal to the ontology in question. Price, following Brandom, sketches a response to the first explanatory demand, ‘How come that we go in for descriptions at all?’ He suggests that for the pragmatist, there is still a story to be told here; the pragmatist’s answer to the general question he suggests, is no less relevant here than in other cases—and might be addressed by, for example, a Brandomian account of the function of assertive talk in general (2013, 159)—an account that doesn’t make use of Representationalist presuppositions. While that seems apt as a response to the first explanatory demand, one might suspect that other explanatory demands remain: not of why we go in for assertions or descriptions at all, but of why we go in for a conceptual scheme that makes use of these particular terms or concepts (regardless of whether we use it in coordinating our commitments, expressing our attitudes, or whatever). Why do we go in for using concepts and terms for medium-sized enduring objects like tables, chairs and trees? For the Representationalist metaphysical realist might still suggest that there must be a reason why we employ this particular conceptual scheme rather than various alternatives; perhaps because these are the things there really are, so that it becomes most important to coordinate our attitudes or practical activities about things of these sorts. But here again there are at least two ways of understanding the (remaining) explanatory demand: (2) why do we make use of concepts for tables and tress rather than, say, concepts for witches or phlogiston or dragons? Or (3) why do we go in for concepts of ordinary objects such as tables and trees and other middlesized dry goods, rather than empirically equivalent, but ontologically distinct, concepts of arrangements of simples, features, temporal parts, or the like? What’s the difference between (2) and (3)? As I have characterized it, (2) involves asking why we make use of the particular terms we do—rather than some (failed) alternatives. It’s when we think of the demand in terms like (2) that we are most tempted to say: because these are the things there really are! But (2) pretty clearly isn’t what Blackburn has in mind. Witches and dragons (if there were any) would also be medium sized dry(ish) goods, and the question seemed to be why we go in for a language like this at all—rather than why we retain certain purported species and kind terms and jettison others. 6 By contrast, (3) involves not contrasting our particular middle sized dry goods terms with others we’ve (mostly) rejected, but rather contrasting the ‘thing’ language at large with what we may call an ‘ontologically alternative’ language. Interestingly, the very possibility of what I call ‘ontologically alternative’ languages has been made clear in post-Quinean ontological debates—and indeed several different ontologically alternative languages have been introduced by ontologists who aim to eschew commitment to ordinary objects. I don’t aim to give a strict set of necessary and sufficient conditions for one language being ‘ontologically alternative’ to another, but instead to just point to the practices and goals of revisionary ontologists who have introduced these new ways of speaking. Those who deny the existence of ordinary objects typically introduce an ‘ontologically alternative’ language, holding that their languages involve distinct ontological commitments, but are equally able to capture our experience and don’t involve any empirical differences from their rivals. Having such ontologically alternative languages to hand is useful to make good on the idea that a revisionary view is “not absurd and …not at variance with Universal Belief” (van Inwagen 1990, 106). So there are three goals in restating claims of one language L1 (in this case, our ordinary object language) with those of an ontologically alternative language L2: 1) to avoid the apparent ontological commitments of the claims in L1 (L2 has different apparent ontological commitments—or would have different commitments according to a Quinean criterion), while 2) still enabling us to properly communicate, to appropriately classify statements of L1 as acceptable/unacceptable, and 3) not altering our empirical commitments: the alternative languages could be employed in stating empirically equivalent theories. In introducing an ontologically alternative language, however, there is no requirement that all the statements of the two languages be intertranslatable, synonymous, or equivalent in expressive power. The idea is simply that a revisionary metaphysician can rephrase our old statements in a way that retains the idea that what ordinary people say in the course of normal conversation is true (though perhaps not ‘perspicuous’2), or (as others would have it) ‘nearly as good as true’,3 thereby enabling eliminativists to distinguish what they are saying from what only a madman (or someone making empirical mistakes) would assert. Developing ontologically alternative languages has been crucial to the projects of revisionary metaphysicians like Peter van Inwagen, who uses his way of talking of particles arranged chairwise in order to show how his position is distinct from the ‘madman’s’ who simply believes there is ‘nothing in the chair-receptacle’, or that people who spoke of chairs were under some sort of illusion (1990, 105-7). Trenton Merricks (2001) similarly aims to distinguish his claim that there are no statues with an ordinary claim that there is no Bigfoot: the latter involves suspecting the believer of having made an empirical mistake—as a result of hallucinating, being the victim of a prank, or something like that. The former does not. The issue between the ontological chair-denier and chairbeliever, unlike the issue between believers and disbelievers in phlogiston or Bigfoot, cannot be determined by straightforward empirical means. As Merricks 2 3 As Hawthorne and Cortens, put it (1995, p. 156). Merricks (2001, 171–2). 