Abstract
According to the conciliatory view in metaontology, there are multiple possible languages corresponding to the popular positions in ontology. In each of these languages, the term ‘exists’ expresses a distinct “existence-like” property, and consequently the claims associated with each of the rival ontological positions come out true in some such language. Species of the conciliatory view can be distinguished based on claims about how the various existence-like properties are related vis-à-vis metaphysical naturalness. On some versions, all of the existence-like properties are held to be equally natural, such that there is no “metaphysically privileged” language. However on others, there is held to be a most natural existence-like property and hence a most fundamental ontological language. In this paper I defend two views, one metaontological and one ontological. I first defend a version of the conciliatory view according to which there is exactly one perfectly natural existence-like property, but the rest are still “reasonably” natural. I then argue that, assuming this metaontological view, we should accept a permissive ontological view in the fundamental language (i.e. one with a very expansive domain).
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Notes
Carnap (1950), Putnam (1987), and Hirsch (2005) are influential proponents of the conciliatory view, but they might not put it in quite the way I do. There are other metaontological views, such as the agnosticism of Bennett (2009), that might also deserve to be called ‘conciliatory’ in some sense, but in this paper, I reserve that label for the more specific sort of view so defined above.
I assume that ‘exists’ is a first order predicate, definable in terms of the existential quantifier ‘something’ (i.e. ‘exists’ = df ‘being an x such that something y is identical to x’). As such, ‘exists’ expresses a (first order) property of individuals, while ‘something’ can be treated as a second order predicate expressing the (second order) property of properties being instantiated (i.e. ‘Something is a dog’ expresses the claim that being a dog is instantiated). Given these connections, a proponent of the conciliatory view will hold that, as the meaning of ‘exists’ varies from one ontological language to another, so does the meaning of any term analytically connected to it (e.g. ‘there is’, ‘something’, ‘all’, ‘instantiated’, etc.).
As I’ll explain in the next section, there are actually multiple possible languages corresponding to universalism, and multiple corresponding to nihilism.
Following Hirsch (2005), I assume that a dispute can be verbal even if both parties express the same thing by the problematic term. Otherwise, few disputes would come out verbal, since disputes typically take place between parties that are members of the same linguistic community and hence speak the same public language. According to Hirsch’s account, a dispute between two parties is verbal (roughly) if the relevant term would mean something different in hypothetical linguistic communities in which the dominant usage of that term corresponds to each parties’ usage. This account is motivated by an attempt to remain faithful to Burge’s (1979) social externalist insight, namely that meaning is (partially) determined by the pattern of usage in one’s wider community rather than merely private usage.
Hirsch (2005) argues that many ontological disputes, specifically those over the existence of “highly visible physical objects,” are merely verbal, while also maintaining that so-called “common sense” ontological positions are true in plain English.
Hirsch would likely hold this view, if not for the fact that he is somewhat skeptical of the application of naturalness to quantifier meanings (Hirsch 2013: pp. 715–716).
Sider (2009, 2011) defends such a position. One might complain that it is odd to classify Sider as a proponent of the conciliatory view, since he is one of the most strident defenders of ontological disputes being substantive. This defense (as I’ll explain in the next section) turns on the metasemantic doctrine of reference magnetism, which purports to secure shared reference of ‘exists’ for all ontologists to the most natural existence-like property (if such there be). However, Sider admits that, if this defense fails, he is happy to allow for a plurality of non-fundamental ontological languages, provided that each can be given a “metaphysical semantics” (Sider 2011: p. 112) in terms of the fundamental one.
Hirsch (2005), for instance, assumes that there are ontological languages corresponding to different positions in the disputes over mereological fusions and temporal parts, though not necessarily ones corresponding to the dispute between platonists and nominalists over abstract objects.
I assume here that there’s nothing incoherent about any of these combinations of views. However, there may be some combinations of ontological views that are incoherent, and we should thus not expect a conciliator to countenance an ontological language corresponding to such a position. For instance, perhaps combining presentism (i.e. only present things exist) with 4Dism is incoherent, in which case a conciliator would reject an ontological language corresponding to this combination.
Different conciliators may disagree on which views count as systematic enough, though I expect broad agreement on clear cases (e.g. mereological universalism and nihilism are systematic, but universalism minus is not).
