Précis of Ontology Made Easy

Philosophy and Phenomenological Research Philosophy and Phenomenological Research Vol. XCIX No. 1, July 2019 doi: 10.1111/phpr.12615 © 2019 Philosophy and Phenomenological Research, LLC Precis of Ontology Made Easy AMIE L. THOMASSON Dartmouth College Many people sense that something is amiss in mainstream metaphysics. Mainstream metaphysicians think of themselves as offering ‘theories’ of a piece with those of the sciences, prominently including ‘discoveries’ about what really exists. In the decades after Quine, disputes about existence questions have massively proliferated—including debates about whether there ‘really’ are tables and chairs, numbers, properties, propositions, fictional characters, mereological sums, and so on. Yet in our metaphysical ‘theories’, in contrast with the scientific, there is nothing close to convergence—quite the opposite. Nor is there even any agreement about what it would take to resolve the disputes. It is no wonder that bystanders have come to look on such debates with perplexity, and to raise doubts about their sense and value. In Ontology Made Easy I aim to articulate where these disputes have gone wrong, giving aid and comfort to the perplexed bystanders. I also aim to develop and defend an alternative deflationary way of understanding existence questions—a way that gets us all we wanted out of existence questions in ordinary life, without the mysteries of mainstream metaphysics. That is what I call ‘the easy approach to ontology’. One can think of it as developing something like a neo-Carnapian approach to existence questions considered as internal questions, though my aim is to develop a clear, coherent and attractive position, not to engage in historical exposition. What is ‘easy’ about the approach is the epistemology. Mainstream metaphysics leaves the epistemology of metaphysics deeply obscure. For mainstream metaphysicians tend to agree that (unlike typical scientific debates) debates about what exists cannot be resolved empirically, nor (they insist) can they be resolved by conceptual analysis. In practice, they often appeal to ‘general metaphysical principles’ in support of a view, but it is unclear how we are supposed to know such metaphysical principles. Many more appeal to certain ‘theoretic virtues’ their view is supposed to exhibit. But (as Bennett (2009) and Kriegel (2013) have argued) rival views typically just trade off one theoretic virtue for another, and it is unclear in any case that such virtues should be thought of as ‘truth-conducive’ in metaphysics. The easy approach to ontology, by contrast, holds that those existence questions that are well-formed and answerable can be answered by straightforward conceptual and/or empirical methods. (I readily grant, though, that both conceptual and empirical work may be difficult, and may not be something we can complete in an idle afternoon.) In general, BOOK SYMPOSIUM 223 as I argue, (for sortal terms K) a question about whether things of a given kind K exist can be addressed by determining whether the application conditions actually associated with ‘K’ are fulfilled (for more on what application conditions are, see Chapter 2). And this, in turn, can often be done simply by making use of our conceptual mastery and engaging in ordinary observation. The easy approach is also ‘easy’ in that it allows that, in many cases, disputed ontological questions (“Are there propositions?” “Are there events?” . . .) can be answered by trivial inferences from uncontroversial premises. So, for example, we can make arguments like the following: • • • • Undisputed claim: May was born on a Monday Conceptual truth: If P was born on D, then P’s birth occurred on D Derived claim: May’s birth occurred on a Monday Ontological claim: There is an event (namely, May’s birth) (2015, 135) Such arguments lead us from an uncontroversial truth, by easy steps, to a conclusion that answers the relevant ontological question—in this case, ‘yes, there are events’. For those who are not committed to idea that ontological debates are deep and serious, the virtues of the easy approach are clear. It makes good sense of the claims we ordinarily want to make about what there is and isn’t and gives them the truth values we normally think they have. Given this approach, we can say that there are tables and chairs, births and heart attacks, and prime numbers between 10 and 20. At the same time, we can deny existence claims exactly where that seems appropriate. We can deny that there are witches (assuming that what it would take for there to be witches is for there to be women who become endowed with supernatural powers by making a pact with the devil, and that those application conditions are not met). We can also deny the existence of phlogiston (assuming that the existence of phlogiston required that a single chemical kind which was a component of substances was released during combustion, and