Hostname: page-component-77c89778f8-m42fx Total loading time: 0 Render date: 2024-07-18T15:52:46.995Z Has data issue: false hasContentIssue false

Against Conservatism in Metaphysics

Published online by Cambridge University Press:  03 July 2018

Maegan Fairchild*
University of Southern California
John Hawthorne*
University of Southern California


In his recent book, Daniel Korman contrasts ontological conservatives with permissivists and eliminativists about ontology. Roughly speaking, conservatives admit the existence of ‘ordinary objects' like trees, dogs, and snowballs, but deny the existence of ‘extraordinary objects', like composites of trees and dogs (‘trogs'). Eliminativists, on the other hand, deny many or all ordinary objects, while permissivists accept both ordinary and extraordinary objects. Our aim in this paper is to outline some of our reasons for being drawn to permissivism, as well as some of our misgivings about conservative metaphysics. In the first section, we discuss a tempting epistemic line of argument against conservatism. This isn’t a line of argument we find especially promising. Our most basic complaint against conservatism is not that conservatism has poor epistemic standing even if true, but instead that conservatism is weird. We develop this thought in the second part of the paper. In the final section we discuss some larger methodological issues about the project of ontology.

Copyright © The Royal Institute of Philosophy and the contributors 2018 

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)


We are grateful to Daniel Korman, Gabriel Uzquiano, Cian Dorr, and an audience at the Royal Institute of Philosophy for feedback on earlier versions of this paper.


2 There are also lots of ways to be an ‘eliminativist’, though we won't get into that here.

3 Korman, , Objects: Nothing Out of the Ordinary (Oxford University Press, 2015), 13Google Scholar, emphasis added. See also e.g. Hirsch, , ‘Against Revisionary Ontology’, Philosophical Topics 30(1) (2002): 107CrossRefGoogle Scholar, 117.

4 See especially Korman, and Carmichael, (‘What Do Folk Think about Composition and Does it Matter?’, in Experimental Metaphysics (Bloomsbury Academic, 2016): 187206)Google Scholar, as well as Korman's discussion of ‘folk capitulation’ (Objects: Nothing Out of the Ordinary, 59–62). See also Rosen, and Dorr, Composition as a Fiction’, in Gale, R. (ed.), The Blackwell Companion to Metaphysics (Blackwell, 2002)Google Scholar, §4 for the observation that Universalism isn't ‘manifestly incompatible with what we take for granted’.

5 Korman, Objects, 13.

6 From Hawthorne, , Metaphysical Essays, (Oxford University Press, 2006), 53Google Scholar.

7 Other versions of permissivism that allow for a plenitude of coincident objects can be found in, for example, Fine, K., ‘Things and Their Parts’, Midwest Studies in Philosophy 23(1) (1999): 6174Google Scholar; Yablo, S., ‘Identity, Essence, and Indiscernibility’, Journal of Philosophy 84(6) (1987): 293314Google Scholar; Bennett, K.Spatio-temporal Coincidence and the Grounding Problem’, Philosophical Studies 118(3) (2004): 339371Google Scholar; Thomasson, A.T. Ordinary Objects (Oxford University Press, 2007)Google Scholar, §9.6 and 10.3; Thomasson, A.T., Ontology Made Easy (Oxford University Press, 2015)Google Scholar §6.1;  Eklund, M., ‘Neo-Fregean Ontology’, Philosophical Perspectives 20(1) (2006): 95121Google Scholar; Leslie, S-J., ‘Essence, Plenitude, and Paradox’, Philosophical Perspectives 25(1) (2011): 277296Google Scholar; Inman, R., ‘Neo-Aristotelian Plenitude’, Philosophical Studies 168(3) (2014): 583597Google Scholar; Jago, M., ‘Essence and the Grounding Problem’, in Reality Making (Oxford: Oxford University Press, 2016), 99120Google Scholar.

8 Korman, Objects, 17.

9 See Fairchild, ‘Unlimited Ontology and Its Limits’ (PhD Thesis), for discussion of the varieties of permissivism.

10 Except in very special cases, a doxastic explanation is unavailable.

11 Another way of relaxing things is to allow that the beliefs that form the justificatory basis of the belief is explanatorily connected, even if the resulting belief is not. For discussion of a version of this idea (given an extra internalist tweak), see Korman ‘Debunking Arguments in Metaethics and Metaphysics’, in A. Goldman & B. McLaughlin (eds), Metaphysics and Cognitive Science (forthcoming).

12 An aside: Korman's discussion focuses on justification. We think it is better to begin with the question whether there is an explanatory requirement on knowledge. (If not, then it is hard to see why believing one is absent should defeat knowledge.) But we shall not argue for our preferred order of inquiry here.

13 Lewis, David, ‘Causal Explanation’, in Philosophical Papers Vol. II (Oxford University Press, 1986), 214240Google Scholar.

