Abstract
A number of authors have suggested that a conditional analysis of dispositions must take roughly the following form:
Thing X is disposed to produce response R to stimulus S just in case, if X were exposed to S and surrounding circumstances were auspicious, then X would produce R.
The great challenge is cashing out the relevant notion of ‘auspicious circumstances’. I give a general argument which entails that all existing conditional analyses fail, and that there is no satisfactory way to define ‘auspicious circumstances’ just in terms of S, R, and X. Instead, I argue that the auspicious circumstances C for the manifestation of a disposition constitute a third irreducible element of that disposition, and that to pick out (or to ‘individuate’) that disposition one must specify C along with S and R. This enables a new conditional analysis of dispositions that gives intuitively satisfying answers in cases that pose problems for other approaches.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Notes
This formulation of the simple conditional analysis is very similar to one considered by Lewis (1997, p. 143).
This paper will focus upon what Prior et al. (1982) call “surefire dispositions”: ones that involve only a single (type of) response. One might wish eventually also to have an account of “probabilistic” dispositions that involve probabilities for multiple (types of) responses. E.g., a fair coin might be said to have a disposition when flipped to produce heads 50% of the time and tails 50% of the time. Surefire dispositions will raise enough puzzles to keep us busy without needing to track the added complexities of probabilistic dispositions. However, the moves we consider regarding surefire dispositions will have obvious analogs in the debate about probabilistic dispositions.
Following much of the literature, I draw upon our intuitive assessment of what conditionals are true in these cases. Some sophisticated accounts of conditionals might deny these intuitive assessments and hence evade these counterexamples. Bonevac et al. (2006) discuss constraints such sophisticated accounts would need to meet in order to avoid these counterexamples. Such maneuvering offers little hope to the popular Lewis/Stalnaker account (which analyzes “If P then Q” as roughly “The nearest P-worlds are Q-worlds”). Supposing I already have the antidote in my mouth, then the “nearest” worlds where I ingest the poison will pretty clearly be worlds in which I also ingest the antidote, and hence I would not become ill in those worlds. But this maneuver might offer more hope to other understandings of the conditional. E.g., Asher (1995) and Asher and Morreau (1995) analyze “If P then Q” as roughly “The normal P-worlds are Q-worlds.” It’s hard to say (and Asher and Morreau offer sadly little guidance) what would be a “normal” world for me to ingest a poison in. I suspect (and selfishly hope) that, at least for some poisons, this would be a world in which the antidote is close at hand, but some fans of an Asher/Morreau conditional analysis might hope that, for every poison I might ingest, antidotes will “normally” not be available, which would prevent antidotes from being a “blocking” counter-example to their version of a conditional analysis. Regardless, my argument in Sect. 3 will show that even their sort of conditional analysis is wrong.
This general idea has been advanced by many authors under various labels other than my “auspicious circumstances.” These include Mumford’s “ideal conditions” (1998, pp. 88–90), Bird’s “normal circumstances” (1998, pp. 233–4), Malzkorn’s “normal conditions” (2000, pp. 456–459), Gundersen’s “standard conditions” (2002, p. 407), Cross’s “background conditions” (2005, p. 324), and Choi’s “ordinary conditions” (2009, p. 576).
Where I can do so without confusion, I will abbreviate “produce a response of type R” as “produce R”. I will do similarly for other variables.
Many formulations of conditional analyses also include time t. For simplicity, I have omitted that, as issues about object persistence through time are largely orthogonal to the present topic. However, if one does include t in the analysandum, then t should also be fair game as an argument to F. This would allow that the auspicious circumstances for a given object to display a given disposition might change over time.
Some versions of a two-parameter approach might even allow that auspicious circumstances for manifesting a particular disposition might differ even for duplicate objects found in different circumstances (e.g., if the natives of two islands have different diets, it could be that auspicious circumstances for a mushroom on one island to manifest poisonousness are quite different from auspicious circumstances for a duplicate mushroom on the other island to manifest poisonousness). However, allowing that the auspicious circumstances for an object to manifest a disposition depend in part upon features extrinsic to that object (while accepting the modified conditional analysis proposed above) leads to the conclusion that dispositions are extrinsic properties of objects, a conclusion that many theorists oppose—see Sect. 6 below.
This is a case that motivates Lewis’ (1997) proposal.
More carefully: on Fara’s account, “This alarm is disposed to go off when someone enters through the window” gets cashed out as, “In virtue of some of its intrinsic properties, this alarm (habitually) goes off when someone enters through the window.” But it may be that, due to the electronic proficiency of local criminals, the alarm doesn’t (habitually) go off when people enter through the window, even though it clearly is now disposed to do just that.
This does not mean that dispositions themselves are context-dependent. It would be very strange indeed if what dispositions my vase actually has changes in tandem with changes in conversational context—and this would be a flagrant violation of the commonly accepted claim that dispositions are intrinsic (see Sect. 6 below).