The predicate ‘grue’ originated in Goodman (1955). An object is grue iff it is either green and discovered before some arbitrary future date (say, 3000 AD) or blue and not so discovered.
In general, it is always possible to perform systematic permutations on the intuitively intended interpretation that yield bizarre (and intuitively unintended) interpretations that nevertheless fit with our usage equally well.
In the next section, I’ll clarify and defend this claim that a language is rationally defective if its simple terms fail to express reasonably natural properties. See Hirsch (1993) for a more thorough discussion.
Giving a clear taxonomy of metaontological views can be tricky, since different theorists use terminology differently. Sider (2009), for instance, calls his view ‘ontological realism’, and intends it to contrast with the view he calls ‘ontological deflationism’, which he attributes to theorists like Hirsch (2005). However, Sider defines ‘ontological realism’ as the view that there is a most natural existence-like meaning, and so that’s compatible with the moderate or weak versions of the conciliatory view as I define them. Insofar as Sider thinks that reference magnetism will guarantee shared reference for all possible linguistic communities using a term that plays the “existence” role, then he counts as an opponent of the conciliatory view (denying the plurality of ontological languages). However, he is on record as being open to the possibility that reference magnetism will fail to secure such shared reference, and so his “Plan B” is to grant the plurality of ontological languages but shift his focus to Ontologese (Sider 2009: pp. 411–416, 2011: p. 74). Falling back on his “Plan B” would then make him count as a conciliator of the moderate or weak variety, accepting multiple ontological languages but a uniquely most natural one. It is also worth mentioning that Hirsch (2002), one of the most prominent contemporary conciliators, is very adamant that his conciliatory view, which he calls ‘quantifier variance’, does not stand in opposition to realism, at least construed as mind-independence. To suppose so is based on a use-mention confusion. After all, what exists does not depend on our minds, though what we mean by ‘exists’ does, and consequently so do the truth-values of sentences involving the term ‘exists’. To make this point, Hirsch (2002: pp. 52–52) reminds us of the old saying that if you called a leg ‘a tail’, this wouldn’t change the fact that dogs still only have one tail (though it would change the truth-value of the sentence, “Dogs only have one tail.”).
Although, as I’ll discuss later, theorists like Sider (2009) maintain that there are ways to put the conciliatory view that are more congenial to nominalists who reject properties altogether. Whether one can be conciliatory towards the dispute between platonists and nominalists itself is not something I will discuss.
We’ve already seen one argument for this claim, based on the idea that reference magnetism will guarantee shared reference of ‘exists’ for the alternative linguistic communities. But another argument, known as the ‘collapse argument’, purports to show that it is incoherent to suppose that there are multiple distinct quantifier meanings that are all inferentially adequate. See Hale and Wright (2009: pp. 183–184) for the argument, and see Turner (2010) and Warren (2015) for some responses.
This “gerrymanderedness-of-definitions” account of comparative naturalness comes from Lewis (1983: p. 347, 1984: p. 228, 1986: p. 61), who is arguably the father of all contemporary discussion of metaphysical naturalness. Although it has been challenged by many in the literature (Sider 1995; Williams 2007; Hawthorne 2007), I think most of the standard worries are overstated. See Williams (2005: §8.3), Sider (2011: pp. 130–131), and Guigon (2014: pp. 7–9) for responses to the common objections, most of which stem from mistakenly taking length to be the only factor that contributes to how gerrymandered a property’s definition is. Other factors include (some suitable measure of) complexity and how miscellaneous its constituents are.
Although Sider (2011) himself prefers a fundamentality-based conception of naturalness (or structure, in his preferred terminology), he elsewhere discusses the possibility of adopting an alternative, similarity-based conception (Sider 1993: §3.3). Other theorists instead adopt a primitivist account of naturalness, on which comparative naturalness is not reducible to definability or fundamentality (e.g. Dunaway and McPherson 2016: §3.5). There are also some completely different uses of ‘natural’, particularly in metaethics, on which it means something more like empirical or investigable (in principle) by the sciences.
In particular, my argument against the strong conciliatory view requires the identification of perfect naturalness with absolute fundamentality, and my argument for the permissive view in the fundamental language requires the “gerrymanderedness-of-definitions” account of comparative naturalness.
