Superficial Dispositionalism

Abstract

Dispositional ascriptions do not entail the counterfactuals we might expect, as interfering factors may be poised to prevent the disposition from manifesting in its very stimulus conditions. Such factors are commonly called finks and masks. It is thought, however, that finks and masks cannot be intrinsic to the disposition bearer; if an intrinsic property of the object would prevent a particular response in certain conditions, the object fails to have the corresponding disposition. I argue that we should accept intrinsic finks and masks if we think there are finks and masks at all, and also if we maintain that paradigmatic dispositions are intrinsic. This last point is particularly problematic for the claim that there cannot be intrinsic finks and masks, for if paradigmatic dispositions are not intrinsic then the central argument for the impossibility of intrinsic finks and masks is undermined.

Notes

²I do not mean to imply that every dispositional ascription is equivalent to some single-track overt dispositional ascription: ⌈disposed to M if S ⌉. See Bird 2007: 18–24] for discussion of the difficulties of translating covert dispositional predicates like ‘fragile’ to ones overtly concerning stimulus-manifestation pairs. Although there may be no single stimulus-manifestation pair associated with a particular disposition, the type of dispositional interference considered only requires the falsification of one associated counterfactual: if x were to undergo S, then x would M. Though fragility-ascriptions may not be equivalent to the disposition-ascription disposed to break if struck, SD is meant to rule out the possibility of fragile glasses not breaking when struck, due to some intrinsic property. Also, note that throughout this paper I will slide between using the schematic letter ‘M’ in place of a noun and in place of a verb—treat ⌈to M⌉ and ⌈to give response M⌉ equivalently.

¹See Johnston 1992: 231–3], Martin 1994: 2–4], and Bird 1998: 228] for counterexamples to the claim that dispositional ascriptions entail their associated counterfactuals.

³See, for example, Handfield and Bird's 2008 response to Martin and Heil 1998, and Cohen and Handfield's 2007 response to Smith 2003. I have also heard similar objections raised to Armstrong's 1961 dispositional analysis of the content of perception.

⁴Photochromatic lenses work somewhat like this, although they respond to UV rather than visible light. For this case, simply imagine super-photochromatic lenses that, instead of turning dark upon UV exposure, turn dark upon the assault of any amount of any kind of light. Note that photochromatic lenses are different from polarized lenses, which do not themselves change in response to light, but instead only let through light waves of certain orientations.

⁵See Smith 1977: 440], Johnston 1992: 231, 232], and Martin 1994: 2–3].

⁶See Martin 1994: 3–4] for this example.

⁷See Johnston 1992: 233]. Alexander Bird 1998 instead calls this kind of property an ‘antidote’. This is potentially confusing, as an antidote can work in (at least) two different ways. An antidote for a poison might block the poison's ability to be absorbed by the poisonee's body, in which case it would mask the poison's disposition—but it might also work by removing the poison's disposition to kill, in which case it would be a fink. Both are perfectly good as an antidote. Bird notes that there is a difference between what he calls the philosophical use of ‘antidote’, and the ordinary use. The ordinary use of the word ‘mask’, however, implies that what is masked—in this case, the disposition—is not removed or altered.

⁸In this paper I am using the idea of intrinsicality most often found in the literature on dispositions—a property P is intrinsic to an object x (at world w) iff P is had by all of x's duplicates in worlds that have the same natural laws as w. This corresponds to a notion of intrinsicality that is referred to as ‘Kim+-intrinsic’ in [Humberstone 1996: 238]. What I say here is not meant to turn on a particular way of defining duplication, but see Langton and Lewis 1998: 336–7] for one definition of duplication: they define duplicates as those that share basic intrinsic properties, which are properties that are 1) independent of loneliness or accompaniment, 2) not disjunctive properties and 3) not negations of disjunctive properties.

⁹Note that dispositions aren't automatically extrinsic simply because we often use their stimulus/manifestation pair—which are generally extrinsic to the object—to pick out the disposition that we are referring to. Conceptual relationality concerns concepts, not properties directly. Dispositions may be conceptually relational (requiring reference to relations the object has to other extrinsic matters in order to specify the concept of the disposition), but, for all that, intrinsic.