One might suspect that in most actual cases, it will turn out that people actually have some sort of subtle disagreement about what sort of stimulus or response is relevant. However, all I need is one possible example where they do have exactly the same stimulus and response in mind. Perfect agreement may be exceedingly rare in actuality, but as long as it is possible, that’s enough for my argument.
Choi proposes a version of Alternative #1 which takes the complex stimulus for fragility to be “being struck under the ordinary circumstances for fragility” (Choi 2009, p. 576).
In theory, one could take any n-parameter proposal involving parameters P 1, P 2, … P n and propose a “one-parameter alternative” that takes the ordered n-tuple <P 1, P 2, … P n > as its “single” parameter, and then unpacks P 1, P 2, … P n out of this n-tuple so that they can play distinct roles in the analysans. This “alternative” might, in some technical sense, be a one-parameter proposal. But there is another good sense in which this “alternative” is really just an n-parameter proposal trickily disguised. As we’ll see in a moment, I think Alternatives #1 and #2 are similarly just trickily disguised version of my own three-parameter proposal.
In a later book, Bird (2007, pp. 36–39) argues that a conditional analysis of covert dispositional locutions (like “is fragile”) might be defended from various counterexamples by holding that these covert locutions are best translated into overt dispositional locutions involving complex stimuli as in Alternative #1 (e.g., “is disposed to break if struck when not encased in foam”), rather than the relatively simple stimuli (“is disposed to break if struck”) they’ve traditionally been taken to involve. I agree that the mapping from covert dispositional locutions to overt ones might be quite complex and context-sensitive. E.g., I think the sort of stimulus that is typically involved in talk of “fragile vases” is quite different from that involved in talk of “fragile parchment”, and both are quite different from that involved in talk of “fragile peace accords”. However, these complexities are of no help in response to cases like those in Sects. 2 and 3, as these cases hinged upon no covert dispositional locutions, but only overt ones like “This alarm is disposed to go off if someone enters through the window” or “This vase is disposed to break if struck.” So, contra Bird, I think the fan of conditional analyses can make little headway by attending to the complexities in translating from covert to overt dispositional locutions. Instead, the pressing philosophical problem is in determining how auspicious circumstances are determined even in overt dispositional locutions.
I am not opposed, however, to holding that the word ‘fragility’ involves a hidden contextual parameter that specifies force of striking. I would then say that fragility simpliciter is not a disposition; instead, the term ‘fragility’ is used, in different contexts, to pick out different dispositions involving (among other things) different degrees of forcefulness in their characteristic stimuli. Insofar as this is all Prior intended to claim, I’m not at odds with her. I am at odds with the Prior-esque view that each such linguistic parameter should correspond to a distinct parameter in our general account of the nature and individuation of dispositions.
I am also inclined toward a “no-backtracking” account of “nearness” akin to the one proposed by Lewis (1979), upon which the “nearest” striking worlds are ones that match the events of the actual world up to the time of the striking, that perhaps include a small miracle to enable the striking to occur, and that subsequently match the laws (but not necessarily the ensuing events) of the actual world.
C.f., Molnar (1999, p. 3).
This argument is an instance of the general methodology I call Pragmatic Conceptual Analysis. This methodology first identifies useful work we’ve been calling upon a concept to do (e.g., to identify conditionals that will stably remain true of an object) and then analyzes the concept in a way that would best enable it to continue that work (e.g., in a way that demands that dispositions are intrinsic). I discuss the merits of this general methodology in Fisher (2006).
For the purposes of this paper, I remain neutral regarding categorical bases. I am open to the idea that there might be some brute dispositions with no categorical bases, and that objects might have other dispositions at least partly in virtue of the brute dispositions that they (or their parts) possess. It is doubtful whether such brute dispositions could be ontologically reduced to conditionals, but a proposed analysis (like my Extrinsic Proposal, or my soon-to-be-advanced Intrinsic Proposal) might still state a true biconditional linking brute dispositions to conditionals. (I will return in the concluding section to the question of what sort of analysis of dispositions we should expect.)
Possible notions of intrinsicness include independence from properties instantiated outside the spatial region occupied by an object (Dunn 1990), independence from relations to distinct objects (Francescotti 1999), or dependence upon the qualitative properties that an object shares with all its duplicates (Lewis 1986). It is controversial which of these general notions captures “the” meaning of ‘intrinsic’, and it is also controversial what the best formal way of analyzing any of these notions is. Candidate analyses include naturalness theories which take (at least a core set of) intrinsic properties to be among some privileged class of ‘natural’ properties (Lewis 1986), and combinatorial theories which take intrinsic properties to be ones that can be freely combined with various particular sorts of extrinsic configurations (Vallentyne 1997; Langton and Lewis 1998; Weatherson 2001; Witner et al. 2005).