If one adopts the approach I mentioned in n 2 above, in which existence is definable in terms of existential quantification (construed as a second order property), then it is this latter property that is fundamental. I will ignore this complication throughout this paper and speak as if it is the existence-like property itself that is fundamental.
See Sider (2011: §9.6.2) for some support for this assumption, or for the challenges of doing without fundamental quantification.
Schaffer (2015: §4) defends this idea by analogy with ideological parsimony, or what he calls ‘conceptual economy’. If a theorist takes a concept as primitive, then she incurs an ideological cost. However, such a theorist incurs no new cost when she merely defines a new concept out of prior accepted ones.
One might think that parsimony similarly favors the weak view over the moderate view, since the moderate view holds all the rival existence-like properties to be reasonably natural while the weak view does not. However, on the stated account of parsimony, only fundamental properties (or entities more generally) detract from ontological parsimony, and reasonably natural properties are still less-than-perfectly natural and hence non-fundamental (i.e. definable).
Since all of the ontological languages are fundamental on the strong view, there is no such thing as the fundamental language, and so no such debate to be had.
Whichever version of the conciliatory view we favor, we can all countenance this property, being such that David Lewis would say it exists, at no ontological cost. After all, it is clearly non-fundamental. However, the point is that only the weak view holds that such a property could be existenceu, and thus that the community of universalists could be expressing such a gerrymandered mess by ‘exists’ rather than a more metaphysically significant phenomenon. The strong and moderate views instead would say that such a property, due to its unnaturalness, couldn’t be existenceu, despite being coextensive with it. It is worth pointing out that coextensive properties can differ greatly in naturalness—for instance, being negatively charged versus being-negatively-charged-and-part-of-a-donut-or-negatively-charged-and-not-part-of-a-donut.
See Sider (2011: preface) for a defense of the idea that the reasonableness of a description depends not only on its accuracy but also on the naturalness of the properties it expresses.
For simplicity, I will ignore the even more permissive view of modal plentitude, which countenances a multitude of coincident material objects corresponding to every (coherent) modal profile.
A trout-turkey is the mereological fusion of the front half of a trout and the back half of a turkey (Lewis 1991: pp. 7–8).
Eklund (2009: §9) gives a somewhat similar argument in favor of a permissive ontological view (though in English rather than in the fundamental language) based on the alleged asymmetry between the languages’ ability to give a systematic semantics for all of the alternative languages. He argues that only the permissive view, which countenances arbitrary fusions, can explain why, say, a universalist’s utterance of “Blob is in the yard,” comes out true, where ‘Blob’ is a name that purports to pick out an arbitrary fusion in the yard. I think there is an important connection between this sort of argument and my own. The more natural a property is, the more fundamental and hence metaphysically explanatory it is (i.e. capable of metaphysically defining more phenomena). Thus it is no coincidence that the most natural existence-like property existencef is capable of metaphysically explaining the most facts—in particular, the semantic facts about truth in all the alternative ontological languages—since the other existence-like properties (and facts about these properties) are all definable in terms of existencef.
Being simple can then be reductively analyzed using one’s preferred mereological primitive (typically parthood, proper parthood, or overlap) as well as other logical operators. Whether being alive is itself a fundamental property or can instead be reductively defined is not something I’m qualified to comment on.
One might wonder how the properties corresponding to other forms of restricted mereological composition fare when giving a reduction. Consider, for instance, the property existencec that corresponds to the more “commonsense” ontological view according to which “ordinary” composite objects exist (tables, chairs, people, etc.) but extraordinary ones (e.g. arbitrary fusions) do not. See Korman (2015) for a defense. It’s unclear whether such a view is systematic enough for existencec to count as “existence-like”. If it is, then the reduction may look something like: existingc is the property existingu and being sufficiently spatiotemporally and/or causally unified. If more than this is needed, perhaps the resources used in a plausible “series-style” answer to the special composition question can help with this reduction. See van Inwagen (1990: p. 63) for discussion. The inevitable vagueness in such a description will result in vagueness of ‘existencec’, though this is to be expected. It is widely acknowledged that most plausible restrictions on mereological composition would need to be vague and would result in a vague notion of existence. See Lewis (1986: pp. 212–213) for discussion.