7 puts it, “one’s visual evidence would be the same whether or not those atoms [arranged statuewise] composed something”, making these debates not ‘straightforwardly empirical’ (2001, 9). And again this is crucial to the revisionist’s project, enabling them to respond to those who say, “of course there are statues or chairs: I see them!” In short, the seriousness with which debates about the existence of ordinary objects have been pursued has relied on the possibility of developing ontologically alternative languages—without these, it is hard to make the eliminativist view plausible, and to distinguish it from the ‘madman’s’ view. Yet the very availability of ontologically alternative languages also enables us to make it clear why the pragmatist view, which would deflate these very debates, is plausible.4 With that much clarified, let’s go back to our questions (2) and (3): 2. Why do we go in for descriptions in terms of tables and trees, as opposed to descriptions in terms of dragons and phlogiston? 3. Why do we go in for descriptions in terms of tables and trees, as opposed to in terms of particles arranged table-wise, in terms of its tabling here, or in terms of sequenced temporal parts rather than enduring objects—that is, instead of employing some ontologically alternative (but perhaps empirically equivalent) language? It is the former question, I suspect, that we have in mind when we are tempted to say simply: It is because there are tables and trees, and there aren’t dragons and phlogiston and dragons. But the pragmatist has a ready line of response if we interpret the question this way: the first set of concepts (unlike the second) turns out to be successful in prediction, explanation, navigation; to not be based on imaginings, misperceptions, or other empirical mistakes. The need for this kind of success enables us to respect Blackburn’s insistence that sufficient attention be paid to the need for a sense of accuracy in our practices that enable us to successfully measure, mark, or navigate (2013, 73). Indeed I would say that, arguably, there are empirical preconditions for the successful use of the relevant vocabulary that are met by concepts of the first but not the second group. Some may worry about a problem arising here: for any way of stating what these empirical preconditions are would seem again to commit us to a particular ontology. But one may hold that there are empirical preconditions without holding that these are stateable (we begin learning language without stating application conditions, simply responding to ostensions in various situations), and (more importantly) without holding that there is a uniquely ontologically privileged statement of what these conditions are.5 What I am rejecting is not the idea that the 4 As I argue elsewhere (2014), the availability of ontologically alternative languages also shows why Stephen Yablo’s quizzical view should be writ large—to show that there is nothing to settle debates about the existence of pants any more than there is about the existence of numbers or propositions. For it gives a way of isolating the ‘assertive content’ of sentences like ‘the pants are at the cleaners’ from the ontological presupposition that there is a unified, countable material object. 5 Suppose, for example, one attempts to introduce a term for a natural kind—a kind of animal, say. There may be empirical preconditions on this, for example, that I am not subject to some sort of visual illusion when I think I am perceiving something to baptize. But one can accept that there are such empirical preconditions for successful employment of a term without thinking that there are also ontological preconditions: that there really be animals rather than mereological sums or time slices or… 8 world may play a role in determining whether our use of certain terms is successful, but rather the idea that there is any uniquely privileged language in which we are to describe the role the world plays—a language that would give us a read on ‘the true ontology’.6 But I think it is really the third sort of descriptive demand (not the second) that is at issue in preserving a global pragmatism: why do we go in for middle-sized object concepts (like table or tree) rather than for any of various ontological alternatives? The prior pragmatist response won’t work here—for we could also succeed at mapping a channel by employing concepts of particles arranged rockwise, or of rocks. And it is this third demand that is what is at issue in determining whether or not we require some specifically ontological—not scientific or empirical—justification for employing certain linguistic forms. It is also with respect to this question that the no-exit problem gains traction: the thought is that we can explain why we make use of and introduce concepts like those of right and wrong, of truth, or of number in terms that make no reference to such things— showing how these terms might be introduced on the basis of what Carnap would have called our ‘thing language’, to serve other purposes. But it is hard to see how to get a similar story going for explaining why we use a ‘thing language’—or the corresponding concepts—at all. It may well be true that we cannot, in natural language, give a rule for introducing ordinary object talk (a thing language) in terms that don’t appeal to ordinary objects themselves, whereas by contrast we can do so for numbers, properties, and propositions.7 Of course one might locally introduce terms for particular kinds of things like blazers by rules that there is blazer, say, when a person creates an article of clothing with this precise structure, to be worn in the following contexts, by the following sorts of people… (and give a pragmatic story about why we would want such a term