Source
Barbara Vetter (2014), “Dispositions without Conditionals”, LIT-109, Mind 123(489): 129–156, read in full in NOTE-125 from the penultimate draft (28 pp., all 27 footnotes). Page numbers are the draft’s.
What was actually shown
The two conceptions (§2.1, pp. 6–7). The standard conception individuates a disposition by stimulus and manifestation, “a disposition to M when C”, and gives it a counterfactual modality. The possibility conception individuates it by manifestation alone, “a disposition to M, fullstop”, and gives it the modality of possibility, “x can M”. The relevant worlds hold the bearer’s intrinsic make-up fixed and vary its surroundings, the device Lewis (1976) used for ‘can’.
Against closeness (§2.2, pp. 9–11). Suppose “easily possible” means true in some relevant world close enough to an ideal world. The ideal cannot be actuality, or anything that actually breaks would be maximally fragile. For any other ideal, stipulate a block with an Achilles’ heel that breaks only in that world. It then counts as fragile, and as more fragile than a champagne glass. “One world is not enough.” The recipe is general across ideals, and rests on the intuition that the block is not fragile.
Proportion instead (§§2.2–2.3, pp. 11–14). x is more fragile than y iff x breaks in more of the relevant cases. A case is a world, a time and an object, because the irascible get angry many times in one world. To avoid counting non-denumerable sets, count over a finite subset of cases with the greatest variation in causally relevant external circumstances. The needed proportion is small, near “at least one”.
Plain possibility (§2.4, pp. 14–16). For ‘breakable’, ‘soluble’ and ‘transmissible’, one actual manifestation suffices, so the modality is possibility simpliciter; even the Achilles’-heel block is breakable. Across terms the variation runs from “at least one” to “a few”, never toward necessity.
Discrediting the contrary data (§3.1, pp. 16–20). In COCA, 341 hits for “disposed to” left 226 relevant uses. 221 had agents as subjects and 5 abstracta; only 5 interacted with ‘if’ and 1 with ‘when’. None ascribed a permanent intrinsic tendency to a concrete inanimate subject. So “disposed to” is a theoretical term, and intuitions about it encode the conditional theory.
Stimuli recast (§3.2). Where a stimulus seems to matter, it becomes part of a causally complex manifestation (claustrophobia: being caused by enclosed rooms to feel anxiety), or a fact about the typical causes of that kind of process.
What could have come out otherwise. Corpus uses of “disposed to” could have ascribed standing tendencies to objects, and closeness grading could have survived some choice of ideal. Neither happened.
What this does not say
- That the conditional analysis is wrong. The paper will “not … address the question which conception is to be preferred” (p. 6) and leaves “the argument for another day” (p. 26). The comparative case is deferred to “Multi-Track Dispositions”.
- That degrees of disposition are defined. Counting cases in a non-unique finite subset is figurative. The cardinality problem is sidestepped, and vagueness is offered as the reason no canonical subset is needed.
- That finks, masks and antidotes are solved. The proposal “is not going to solve” them (p. 5); they arise on both conceptions (p. 7).
- That it covers the dispositions of physics. Natural properties that ground laws, such as charge, need nomological necessity. The clash with dispositional essentialism is called “merely nominal”, which restricts the view’s scope rather than answering the essentialist (pp. 24–25).
- That model capabilities are dispositions of this kind. NOTE-125 notes the analogy between “can do X” (one elicitation) and “reliably does X” (robust across varied prompts). The paper does not draw it, and the record has not tested it.
- That it holds beyond English. The linguistic evidence is English-centred, with a cross-linguistic assertion (fn. 2), and the corpus count is small and was coded by hand with no inter-rater check reported.
Connections
- Pernu (LIT-144) relies on a counterfactual semantics that this paper argues is the wrong modality for ordinary dispositions, though the paper says nothing about counterfactual theories of causation.
- Behavioural capacity and intelligence (THEORY-092) is the record’s other account that turns on what a capacity ascription says.