To be fair, the sorts of theorists who are typically more inclined towards conciliatory views in general may be less likely to feel the need to appeal to naturalness on any of these topics. After all, the appeal to naturalness tends to be associated with “realism” in some form or other, while conciliators often—though not always (e.g. Hirsch 2002)—see themselves as opposed to “realism”. However, given how broad the proposed theoretical role of naturalness is, and given that one can (and should, in my view) be conciliatory about some disputes but not others, conciliators in metaontology should not be too quick to write off the notion of naturalness entirely. Thanks to an anonymous referee for encouraging me to address this point.
One complication arises in connection to ontological parsimony. Given that the stated account of parsimony appeals to fundamentality, this raises the question: Does parsimony demand that we avoid positing fundamental1 entities or fundamental2 (fundamental3, etc.) ones? If we go conciliatory about parsimony as well, then which should we appeal to when theorizing: parsimony1 or parsimony2 (parsimony3, etc.)? A further complication arises in spelling out how all the definability relations are related to one another with respect to all the definability relations. Is each definability relation indefinable in its own sense but definable in each different sense in terms of each different definability relation? For instance, is definability1 indefinable1 and yet definable2 in terms of definability2? What would such a definition2 look like? These are just some of the difficult questions that a conciliator about naturalness/fundamentality faces.
Korman (2015: §5.2) makes this sort of argument in response to claims that universalists often make, e.g. Lewis (1986: p. 213), that ordinary speakers are restricting their quantifiers to exclude extraordinary objects such as arbitrary fusions. To be fair, Lewis may not actually be guilty of making this claim himself. He says, “Restrict quantifiers, not composition” (1986: p. 213), which can be interpreted as an imperative targeted towards ontologists, rather than a linguistic hypothesis about actual speakers. Thanks to Dan Korman for making this point.
I should clarify that by ‘primitive’ I mean not synonymous with any complex expression. Thus a term can be primitive without denoting a fundamental entity and vice versa. Some writers, however, use ‘primitive’ and ‘fundamental’ interchangeably.
Here I am not assuming that positing objects is more costly, in terms of ontological parsimony, than positing properties. After all, on the stated account of parsimony, what makes a difference is not the property/object distinction but rather the fundamental/non-fundamental distinction. Insofar as the objects one posits are non-fundamental (e.g. tables, trees, arbitrary fusions, etc.), definable in terms of fundamental objects (e.g. particles, spacetime points, etc.), there is no ontological cost to positing them. The difference in controversiality between positing objects versus properties seems less to do with considerations of parsimony and instead more to do with the idea that “commonsense” takes a stronger stand on what objects exist, especially when it comes to material objects that are allegedly right in front of our eyes, than it does on what properties (or abstract entities more generally) exist.
See Sider (2011: especially §6) for an account of naturalness, or what he calls ‘structure’, that is compatible with nominalism. One might wonder what the nominalist analogue of the abundant conception of properties is, given that nominalists reject properties. I have in mind the view according to which one is very permissive about what true predications one can make. So, just as on the abundant conception of properties there are all sorts of bizarre properties like being grue in addition to “normal” ones like being green, on its nominalist analogue, there are true predications involving all sorts of bizarre terms like ‘grue’ just as there are true predications involving “normal” ones like ‘green’. So, a nominalist conciliator will hold that there are plenty of true predications one can make, including those involving the terms ‘existsu’, ‘existsn’, ‘existso’, etc., while rejecting that these terms express properties. Such a conciliator can then adopt Sider’s nominalist account of naturalness (i.e. structure) to take a position on the strong versus moderate versus weak dispute.
Thanks to Dan Korman, Teresa Robertson Ishii, and the other members of the Santa Barbarians* metaphysics reading group—arguably numerically distinct from the original Santa Barbarians reading group, but quite similar—for helpful feedback on earlier drafts. And thanks to two anonymous referees for many helpful comments and suggestions throughout the process.
References
Bennett, K. (2009). Composition, colocation, and metaontology. In D. J. Chalmers, D. Manley, & R. Wasserman (Eds.), Metametaphysics (pp. 38–76). Oxford: Clarendon Press.
Burge, T. (1979). Individualism and the mental. Midwest Studies in Philosophy, 4(1), 73–122.