in our language). But that itself requires prior reference to persons and objects—the latter in the sense of middle-sized dry goods. So that doesn’t undermine the idea that in general we need ontological justification for thinking there are middle-sized dry goods to justify introducing terms for middle-sized dry goods. So the core challenge for the global expressivist really should be presented as explaining why it is that we go in for a thing-language at all, rather than for any of various ontologically alternative languages, and doing so in a way that doesn’t ‘make any use of the referring expressions of the discourse’—that doesn’t say: because the world really does contain tables and trees rather than particles arranged tablewise or tablings or time slices. Nonetheless, I don’t think that these difficulties suggest that we require ‘ontological justification’ for introducing the relevant terms for medium-sized dry goods (so that we must reject pragmatism for those cases, and return to 6 So whether we describe the situation on Johnsonville Farm as one in which there are turkeys, or particles arranged turkeywise, or time-slices of 4-dimensional turkeys: the empirical preconditions for use of the term ‘turkey’ are fulfilled. If we describe it as one on which there is merely dirt (or particles arranged dirtwise, or…) they are not. 7 Though one might be able to give such a rule in an ontologically alternative, non-natural language, that appealed (say) to particles arranged Xwise. 9 Representationalism). Instead, what they suggest is that at least some such terms are ‘semantically basic’ for us (that is, terms that cannot be learned or introduced just by way of learning definitions stated in other terms) and perhaps further that some such concepts are basic for creatures like us. Indeed I think an interesting response to the ‘no exit’ problem is available by drawing on recent work in developmental psychology. Good psychological evidence has been amassed for the idea that there are certain basic concepts that are the products of natural selection, and are tied to what Susan Carey (2009) calls ‘core cognition’ (Carey 2009, 71-72). Carey identifies these basic concepts as including concepts for middle-sized objects, agents, causation, and quantity (Carey 2009, 449). All of these concepts, Carey argues, are generated by innate input analyzers, which act on perceptual input in accord with rules built-in as a product of a long evolutionary process.8 The object concept, for example, involves rules in which tracking spatio-temporal continuity plays a key role in object individuation and identification, as do features like perceived rigidity and cohesiveness. For, as Carey writes: “All the work to date suggests that the core cognition of objects exhibited by young infants has a long evolutionary history” shared with cottontop tamarins (with whom our common ancestor reaches back more than 100 million years) and our closer relatives, the Rhesus macaques (2009, 96). The idea that such concepts are conceptually basic for us gives us a way to embrace what Blackburn calls ‘the priority of the everyday’ (2013, 78) and to explain why it is that our other genealogical stories, about the origins of moral or modal or mathematical vocabulary, tend to bottom out here, and why, when asked why we go in for talk of this sort, we commonly can’t explain it in other terms—can’t do more than ‘stutter’. It is also an interesting option because it shows a way of distancing Carnapian pragmatism from conventionalism: one needn’t hold that the conceptual scheme that underlies the most basic terms of our language (the ‘thing language’) is merely arbitrary or conventional to deny that use of those or other terms requires ontological justification. The response we give for these basic concepts is still pragmatist in spirit—we appeal to the use that possession of the relevant concepts (and ultimately terms to go with them) has for our evolutionary success in explaining why we have these concepts. But it is not conventionalist. And it is still subject-naturalist in Price’s (2011) sense; we appeal to the results of empirical evolutionary psychology to figure out how and why we acquired certain concepts, and what use it served for us to have concepts governed by these rules. But does it involve giving up on the pragmatist project? In a sense it suggests that we answer the question ‘Why do we have such concepts (for middle-sized dry goods, enduring medium sized material objects)?’ not exactly by way of anthropology but by evolutionary psychology. Some might suggest that that still preempts the possibility of a pragmatist approach to everyday talk—for the right 8 Thus we shouldn’t think of basic experience as merely involving perceptual primitives, and on that basis learning and constructing representations of objects, number, agency, or causality: indeed Carey says that there is no known proposal for how this learning could work (2009, 456). 10 evolutionary story must be that we have these concepts of medium-sized dry goods because that’s what there is in the environment, and that accuracy explains why the concepts work for us and why possession of those concepts helps us and our evolutionary relatives to survive and reproduce. But this is a totally unnecessary interpretation of the data. Natural selection doesn’t care a bit about correspondence to a Uniquely True Ontology of the World. It cares about the success of the organism: success at surviving, success at breeding. All of this can be understood in perfectly pragmatic terms that simply appeal to the fact that having concepts governed by these rules (rather than ontological alternatives) is very useful to us. In fact, there are good reasons for thinking that the only plausible route for explaining (in causal-evolutionary terms) why we have the concepts we do is in terms of ways in which it might