14 Albeit one that he refines in ‘Debunking Arguments in Metaethics and Metaphysics’ (forthcoming).

15 Of course, there are many subtle decision points related to the safety approach, including how to control for methods, whether to impose counterclosure, etc. We are not going to pursue these questions here.

16 Dorr, and Hawthorne, (‘Semantic Plasticity and Speech Reports’, Philosophical Review 123(3) (2014): 281338)Google Scholar explore this kind of multiplicity view at length.

17 Granted, there are some residual questions about lines of efficient causation. When one looks at a hammer, do the hammer, hammer*, and hammer** all cause the experience in a way that is causally over determines the experience? Or do certain of the coincident things play a special role viz a viz causation? Again, we won't purse those questions here – we don't think that knowledge about hammers turns on any particular answer. Here's a helpful analogy: suppose we think that beliefs about holes in cheese are caused by hole-linings and not holes, or that it is the table surface rather than the table that, strictly speaking, causes the experience as of a table. These hypotheses do not seem to involve a kind of deviance that would block the acquisition of knowledge of the presence of holes or tables by vision.

18 Verbs like ‘aware’ also take noun phrases (at least if we insert a preposition): ‘She is aware of the cat.’ Such constructions do not require knowledge. Might a verb plus noun phrase construction be more helpful to Korman's purposes, picking out a state that can explain knowledge without presupposing it? One thing to note here is that when used non-idiomatically, the noun-phrase occurs transparently: someone can be aware of Michelle Obama's best friend without having any knowledge of who is friends with who. On a transparent reading, being aware of the truth that P can occur even if one has no clue about the truth value of that proposition. Granted, the normal use of ‘noticing the fact that P’, ‘being aware of the truth that P’ do not seem to be compatible with just ignorance. But that is because these are merely idiomatic ways of conveying that someone is aware that P, noticed that P and so on.

19 Korman, Objects, 112.

20 See Korman Objects, 106 fn.30, for one attempt to say what deviance amounts to.

21 Hawthorne, Metaphysical Essays, 109.

22 Cf. Korman's Birds (Objects, 100) and Citrus Detector (Objects, 109). See his discussion for more on the problems with this kind of reasoning.

23 This line of argument is extremely pervasive. See, for example, Yablo ‘Identity, Essence, and Indiscernibility’, 307; van Inwagen, , Material Beings (Cornell University Press, 1990), 6669Google Scholar, 126; Sosa, E.Subjects Among Other Things’, Philosophical Perspectives 1 (1987)CrossRefGoogle Scholar, 178; Sosa, E., ‘Existential Relativity’, Midwest Studies in Philosophy 23(1) (1999), 178CrossRefGoogle Scholar; K. Bennett ‘Spatio-temporal Coincidence and the Grounding Problem’; Hawthorne, Metaphysical Essays, vii; van Cleve, J.The Moon and Sixpence: A Defense of Mereological Universalism’, in Sider, T., Hawthorne, J. and Zimmerman, D.W. (eds), Contemporary Debates in Metaphysics (Blackwell, 2008), 323333Google Scholar; S-J. Leslie, ‘Essence, Plenitude, and Paradox’, 281; Fairchild, M.A Paradox of Matter and Form’, Thought: A Journal of Philosophy 6(1) (2017), 34Google Scholar.

24 Van Inwagen, Material Beings, ch 2.

25 Korman characterizes ‘ontologically significant differences’ between a pair of cases as differences that ‘could account for’ why one is a case of existence and the other a case of non-existence (Objects, 124). On one way of reading this, ‘could’ is epistemic and ‘account for’ means something like ‘provides an explanation’, but we are unsure whether a special purpose notion of metaphysical explanation is in play. We are much less comfortable with making some notion of explanation the theoretical centerpiece of this dispute (especially if unexplicated). However, we don't think that this construal ultimately alters the dialectic very much: the Tabler will say that the presence of the property being arranged tablewise in one case but not the other does explain the difference between the cases. Of course, we think we know it doesn't, and so couldn't explain differences (in the epistemic sense connected to knowledge). Similarly, by our lights it is metaphysically impossible that this sort of difference explain divisions in the way the Tabler suggests. And so we have the very kind of standoff envisaged in the text. It is not as if the proposition that nothing could account for a certain difference is common ground between us and the Tabler.

26 It bears emphasis that when we say a theory is bizarre, we do not mean to imply that no one could have a reason for embracing it (especially if we are liberal enough to count ‘intuitions’ or ‘seemings’ as reasons). Insofar as arbitrary theory choices are ones that lack reasons we thus prescind from calling bizarre theory choices arbitrary.