¹⁰Note that, depending on what one thinks about the modal status of natural laws, there may be another intrinsic/extrinsic distinction that I am not appealing to. If the natural laws are contingent, then one might think there is a notion of intrinsic simpliciter, where a property is intrinsic if it is shared by any duplicate whatsoever (though if the laws are necessary then these two distinctions collapse into each other). Then dispositions may be extrinsic in this sense. However, this is not the intrinsic/extrinsic distinction that is in play in the dispositions literature.

¹¹See Choi 2003. For an example of this put to work, see Bird 2007: 24–5].

¹²The case referred to in this quotation comes from [Lewis 1997: 147]. A sorcerer finkishly protects a glass by standing by, ready to cast a spell which will change the glass's micro-structure to a non-fragile one, should the glass ever be about to be knocked or dropped.

¹³I take it that Handfield and Bird's intention in giving this example is not so much to argue decisively for SD, as to merely give it some plausibility—though this might be questioned given that following this discussion they say, ‘We can conclude therefore that as regards both fink and antidote cases if S contains an intrinsic fink or antidote to some disposition D, then S does not possess D' [Handfield and Bird 2008: 291].

¹⁴Though Choi does not clarify what it is to clearly possess the disposition in question, I assume that being able to manifest the disposition in the disposition's stimulus conditions is what is required.

¹⁵Actually, this is not correct, as we will see in §3.3.

¹⁶However, see Clarke 2008: 3–4] for an argument that Choi's tests give contradictory results in the case of intrinsic finks which have been extrinsically disabled.

¹⁷This was originally given simply as a case of finking, not particularly as an example of intrinsic finking. The fact that the shy intuitive chameleon is a case of intrinsic finking has been independently pointed out by Clarke 2008: 2].

¹⁸This is Handfield's response (personal correspondence).

¹⁹A possible reply is that what the Superficial Dispositionalists are really worried about are supposed finks and masks operating via a causal chain that is entirely intrinsic—the necessary condition in question should be: if the glass's enchantment were to operate via an entirely intrinsic causal chain, then the glass would not be fragile. I discuss this amendment to SD in Ashwell 2009.

²⁰Despite the name, Miraculin is not magical. There is a berry, Synsepalum dulcificum, or ‘miracle fruit’, which contains Miraculin. It is slightly different, however, from the kind of fruit in this example: it does not taste sweet, either because it does not have the acidity which is made to taste sweet by the Miraculin, or because the Miraculin takes some time to work.

²¹Again, if you are worried about the use of magic here, consider a variation on Miraculin where someone has a spray that would prevent the Miraculin from affecting the tastebuds.

²²In §4.1 I will consider a reply to this. It may be that, although these cases show that SD conflicts with the (paradigmatic) intrinsic dispositions thesis, and so even paradigmatic dispositions like fragility are extrinsic, there may still be a sense in which paradigmatic dispositions can be intrinsic to objects.

²³This reply was suggested by an anonymous referee for this journal.

²⁴See Martin 1994: 2–3] and Lewis 1997: 144]. Martin's original electro-fink is what is now commonly called a ‘reverse fink’. In this case, a wire that is not conductive is made so by an attached device, should the wire ever have a potential difference applied across it.

²⁵See Humberstone 1996: 227–8] for this distinction.

²⁶See Handfield and Bird 2008: 290-1] for this sort of objection, and this example.

²⁷The same thing holds for the fruit in Miraculin. Suppose that we have, in fact, made the part/whole mistake here. Then we have confused a disposition of a part—the sourness of the fruit sans Miraculin—for a disposition of the whole. But it need not be the case that there is a part of the fruit that is protein-free in the way that the uranium is rod-free. Furthermore, the protein might play a dual role in the fruit such that removing it would change the fruit so it was no longer sour, or its sourness might be intrinsically masked by some new feature of the fruit. So, (1) there need not be any physical part that doesn't contain Miraculin, and (2) if the Miraculin is removed, the fruit need not be still sour, and (3) even if the fruit with Miraculin removed is sour, it need not taste so (removing the Miraculin might make the fruit highly poisonous so that it kills before one has time to taste it). The fact that the Miraculin could play this kind of dual role shows that the mask is in fact some property had by the Miraculin protein, thus not a part, strictly speaking, of the fruit.

²⁸Thank you to two anonymous referees and the editor of this journal for extremely useful comments on earlier versions of this paper. I am also grateful to Richard Holton, Rae Langton, Alex Byrne, Alexander Bird, John Heil and Toby Handfield for particularly helpful discussions, and to audiences at the University of Auckland and Australasian Association of Philosophy.