be cognitively preferable for creatures like us, and might make a causal difference to us to possess one set of concepts rather than another—not in terms of ontological accuracy. For suppose we are ontologically inaccurate, and think using a thing-language, while really there are only particles arranged thing-wise. It makes a causal difference—one relevant to survival--if you make a mistake and think there is a branch rather than a snake. But, even according to those who vehemently debate the issue, it makes no causal difference (and so none that can be relevant to survival or reproduction) if there ‘really’ are only particles arranged snake-wise and not a snake. As they are causally indistinguishable, there seems no way such putative differences in ‘ontological accuracy’ could play a role in evolutionary explanations. Of course, as the snake/branch case makes clear, evolution does care about accuracy in a certain respect: for evolutionary success, it is (on the whole) useful to avoid making empirical mistakes, say, about the locations and movements of danger, or of nutrition, or… But empirical mistakes could be avoided, and similarities and differences tracked, using a variety of ontologically alternative conceptual schemes or languages. As revisionary ontologists since Quine have shown over and over again, there are a variety of different languages one may use in conveying important empirical information, tracking relevant changes in the environment, and so on. One could have accurate tracking skills and avoid dangers regardless of whether one thought or spoke of snakes or particles arranged snakewise, or finding where it’s snaking, etc.9 So what explains why we evolved to use the thing-language rather than one of the ontological alternatives? It is worth remembering, first of all, that evolutionary explanations needn’t tell us this option was optimal or unique. Nonetheless, there are various places one might look for a non-ontological explanation of why we came to use (and persist in using) the thing-language. Perhaps the case could be made that it is more cognitively efficient for creatures like us than simply tracking changing features, or individuating the world in terms of sequenced temporal parts, or tracking particles and ways they are arranged. Plausibly, it fits better with the constraints of our evolved perceptual system (since we can’t perceive particles). We needn’t take a stand on the best such explanation 9 This is not to say that all would be equally successful—there may be differences in cognitive efficiency, expressive power, etc. that plausibly make an evolutionary difference. I return to this point below. 11 here—that is an empirical matter. The point is that there is plenty of room for explanations of why we use a thing-language or conceptual scheme that don’t appeal to its ‘ontological accuracy’ as compared with ontological alternatives. It is also worth noting that this style of pragmatic explanation is fully available without commitment to whether or not these basic concepts are innate, as long as we don’t maintain that they are learned by way of detecting these specifically ontological differences in the world. In short, there are a variety of ways to address this third explanatory demand in pragmatist spirit, without a flatfooted stutter that just appeals to the existence of the very ontology that conceptual scheme apparently refers to. If we accept that certain concepts are basic for creatures like us, and that certain terms that align with them are semantically basic, we can suitably acknowledge the priority of the everyday. And we can do so without thinking that the only explanation that can be given of why we have terms for middle-sized dry goods (rather than employing some ontologically alternative conceptual scheme) is that there are middle-sized dry-goods, taken in the sense of asserting that that (and not the monist, trope-theorist, nihilist’s or organicist’s ontology) is the ‘true ontology’, giving us ‘ontological justification’ for introducing this range of concepts and terms rather than any ontological alternatives. Here as elsewhere, no such ontological justification is needed, and we can justifiably approach the relevant language in a pragmatist spirit.10 2. Respecting differences I have argued so far that ordinary object talk doesn’t require ‘ontological vindication’ or Representationalist presuppositions after all, so that the pragmatist approach can be as applicable there as elsewhere. But are there huge asymmetries between talk of ordinary objects and talk of the moral, modal, or other problem areas—asymmetries that suggest the need for a bifurcated view, rather than a form of global pragmatism? Blackburn argues that, by their own lights, pragmatists should see their view as ‘vindicating’ realism about chairs and tables and other ordinary objects. For the language that “embraces external, independent, public objects earns its living. It works, and nothing else of which we have the faintest conception does so. So we are to embrace it” (2013, 82).11 This might, though, lead us naturally to a merely local form of pragmatism: accepting realism about ordinary objects, say, while giving an expressivist account of discourse about the modal, moral or mathematical. Couldn’t a pragmatist, however, say the same for the suspect discourse about the mathematical, moral or modal, however, and once again adopt a global view? 10 And of course, once we employ these concepts, we very often easily reach the conclusions that there are tables, trees and the like. And then it becomes not at all silly to say ‘we use the concepts of tree and table and not of witch and dragon because there are trees (and not dragons)’, but no ontological read is needed on this: it simply alludes to the success of the first, and the failure of the second. 