We’ll consider one potential threat to my neutrality below: the treatment of de re or haecceitistic properties, which some accounts take to be intrinsic, even though it’s doubtful these properties are of relevance to dispositions.
Perhaps the most plausible account of strict conditionals would be a version of an Asher/Morreau “normality” account which takes “If A then C” to mean “The most normal A-worlds are C-worlds”, and which takes “normality” to be an entirely a priori matter that does not depend at all upon the actual surroundings of the objects in question. However, I doubt that it’s an a priori matter, for example, what sorts of circumstances are “normal” (in any ordinary sense of the term) for caterpillar development. Instead, it seems to me, this is determined by the actual ecological circumstances that evolutionarily successful caterpillars have faced. So, insofar as I’m attracted to a “normality” account at all, I’m attracted to an “unstrict” version that allows a role for such external factors in determining what counts as “normal”.
Thanks to an anonymous reviewer for suggesting this example.
This includes dispositions where C takes the trivially unrestrictive setting: “any type of circumstance, whatsoever”. Dispositions involving trivially unrestrictive settings for C are circumstance-insensitive in that something that possess the disposition to R when S in any circumstance whatsoever will produce R if it receives S, regardless of the surrounding circumstances. Most circumstance-insensitive dispositions that are actually instantiated will either have an incredibly restrictive S that builds in the restrictions that we usually count upon C to place (much as in alternative #1 in Sect. 4), or else an incredibly liberal R that allows for many strange responses in strange circumstances (much as in alternative #2). While I countenance circumstance-insensitive dispositions among the many dispositions that are instantiated, I think that ordinary people typically refer to circumstance-sensitive dispositions instead.
For a good overview see Horgan and Potrč (2008, Chap. 2).
While this paper tentatively accepts an unrestricted view of dispositions, it remains neutral regarding unrestricted composition. Many people are attracted towards nominalism about properties but not objects. If property nominalism allows us to posit large numbers of dispositions at little intuitive or ontological cost, one might plausibly embrace an unrestricted view of dispositions while holding a more restricted (or even nihilist) view of object composition.
Thanks to an anonymous referee for voicing this worry.
I remain neutral regarding whether conditionals are, in some sense, more fundamental than dispositions. One attractive view takes a Humean manifold of categorical properties to be basic, defines laws and conditionals in terms of this manifold, and then defines dispositions in terms of conditionals. One coherent alternative reverses this order of dependence, taking some brute dispositions to be basic, and defining laws and conditionals in terms of those. A third option would be to take none of these to be ontologically prior, and instead view dispositions, conditionals, and laws as a family of interdefinable modal notions. The aim of my “analysis” is just to state a true biconditional capturing the relation between dispositions and conditionals, not to sort out their order of ontological dependence.
I also remain officially neutral regarding the question of whether my proposed analysis can be confirmed just by careful reflection upon folk concepts, or whether it is instead empirically supported by our experience with various dispositions and conditionals. For this paper, it’s enough just to find an analysis that is true.
References
Armstrong, D. M. (1968). A materialist theory of the mind. London: Routledge.
Armstrong, D. M. (1973). Belief, truth, and knowledge. Cambridge: Cambridge University Press.
Asher, N. (1995). Commonsense entailment: A conditional logic for some generics. In G. Crocco, L. F. del Cerro, & A. Herzog (Eds.), Conditionals: From philosophy to computer science (pp. 103–146). Oxford: Clarendon Press.
Asher, N., & Morreau, M. (1995). What some generic sentences mean. In G. Carlson & J. Pelletier (Eds.), The generic book (pp. 300–338). Chicago: University of Chicago Press.
Bird, A. (1998). Dispositions and antidotes. Philosophical Quarterly, 48, 227–234.
Bird, A. (2007). Nature’s metaphysics: Laws and properties. Oxford: Oxford University Press.
Bonevac, D., Dever, J., & Sosa, D. (2006). The conditional fallacy. Philosophical Review, 115, 273–316.
Boyd, R. (1988). How to be a moral realist. In Sayre. McCord (Ed.), Essays on moral realism (pp. 181–228). Cambridge: Cambridge University Press.
Burge, T. (1979). Individualism and the mental. In P. French, T. Uehling, & H. Wettstein (Eds.), Studies in metaphysics. Minneapolis: University of Minnesota Press.
Choi, S. (2009). The conditional analysis of dispositions and the intrinsic dispositions thesis. Philosophy and Phenomenological Research, 78, 568–590.
Cross, T. (2005). What is a disposition? Synthese, 144, 321–341.
Dretske, F. (1988). Explaining behavior. Cambridge: MIT Press.