that condition was not met). But at the same time, while preserving the sense and truth-values of existence questions asked in the ordinary business of life, the easy approach demystifies the epistemology of metaphysics. The answers we get to existence questions, using the easy approach, are usually positive (except when, as in the witch or phlogiston case, mistakes are being made)—we can get easy arguments for the existence of propositions, properties, numbers, events, and much more. Moreover, as I explain in Chapter 3, this is a straightforward, simple realism about the entities in question: the easy approach tells us not that the entities exist in some special ‘lightweight’ sense, but rather that they exist in the only sense that has sense. One might wonder, then, in what sense the view is deflationary? As I insist in Chapter 3, it is the debates rather than the objects that are deflated: The easy approach gives us a deflationary metaontology. Serious debates about whether or not numbers, properties, events, or tables ‘really’ exist, in which disputants support their views by appealing to which ‘theory’ has the most ‘theoretic virtues’, or which ‘fits’ best with some preconceived ‘general metaphysical principles’, are misguided and lead us astray. Instead, such questions can be answered easily by straightforward conceptual and/or empirical methods—and often by trivial arguments from uncontroversial premises. 224 AMIE L. THOMASSON Let me make clear how this book aims to intervene in debates in metametaphysics. The ‘easy approach’, as mentioned above, is an heir to Carnap’s approach to ‘internal’ questions. But according to the standard lore, Carnap was long ago defeated by Quine, clearing the way for a neo-Quinean approach to ontology.1 To motivate a neo-Carnapian position, I thus need to begin with some historical ground-clearing. I undertake this in Chapter 1, arguing that Carnap was not (as is often thought) ‘defeated’ by Quine, and that it is the neo-Quinean approach that is the outlier, taken in the context of a broader history of philosophy. I also argue that Carnap’s position was unjustly dismissed by false associations with verificationism and anti-realism—and that both of these can be avoided on a deflationary neo-Carnapian position. Even after a neo-Quinean approach came to dominate, deflationary views sometimes surfaced. Eli Hirsch, for example, develops and defends the influential idea that many ontological disputes are merely verbal disputes in which the disputants can often be charitably interpreted as using ‘exists’ or the existential quantifier with different meanings (though at most one is speaking English). For years, the main threat to taking ontological debates seriously was thought to be the threat that the disputes were merely verbal. The best way of cashing that out was thought to lie in the idea that the quantifier can vary in meaning in the mouths of disputants. As a result, for decades, the bulk of discussion in metaontology centered on debates about whether or not we should accept that the quantifier does, or could, vary in meaning. This is, roughly, where I came in. An important feature of the form of easy ontology that I develop in the book is that it gives us a way of deflating ontological debates that not only avoids anti-realism and verificationism, but also does not rely on quantifier variance, and so can side-step debates about whether the quantifier does (or could) vary in meaning. Instead, I argue, we need only accept a univocal, formal sense of ‘exists’ (and the quantifier) to develop the easy approach. In Chapter 2 I take some time to draw out this view of ‘exists’ as a formal term, discuss the function of existence claims, and give a central rule of use governing ‘exist’: For a sortal term ‘K’, Ks exist if and only if the application conditions actually associated with K are fulfilled. I show how this rule of use entitles us to answer existence questions ‘easily’ (in the sense above, often using our conceptual mastery and ordinary processes of observation). I also argue that, if we accept this understanding of ‘exists’, we should reject all substantive across-the-board ‘criteria of existence’ often used in ontological arguments, such as ‘to be real is to have causal powers’, or ‘to be real is to be mind-independent’. ‘Easy’ arguments for some disputed entities have, of course, been made before, and I draw on this crucial and groundbreaking work. For example, neo-Fregeans in the philosophy of mathematics (including Bob Hale and Crispin Wright (2001, 2009)) offer ‘easy’ arguments for the existence of numbers (though these differ a bit in their form and requirements from my easy arguments—see my Chapter 3). Stephen Schiffer (2003) gives easy arguments for the existence of propositions, properties, events, states and fictional characters. I aim to show how this past work can be generalized into an overall approach to addressing existence