27 Perhaps it is better to say ‘destroyed when the constituting rock is completely submerged’.

28 The difference in phrasing here is not meant to track a significant difference between arguments A and B. Rather, we use the vocabulary of essential properties here because the target modal / persistence profile is harder to concisely characterize than the one at issue in A1.

29 Neither sample appears exactly as written in Korman's discussion. Here's his presentation of what we have called argument A:

  1. (AR1)

    (AR1) There is no ontologically significant difference between islands and incars.

  2. (AR2)

    (AR2) If so, then: if there are islands, then there are incars.

  3. (AR3)

    (AR3) There are islands.

  4. (AR4)

    (AR4) So, there are incars.

Put this way, Korman describes his response as a denial of (AR1). However, we find it most helpful to articulate the real locus of our disagreement by presenting the argument (and his response) as we do above.

30 By analogy, when a Lord Mayor stops being Lord Mayor it is not prima facie plausible that something stops existing.

31 Korman offers two diagnoses of this case that are open to the conservative: either the conservative can grant, as we we've suggested, the existential premise – there are snowballs, and snowballs do have the proposed modal profiles – and deny that any violations of parity result. But they might instead handle snowballs and snowdiscalls case more on the model of islands and incars: ‘On the other hand, it's plausible that the snowball is just a clump of snow. When you mash the snowball into a pancake, it ceases to be a snowball. But it doesn't cease to exist, since the clump of snow that it is doesn't cease to exist.’ (Objects, 157) Korman also considers other cases in this chapter that he thinks are clearly better handled in the former way (e.g. statutes and gollyswogles (ibid., 152), pieces of driftwood and wineracks (ibid., 155)).

32 Korman, Objects, 156–157.

33 Korman, Objects, 155.

34 And this is so whatever the details of ‘substantial’ turn out to be – Korman acknowledges that those may be tricky to work out, but the point here stands however the conservative might do so.

35 Of course as plentitude lovers we are happy to allow that there are certain things that require substantial alteration of underling matter to come into existence rather than not-quite substantial. But we are also happy to allow that there are things that require not-quite substantial as opposed to not-quite not-quite substantial. Thus we do not end up treat the line between substantial and non-substantial alterations as metaphysically special.

36 See Sider, , Four-Dimensionalism: An Ontology of Persistence and Time (Oxford University Press, 2001)CrossRefGoogle Scholar.

37 Korman favor approaches to ontology that go in for vagueness, including vagueness in existence and composition. But it is hard for us to figure out exactly what he himself would say about the current line of thought, since he never provides a logic. For example, he doesn't seem to take a clear stand on the status of excluded middle. But it is hard to fill out a view when such questions are left open.

38 Although ‘Strange Kinds, Familiar Kinds, and the Charge of Arbitrariness’ (Oxford Studies in Metaphysics (2010): 119–144) Korman describes himself as a particularist in the sense of Markosian, N. (‘Brutal Composition’, Philosophical Studies 92(3) (1998): 211249)Google Scholar, he disavows that in Korman (Objects, 24). Korman's brand of particularism is elegantly illustrated in his discussion of Cook Ting (Objects, 157–159).

39 Korman, Objects, 139.

40 Korman, Objects, 147–148.

41 Physics need not be special in this respect. The special and social sciences often teach us that apparent differences fall away on closer inspection.

42 We note in passing that expressions like ‘that puddle’ seem to pass Korman's diagnostics for genuine singular terms (Objects, 145) with flying colors.

43 Korman (‘Strange Kinds, Familiar Kinds, and the Charge of Arbitrariness’), although hostile to scattered individuals – at least in the absence of creative intentions – seems much more amenable to scattered stuffs. That idea is not salient in Objects: Nothing Out of the Ordinary.

44 Carmichael (Toward a Commonsense Answer to the Special Composition Question’, Australasian Journal of Philosophy 93(3) (2015): 475490Google Scholar) invokes the notion of ‘dynamic equilibrium’ from Hoffman and Rosenkrantz (Substance: Its Nature and Existence (Routledge, 1997)Google Scholar) as just one component of his series-style response to the special composition question. Hoffman and Rosenkrantz define a relation of dynamic equilibrium as one that holds between two objects iff (a) there are attractive forces between x and y, (b) there are repulsive forces between x and y, and the attractive forces of (a) and the repulsive forces of (b) are in equilibrium (Substance: Its Nature and Existence, 82).

45 See Thomasson, (‘Realism and Human Kinds’, Philosophy and Phenomenological Research 67(3) (2003), 3.1)CrossRefGoogle Scholar for discussion of these questions and closely related cases.

46 See Fairchild, ‘Unlimited Ontology and Its Limits’.

47 See e.g. Hawthorne, J. & Uzquiano, G., ‘How Many Angels Can Dance on the Point of a Needle? Transcendental Theology Meets Modal Metaphysics’, Mind 120(477) (2011): 5381Google Scholar.