11 Of course eliminativists about ordinary objects might beg to differ here—suggesting again that we at least have the ‘faintest conception’ of various ontologically alternative languages that could ‘work’. 12 That is, we might in each case say that, given that the form of discourse earns its keep, we should be realists about the relevant entities under discussion. Blackburn cautions against generalizing this result to matters moral, modal, and the like, noting: …there is a huge asymmetry between the case of common sense and what I called the coastal waters of science, on the one hand, and cases like possible worlds, numbers, and rights and duties or the passage of time on the other. For in embracing the common-sense scheme we embrace not only the tables and chairs it posits but a distinct view about our relation to them. (2013, 82) What are these huge asymmetries, and can the pragmatist do them justice, while retaining a form of realism about the moral, modal, and mathematical? Blackburn identifies several points of contrast that seem onto something: • Tracking and sensitivity: We think of ourselves as causally influenced by tables and chairs, and sensitive to their ‘appearances and changes’: “if my chair collapses I will notice it, …if the table dances around or bursts into flames I will register that, … were it to grow in size it would have all kinds of other consequences that I could also register, and so on…” (2013, 82) • Automaticity: Our (perceptual) registering of common-sense entities is “outside our control and outside the influence of other cognitive functions” (2013, 83) whereas this is not the case, e.g. for ethics. • Explanation: Common sense entities are ‘directly witnessed’ and ‘Their whole life… consists in their role as systematizers and explainers of experience’ (2013, 83) As a result ‘There is therefore no option of embracing the scheme while holding back on its own explanations of why we do so’ in these cases, though doing so is quite plausible for possible worlds and the like (2013, 83). These asymmetries, Blackburn seems to suggest, give us reason to hesitate about simply adopting a form of global pragmatism, for they might give us reason to reserve (a form of) realism for the ordinary objects of the common sense world that we hold back for the objects apparently referred to by the areas of discourse we are (locally) pragmatist about. I agree that there are important differences here that ought to be respected. But I think they can be respected from within a kind of global pragmatist view that denies that we ever need ontological justification for introducing or retaining a certain form of discourse (though we may make inferences from truths of the discourse to simple realist claims about the objects). The first asymmetry, marking the role terms for ordinary objects (but not for numbers or moral properties) play in tracking and world-sensitivity, is something Price has already tried to respect and preserve in drawing a distinction between E(xternal)-representations, which involve environmental tracking, covariation with environmental conditions, etc., and I(nternal)-representations, which give priority to internal connections, and are governed primarily by functional/inferential role (2011, 20-23 and 2015). This kind of insight can be preserved on the current suggestion as well. Indeed we can see our ordinary object concepts as having primarily this sort of world-tracking function. By contrast, we introduce terms for 13 moral facts, modal properties, or numbers with other functions—whether of expressing attitudes, conveying semantic rules, or enabling us to state in finite form what would otherwise require infinite expression. But we may acknowledge these differences in function without thinking that we require ontological justification for introducing the thing language rather than some ontological alternative (one that would also enable us to engage in accurate tracking etc.). In their functioning as Erepresentations, ordinary object terms need not differ from ontologically alternative languages in which we could track and retain sensitivity to similarly relevant elements of the environment. (We could with equal sensitivity report that the table was moved to the left or that the particles arranged tablewise were moved to the left [without affecting their relative arrangement]). So we can retain our pragmatism while respecting the difference between terms that are introduced to serve a tracking and world-sensitivity function from those that serve other functions. The automaticity point comes naturally with the idea floated above, that our basic object concept is fundamental for us, perhaps even built-in as a product of a long evolutionary process—and so not conventional, or under our cognitive control, nor an optional, if useful ‘add on’ to our conceptual scheme, but its very starting place. Again, we can acknowledge and accept that difference between our ordinary object concepts and terms and those for the moral, mathematical, or modal, without thinking that the difference lies in that the former (and not the latter) requires an ontological explanation. The ‘explanation’ point can be preserved as well. Blackburn says that for common sense entities, their “whole life… consists in their role as systematizers and explainers of experience” (2013, 83). But we can acknowledge that by noting that the ‘entry rules’ for our ordinary object terms and concepts rely on certain empirical presuppositions in such a way that, if those fail, we are not entitled to say that there are the relevant objects, the use of the terms in explanations being, in Yablo’s terms, ‘wrecked’ (2009, 515). Indeed terms for ordinary objects might have certain empirical preconditions for their successful use—marking the difference between conditions under which we succeed or fail in our attempts to introduce a term for a new person, species, or artifact (we might