questions, and to defend it against common objections. As I argue in Chapter 3, easy inferences can be made not only to resolve debates about abstract entities such as numbers, propositions and fictional characters, and entities 1 While ‘neo-Quinean’ has become a standard label for this sort of approach to metaphysics, as I mention in the book (drawing on Price 2009), this is not to attribute it to the historical Quine himself. BOOK SYMPOSIUM 225 like events or states, but also about ‘ordinary objects’ such as tables and chairs. For we may begin from a statement that is not a point of controversy between those who do and do not accept that tables exist, say “There are particles arranged tablewise”2 and go on to make use of (what I claim is) a conceptual truth, “If there are particles arranged tablewise, then there is a table”, to get the ontological claim “there is a table”, settling the question about the existence of tables.3 Chapters 4 and 5 are dedicated to distinguishing the easy approach from other ways of being suspicious about ontological debates, including Hirsch’s quantifier variance view, the epistemicist view (popularized by Karen Bennett (2009)) that we sometimes lack sufficient grounds for knowing the right answer to ontological questions, and the ‘ontological anti-realist’ view, which denies that every unproblematic ontological existence claim has an objective and determinate truth-value (Chalmers 2009). Fictionalists raise one of the most recurrent criticisms of easy ontological arguments: that the ontological conclusions of easy arguments should not be taken seriously at all, as they are nonliteral speech or implicitly in the scope of some sort of pretense. Fictionalists also provide a rival way of diagnosing what is wrong with many traditional debates between, say, Platonists and eliminativists. Chapter 5 is devoted to a more extensive evaluation of fictionalism versus easy ontology, in which I respond to the fictionalist objection and argue that fictionalists face a crucial problem of their own in maintaining an analogy between the disputed ontological discourse and overtly fictional discourse. I also argue that the motivations for fictionalism are as well or better served by the easy approach to ontology. Although easy arguments have been raised before, typically they are quickly dismissed. Most of the book’s second half is directed to dealing with various objections head-on and showing why they should not lead us to dismiss easy ontological arguments. Some claim that it ‘just can’t be right’ that these questions can be settled easily, given the seriousness with which smart people engage in ontological debates (section 4.4). Others claim that easy arguments quickly lead to troubles—whether of a ‘bloated’ ontology, of making it seem we can ‘magic’ things into existence (Chapter 6), or of keeping ‘bad company’ with clearly problematic arguments (Chapter 8). One of the central reasons easy arguments are not widely accepted is that they apparently rely on conceptual truths—but the idea that there are conceptual (or analytic) truths has been unpopular since Quine, and recently subjected to new criticisms by Timothy Williamson. Since I addressed the Quinean arguments against analyticity in an earlier book (2007), here (Chapter 7) I focus on Williamson’s arguments, aiming to show that they do not undermine the trivial arguments easy ontology accepts, or its claim that those who have mastered the relevant linguistic/conceptual rules are entitled to make the inference from the undisputed claim and accept the ontological conclusion. Another interesting objection (to which I respond in Chapter 9) comes from Thomas Hofweber’s influential and intriguing arguments that the conclusions of easy arguments use only an ‘internal’ (substitutional) 2 Roughly, in the sense that there are particles suitably arranged and bonded to perform the characteristic functions of tables, and so arranged in the right cultural context, by individuals with the right intentions, etc. 3 It is worth noting, however, that I do not think that this is our normal, or only way to come to know that there are tables—certainly one needn’t master particle arrangement talk in order to come to know that there are tables. Our ordinary conceptual mastery (which includes mastery of the relevant application conditions) and observational practices are enough. 