Dunn, J. M. (1990). Relevant predication 2: Intrinsic properties and internal relations. Philosophical Studies, 60, 177–206.
Evans, G. (1973). The causal theory of names. In A.P. Martinich (Ed.), The Philosophy of Language. New York: Oxford University Press.
Fara, M. (2001). Dispositions and their ascriptions. Dissertation, Princeton University, Princeton.
Fara, M. (2005). Dispositions and habituals. Noûs, 39, 43–82.
Fisher, J. (2006). Pragmatic conceptual analysis. Dissertation, University of Arizona, Tucson.
Fodor, J. (1990). A theory of content and other essays. Cambridge: MIT/Bradford.
Francescotti, R. (1999). How to define intrinsic properties. Noûs, 33, 590–609.
Gundersen, L. (2002). In defence of the conditional account of dispositions. Synthese, 130, 389–411.
Horgan, T., & Potrč, M. (2008). Austere realism: Contextual semantics meets minimal ontology. Cambridge: MIT Press.
Jackson, F. (1998). From metaphysics to ethics: A defense of conceptual analysis. New York: Oxford University Press.
Johnston, M. (1992). How to speak of the colors. Philosophical Studies, 68, 221–263.
Kripke, S. (1972). Naming and necessity. Cambridge: Harvard University Press.
Langton, R., & Lewis, D. (1998). Defining ‘intrinsic’. Philosophy and Phenomenological Research, 58, 333–345.
Lewis, D. (1973). Counterfactuals. Oxford: Blackwell.
Lewis, D. (1979). Counterfactual dependence and time’s arrow. Noûs, 13, 455–476.
Lewis, D. (1984). Putnam’s paradox. Australasian Journal of Philosophy, 62, 221–236.
Lewis, D. (1986). On the plurality of worlds. Oxford: Blackwell.
Lewis, D. (1997). Finkish dispositions. Philosophical Quarterly, 47, 143–158.
Malzkorn, W. (2000). Realism, functionalism and the conditional analysis of dispositions. Philosophical Quarterly, 50, 452–469.
Martin, C. B. (1994). Dispositions and conditionals. Philosophical Quarterly, 44, 1–8.
McKitrick, J. (2003). A case for extrinsic dispositions. Australasian Journal of Philosophy, 81, 155–174.
Millikan, R. (1984). Language, thought, and other biological categories. Cambridge: MIT Press.
Molnar, G. (1999). Are dispositions reducible? Philosophical Quarterly, 49, 1–17.
Mumford, S. (1998). Dispositions. Oxford: Oxford University Press.
Prior, E. (1985). Dispositions. Aberdeen: Aberdeen University Press.
Prior, E., Pargetter, R., & Jackson, F. (1982). Three theses about dispositions. American Philosophical Quarterly, 19, 251–257.
Putnam, H. (1973). Meaning and reference. In A.P. Martinich (Ed.), The Philosophy of Language (3rd ed.). New York: Oxford University Press (1996).
Reiter, R. (1980). A logic for default reasoning. Artificial Intelligence, 13, 81–132.
Russell, B. (1905). On denoting. In A. P. Martinich (Ed.), The Philosophy of Language. (3rd ed., pp. 199–207). Oxford: Oxford University Press (1996).
Shoemaker, S. (1980). Causality and properties. In Peter. van Inwagen (Ed.), Time and cause (pp. 109–135). Dordrecht: Reidel.
Sider, T. (1996). Intrinsic properties. Philosophical Studies, 83, 1–27.
Smith, M., & Stoljar, D. (1998). Global response-dependence and noumenal realism. The Monist, 81, 85–111.
Stalnaker, R. (1968). A theory of conditionals. In Studies in Logical Theory, American Philosophical Quarterly Monograph Series, 2 (pp. 98–112). Oxford: Blackwell.
Strawson, P.F. (1950). On referring. In A. P. Martinich (Ed.), The Philosophy of Language (3rd ed., pp. 215–230). Oxford: Oxford University Press (1996).
Vallentyne, P. (1997). Intrinsic properties defined. Philosophical Studies, 88, 209–219.
Weatherson, B. (2001). Intrinsic properties and combinatorial principles. Philosophy and Phenomenological Research, 63, 365–380.
Witner, G., Butchard, W., & Trogdon, K. (2005). Intrinsicality without naturalness. Philosophy and Phenomenological Research, 70, 326–350.
Yablo, S. (1999). Intrinsicness. Philosophical Topics, 26, 479–505.
Acknowledgments
The author especially grateful for helpful comments from L. A. Paul and Doug Ehring and an anonymous referee, as well as audiences at the Central APA and the University of North Carolina.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Fisher, J.C. Dispositions, conditionals and auspicious circumstances. Philos Stud 164, 443–464 (2013). https://doi.org/10.1007/s11098-012-9862-2
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11098-012-9862-2