fail in introducing such a term if we are suffering from some sort of perceptual illusion, for example). Other, more derivative terms might serve a different purpose; terms for numbers, propositions, properties, etc. may serve a useful role in explanation (enabling us to state our explanations and laws more succinctly, for example).12 But the use of these terms seems to carry with it no empirical presuppositions that might fail. Introduction rules for these derivative terms may be insulated from such potential failings (we can move from ‘there are five cups’ to ‘there is a number, five’, regardless of whether we are suffering from perceptual illusion). In the words of Stephen Schiffer there is no ‘algorithm for their elimination’ (1996, 152). Moreover, even if one thought that numbers or propositions could turn out to fail to exist (a thought I myself deny we can really make sense of), as Stephen Yablo has pointed out (2005, 94-5), their failure to exist would make no difference to the success of the 12 See, e.g. Yablo 2005 on number talk; Schiffer 2003, Chapter 8, on proposition talk. 14 explanations they contribute to. (In this regard, they are quite unlike platypuses, planets or electrons.) So we can still capture a difference in the roles terms of these kinds play in explanations, without being Representationalists about terms of the first sort while holding this back from terms of the second sort. Blackburn says that, given that their whole point is to systematize experience, there is no prospect of embracing the common sense conceptual scheme without explaining our possession of it by saying roughly that we have that conceptual scheme because those are the things there are. But again I suspect that the feeling here comes from considering the second rather than the third explanatory demand. We can’t say why we have this conceptual scheme rather than an ontologically alternative but empirically equivalent one by saying that only the former can play a role in systematizing and explaining experience. For (given empirical equivalence) we could systematize and explain experience in terms that appeal to particles arranged tablewise rather than in terms that appeal to tables. The pragmatist may acknowledge the role of common sense terms in systematizing and explaining experience, while denying that ontological justification is needed or available for adopting one set of worldly concepts or terms over an ontologically distinct set (there, we might again look to pragmatic criteria that make our objectual scheme more efficient or user-friendly, or more suited to creatures with our perceptual apparatus (which can’t distinguish particles)). In sum, I think that it’s undeniably true that there are ‘huge asymmetries’ between terms that serve a tracking function (including but not limited to our terms for ordinary objects) and those that serve other functions instead. We need to respect those differences in our theorizing. But once one sees key question as (3) why we go in for a thing-language rather than adopting one of a number of merely ontologically alternative conceptual schemes, the feeling that there is a difference in the sense that some form of ontological justification is needed for introducing thing talk but not number talk—begins to fade away. Our basic pragmatist approach, that insists that our use of a language form doesn’t require ontological justification, remains intact (though proper use of some language forms (and not others) may require that certain empirical preconditions be met). If what I’ve said is correct, then even in the case of terms for ordinary objects we can begin by aiming to explain why we would want to have a certain form of discourse, answering that without prior commitment to the relevant ontology. We can then go on (with Michael Williams (2011) to ask what rules the terms of that discourse follow that enable them to fulfill that function. Finally, using those rules, we may ask whether the terms of that discourse refer—whether there are the entities in question. But as I have argued elsewhere (2015), once the question is asked in that way, it is typically easy to answer. In some cases, the way the terms (say for numbers or propositions) are introduced guarantees that they refer. In other cases (say, terms for trees or platypuses), one must look to the world—but just engage in normal observation or empirical investigation to check if empirical preconditions are met—to see if they refer. So just as these differences in the origin and functions of our terms don’t make a difference with respect to whether one can retain a pragmatist approach, I also think they don’t lead to differences in whether one should say there really are 15 the entities in question. Instead, as I have argued elsewhere (2015, 145-57), we should adopt a form of simple realism about ordinary objects, numbers, duties, and modal facts alike; that there are such things in the only sense these terms have. We don’t need a bifurcated realism about the entities we accept. And in each case, we can retain the core pragmatist idea that introduction of the relevant conceptual or linguistic form does not require ‘ontological justification’ (though in some cases and not others it has empirical presuppositions). Instead, we may introduce, for various purposes, different ontologically alternative languages and build on these for other purposes. And in each case we can hope to get a pragmatic—anthropological, genealogical, or perhaps even evolutionary—story about the use and function of linguistic forms like these, where that needn’t be a matter of saying (in flatfooted Representationalist mode): we employ these linguistic forms because they correspond to the things there really are. 4. The Consequences of Global Pragmatism So far I have aimed to show how a contemporary pragmatist can extend her view globally, to include talk