226 AMIE L. THOMASSON reading of the quantifier, and so do not make genuine existence claims (which require an external, ‘domain conditions’ reading of the quantifier). There is not enough space to rehash those responses here, but I hope those inclined to dismiss easy arguments will take a look. One increasingly popular response to easy ontological arguments among those who engage in serious ontological debates, is to simply accept that ontological questions, stated in English, are easy to answer, but to suggest that ontologists can retreat to a special language of “Ontologese” in which such disputes remain deep and difficult. Theodore Sider gives the most extensive development of this approach in his (2011), where he argues that one may introduce a fundamental quantifier, the meaning of which is stipulated to carve at the logical joints, and in terms of which one can raise substantive existence questions (2011, 171). He also argues that deflationists like myself cannot resist this move without engaging in deep metaphysics of precisely the kind they claim to reject—for if they reject this kind of ‘joint carving’ fundamental quantifier, they must be denying that the world has quantificational structure, and thus making a deep metaphysical claim. Chapter 10 is devoted to discussing this interesting line of argument. In response to Sider, I argue that, while deflationists like myself mustn’t endorse the idea that there is such a meaningful joint-carving quantifier, that doesn’t require making a claim about what kinds of structure the world has or lacks. By engaging not in serious metaphysical speculation about the world’s structure, but rather linguistic analysis, we can problematize the very idea that there may be a quantifier that ‘carves the world at its joints’. Perhaps the most common strategy critics reach for against deflationism is a generalized version of Sider’s: that this metaontological view must itself somehow be engaging in deep metaphysics, and so be no better than the views it aims to replace. Against this, in the final chapter I aim to make clear that the arguments for the easy approach to ontology are purely pragmatic, empirical and conceptual—thus employing nothing but the demystified methodology the easy approach itself accepts. As a result, the view does retain the ‘epistemological high ground’ over mainstream metaphysics and avoid its mysteries. Overall, in Ontology Made Easy I aim to show that there is room for a deflationary meta-ontological approach, which avoids the mysteries of mainstream metaphysics and is an attractive competitor among deflationary views. I also aim to show that this view remains standing despite the common objections wielded against it. For me, this work has a deeper point. What I aim to do in defending the easy approach to ontology is not (as the positivists would have) to try to put an end to metaphysics. Nor is it just to place a distinctive deflationary meta-ontological position on the map. Rather than killing metaphysics, I aim to reorient it to more productive pursuits, leaving behind morass of ontological debates about existence. If ontology is easy, we should shift our attention from the neo-Quinean’s endless debates about existence questions. One area in which metaphysicians clearly have made, and can continue to make, contributions, is in conceptual work. This may of course involve conceptual analysis, as well as explication of how our actual conceptual system (taken holistically) works—but it needn’t be limited to that. For, more deeply, we may also assess what concepts we should use for various purposes (what concept (if any) of freedom, person, time, species, art. . . should we use?), and evaluate and (re-)engineer concepts where this is needed. Put in Carnapian terms: while existence questions (taken internally) are answerable easily, there remain the external questions of what language or BOOK SYMPOSIUM 227 concepts we should use, and how we should use them. Combining this thought with the easy approach to ontology brings us quite an attractive, modest picture of what the role of metaphysics can be: not engaging in quasi-scientific deep ‘discoveries’ about the world or its nature, but rather doing work on the conceptual side-on understanding how our concepts, or conceptual scheme at large, work—or how they should work. These, in my view, are the more interesting and tractable questions, where we can hope to make more progress—progress that matters. It is to such questions that I, for one, aim to turn my attention. Works cited Bennett, Karen. (2009). “Composition, Colocation, and Metaontology”, in Chalmers et al. Chalmers, David, Ryan Wasserman and David Manley, eds. (2009). Metametaphysics. Oxford: Oxford University Press. Chalmers, David (2009). “Ontological Anti-Realism”, in Chalmers, et. al. Hale, Bob and Crispin Wright (2009). “The Metaontology of Abstraction”. In Chalmers et al. (2001). The Reason's Proper Study: Essays towards a Neo-Fregean Philosophy of Mathematics. Oxford: Clarendon. Kriegel, Uriah (2013). “The Epistemological Challenge of Revisionary Metaphysics”. Philosopher's Imprint 13 (12): 1–30. Price, Huw. (2009). “Metaphysics after Carnap: The Ghost who Walks?”, in Chalmers et. al. Schiffer, Stephen (2003). The Things we Mean. Oxford: Oxford University Press. Sider, Theodore (2011). Writing the Book of the World. Oxford: Oxford University Press. Thomasson, Amie L. (2007). Ordinary Objects. New York: Oxford University Press. 228 AMIE L. THOMASSON