Carnap, R. (1950). Empiricism, semantics, and ontology. Revue Internationale de Philosophie, 4(11), 20–40.
Correia, F. (2017). Real definitions. Philosophical Issues, 27(1), 52–73.
Dorr, C. (2005). What we disagree about when we disagree about ontology. In M. E. Kalderon (Ed.), Fictionalism in metaphysics (pp. 234–286). Oxford: Oxford University Press.
Dorr, C., & Hawthorne, J. (2013). Naturalness. In K. Bennett & D. W. Zimmerman (Eds.), Oxford studies in metaphysics (Vol. 8, pp. 1–70). Oxford: Oxford University Press.
Dunaway, B., & McPherson, T. (2016). Reference magnetism as a solution to the moral twin earth problem. Ergo: An Open Access Journal of Philosophy, 3(25), 639–679.
Eklund, M. (2009). Carnap and Ontological Pluralism. In D. J. Chalmers, D. Manley, and R. Wasserman (eds.), Metametaphysics (pp. 130-156). Clarendon Press.
Goodman, N. (1955). Fact, fiction, and forecast. Harvard: Harvard University Press.
Guigon, G. (2014). Overall similarity, natural properties, and paraphrases. Philosophical Studies, 167(2), 387–399.
Hale, B., & Wright, C. (2009). The metaontology of abstraction. In D. J. Chalmers, D. Manley, & R. Wasserman (Eds.), Metametaphysics (pp. 178–212). Oxford: Clarendon Press.
Hawthorne, J. (2007). Craziness and metasemantics. Philosophical Review, 116(3), 427–440.
Hirsch, E. (1993). Dividing reality. Oxford: Oxford University Press.
Hirsch, E. (2002). Quantifier variance and realism. Philosophical Issues, 12, 51–73.
Hirsch, E. (2005). Physical-object ontology, verbal disputes, and common sense. Philosophy and Phenomenological Research, 70(1), 67–97.
Hirsch, E. (2013). The metaphysically best language. Philosophy and Phenomenological Research, 87(3), 709–716.
Korman, D. Z. (2015). Objects: Nothing out of the ordinary. Oxford: Oxford University Press.
Lewis, D. (1983). New work for a theory of universals. Australasian Journal of Philosophy, 61(4), 343–377.
Lewis, D. (1984). Putnam’s Paradox. Australasian Journal of Philosophy, 62(3), 221–236.
Lewis, D. (1986). On the plurality of worlds. Hoboken: Wiley-Blackwell.
Lewis, D. (1991). Parts of classes. Hoboken: Blackwell.
Putnam, H. (1987). Truth and convention: On Davidson’s refutation of conceptual relativism. Dialectica, 40(1–2), 69–77.
Rosen, G. (2015). Real definition. Analytic Philosophy, 56(3), 189–209.
Schaffer, J. (2015). What not to multiply without necessity. Australasian Journal of Philosophy, 93(4), 644–664.
Sider, T. (1993). Naturalness, intrinsicality, and duplication. Ph.D. Thesis, University of Massachusetts.
Sider, T. (1995). Sparseness, immanence, and naturalness. Noûs, 29(3), 360–377.
Sider, T. (2001). Four dimensionalism: An ontology of persistence and time. Oxford: Oxford University Press.
Sider, T. (2009). Ontological realism. In D. J. Chalmers, D. Manley, & R. Wasserman (Eds.), Metametaphysics (pp. 384–423). Oxford: Clarendon Press.
Sider, T. (2011). Writing the book of the world. Oxford: Clarendon Press.
Turner, J. (2010). Ontological pluralism. Journal of Philosophy, 107(1), 5–34.
van Inwagen, P. (1990). Material beings. Oxford: Cornell University Press.
Warren, J. (2015). Quantifier variance and the collapse argument. The Philosophical Quarterly, 65(259), 241–253.
Williams, J. R. G. (2005). The inscrutability of reference. Ph.D. Thesis, University of St Andrews.
Williams, J. R. G. (2007). Eligibility and inscrutability. Philosophical Review, 116(3), 361–399.
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Mokriski, D. Conciliatory metaontology, permissive ontology, and nature’s joints. Synthese 199, 2335–2351 (2021). https://doi.org/10.1007/s11229-020-02887-7
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11229-020-02887-7