about ordinary objects as much as about discourse about the moral or modal, while still respecting the priority of the everyday, and the huge asymmetries between these different forms of discourse. Seeing global pragmatism in this way also makes the view show up importantly differently from how it is traditionally conceived. This is not your grandfather’s pragmatism. So let’s close by examining its consequences both for first-order ontology and for metaontology. For first-order ontology: If we accept the pragmatist approach (globally), then, following the linguistic priority thesis, we do not begin by asking, “are there numbers, properties, moral facts, or tables?” Instead, we begin by asking what the function of the discourse is. We can then give an account of meaning by identifying the rules that govern these terms, enabling them to fulfill their function (Williams 2011). These rules then enable us to see whether the conditions for properly using the term are fulfilled—and so to draw conclusions about whether there are numbers, properties, moral facts, or tables. Thus the global pragmatist position does not render existence questions unanswerable. Instead, it makes answering them a straightforward matter of seeing whether the relevant rules lead to the entailment, or the relevant conditions are fulfilled.13 As long the rules are coherent and we haven’t made empirical errors, the results we get to our existence questions are generally going to be positive. So the global pragmatist view, in short, fits naturally with the easy approach to ontological questions that I have defended at length elsewhere (2015). This is an important result, since pragmatist views are often shunned on suspicion of being unacceptably anti-realist. But the first-order ontological views that come out of this global pragmatist approach are typically nothing other than straightforward realism about the disputed entities. The pragmatist, so conceived, does not say that we should accept that there are tables or trees because it is useful 13 It may be somewhat less straightforward in cases where we have worries that there may be incoherencies in the rules—but I’ll leave those worries to the side here. For discussion see my (2015, Chapter 8). 16 for us to do so. Rather, she says that we adopt a conceptual scheme that includes objectual terms and the like (rather than an ontologically alternative scheme) because employing such a scheme is useful for creatures like us (with our perceptual apparatus, cognitive limitations, evolutionary niche…). But having adopted that scheme, it is then (given a small amount of ordinary experience) an easy matter to conclude that there (really) are such things as tables and trees—to be straightforward realists about them. (It of course does not rule out also concluding that there are particles arranged tablewise, table time slices, or the like—though we may lack motivation for utilizing these alternative conceptual schemes. So it tells us that there are tables and trees, but not that this is an exclusive True Ontology of the World). And the objects we end up accepting are generally (except in the obvious cases of social and cultural objects) mind and language independent. So the pragmatist does not embrace anti-realism or deny that there are mind-independent objects and kinds. Instead, the global pragmatist approach gives us a straightforward realism about all the entities you ever wanted (and more). So what distinguishes the pragmatist view from that of the serious ontologist is not any standard first-order realist claim, but rather rejecting the ontology-first approach that says we must first settle ontological questions about what exists before we can determine whether or not we are justified in introducing (or retaining) pieces of terminology (though sometimes we must settle empirical questions to be so justified!). For that reason also, it denies that there is a unique answer to the question of which of various ontologically alternative languages is the uniquely correct one. They may differ in their expressive power or other pragmatic (or even moral or political) virtues, but on this view there is no prospect of beginning with ontology to figure out which of these languages correctly matches it—indeed they may (if no internal problems with the rules arise, nor failures of external empirical propositions) all be fine in the sense that we have reason to say that there are the things named by each. Once we see global pragmatism in this light—as part of a rejection of the idea that some forms of language (and not others) properly mirror the ontological structure of reality, but not as rejecting the idea that there really are cats and not dragons, oxygen but not phlogiston—I think it becomes far more palatable and interesting than both its opponents, and even some of its proponents, had realized. And once we see this, perhaps more will be ready to go global—extending the neopragmatist approach even to discourse about ordinary objects. As I mentioned at the outset, a standard motivation for adopting a local form of pragmatism is that it enables us to sidestep the traditional ontological ‘placement’ problems and epistemic problems plaguing areas of discourse like the moral and the modal. It might at first seem that—even if a route is available for doing so—taking a pragmatist approach must be totally unmotivated for discourse about ordinary objects. For here, we certainly don’t have the ontological problem of ‘placing’ such objects among the natural furniture of the world—these ordinary objects are the natural furniture of the world within which we had trouble ‘placing’ moral or modal properties! We also lack the epistemic problems that notoriously arise in saying how we can acquire moral, modal or mathematical knowledge—again our ability to 17 know of ordinary objects through experience is (while open to skeptical doubts) the paradigm against which our knowledge of the moral, mathematical or modal is contrasted. But once we have clarified the explanatory demand at issue, we can see that there are parallel problems for ontology. The differences between an ontology of ordinary objects versus an ontology of particles arranged ordinary object-wise, or a nihilist ontology expressed in a feature-placing language, are (as all sides agree) not empirical differences. They are merely ontological differences. But now the same kinds of question arise for these supposed ontological features: how can we ‘place’ in the natural world those features that would make a merely ontological difference –making a difference, say, between situations in which particles arranged in certain ways did versus did not compose something? Epistemically, how could we come to know what the uniquely true ontological view is? These are the problems that motivate skeptical and deflationary metaontological approaches, and they are as formidable as those that motivate meta-ethical expressivists and modal nonrepresentationalists. And once again, taking the pragmatist stand helps relieve us of these problems: for we do not need to ‘posit’ features that make such ontological differences and justify us in using one language rather than an ontologically alternative one. And we do not need to provide a story about how we could acquire knowledge of the uniquely true ontological view, in a way that could again justify us in adopting a particular language over an ontologically alternative language. Where justifications are available, on the global pragmatist view, they will be given not in metaphysical terms but rather in terms of pragmatic or even moral issues about what conceptual scheme we ought to adopt in order to better meet some pragmatic, social, or moral goal. Taking a pragmatist approach here, as elsewhere, is motivated by its ability to diminish ontological and epistemological mysteries.14 Works Cited: Blackburn, Simon (1993). Essays in Quasi-Realism. Oxford: Oxford University Press. Blackburn, Simon (2013). “Pragmatism: All or some?”, in Price (2013): 67-84. Brandom, Robert (2008). Between Saying and Doing: Towards an Analytic Pragmatism, Oxford: Oxford University Press. Carey, Susan. (2009). The Origin of Concepts. Oxford: Oxford University Press. 14 I would like to thank audiences at the Global Pragmatism conference in Szczecin, Poland, at Cornell University, and at Notre Dame University for helpful comments on and discussion of earlier versions of this paper. 18 Carnap, Rudolph. (1950). “Empiricism, Semantics, and Ontology”. Reprinted in Meaning and Necessity, Second edition. Chicago: University of Chicago Press. (second edition 1956) Chrisman, Matthew (2007). “From Epistemic Contextualism to Epistemic Expressivism”. Philosophical Studies 135, 2: 225-254. Garfinkel, A. (1981). Forms of Explanation: Rethinking the Questions in Social Theory. New Haven: Yale University Press. GIbbard, Allan (1990). Wise Choices, Apt Feelings. Cambridge, Massachusetts: Harvard University Press. Hawthorne, John and Andrew Cortens, “Towards Ontological Nihilism,” Philosophical Studies, LXXIX (1995): 143–165, at p. 156. Horwich, Paul (1998). Truth. Second edition. Oxford: Oxford University Press. Merricks, Trenton (2001). Objects and Persons. Oxford: Clarendon. Price, Huw (2011). Naturalism without Mirrors. Oxford: Oxford University Press. Price, Huw, with Simon Blackburn, Robert Brandom, Paul Horwich and Michael Williams (2013). Expressivism, Pragmatism and Representationalism. Cambridge: Cambridge University Press. Price, Huw (2015). “From quasirealism to global expressivism — and back again?” In R. Johnson and M. Smith (eds.), Passions and Projections: Themes from the Philosophy of Simon Blackburn. Oxford: Oxford University Press: 134—152. Restall, Greg (2012). “A Cut-Free Sequent System for Two-Dimensional Modal Logic, and Why it Matters”. Annals of Pure and Applied Logic. (163: 11): 1611-1623. Ryle, Gilbert. (1949). The Concept of Mind. London: Hutchinson. Schaffer, Jonathan (2005). “Contrastive Causation”, Philosophical Review 114 (3): 327-358. Schiffer, Stephen (1996). “Language-Created, Language-Independent Entities”. Philosophical Topics 24 (1): 149-67. Schiffer, Stephen (2003). The Things We Mean. Oxford: Oxford University Press. Sellars, Wilfrid (1958). “Counterfactuals, Dispositions and the Causal Modalities”. In Herbert Feigl, Michael Scriven, and Grover Maxwell, eds. Minnesota Studies in 19 Philosophy of Science Volume 2: Concepts, Theories and the Mind-Body Problem. Minneapolis: University of Minnesota Press: 225-308. Sider, Theodore (2011). Writing the Book of the World. Oxford: Oxford University Press. Thomasson, Amie L. (2007). “Modal Normativism and the Methods of Metaphysics”, Philosophical Topics, 35/1&2: 135-160. Thomasson, Amie L. (2014). “Quizzical Ontology and Easy Ontology”. The Journal of Philosophy CXI, 9/10: 502-528. Thomasson, Amie L. (2015). Ontology Made Easy. New York: Oxford University Press. Thomasson, Amie L. (forthcoming). Norms and Necessity. Van Fraassen, Bas (1980). The Scientific Image. Oxford: Clarendon Press. Van Inwagen, Peter (1990). Material Beings. Ithaca, New York: Cornell University Press. Williams, Michael (2011). “Pragmatism, Minimalism, Expressivism”. International Journal of Philosophical Studies 18 (3):317-330. Williams, Michael (2013). “How Pragmatists can be Local Expressivists”, in Price (2013). Yablo, Stephen (2005). “The Myth of the Seven”, in Mark E. Kalderon, ed. Fictionalism in Metaphysics. Oxford: Oxford University Press, 88-115. Yablo, Stephen (2009). “Must Existence-Questions Have Answers?”, in David Chalmers, David Manley and Ryan Wasserman, eds. Metametaphysics. Oxford: Oxford University Press. Yablo, Stephen (2014). “Carnap’s Paradox and Easy Ontology”. Journal of Philosophy CXI, 9/10: 470-501. 20