Tuesday, April 20, 2010

Draft: Do Extrinsic Dispostions Need Extrinsic Causal Bases?

My paper 'Do Extrinsic Dispositions Need Extrinsic Causal Bases' has been accepted by Philosophy and Phenomenological Research (which, I have to say, is every bit as well-run as people say it is)!

I'm going to have to submit the final version soon (as they will be running an online early program starting this summer!). So this is my last chance to pick your bloggin' brains about this. Any last-minute comment no matter how big or small or whether here or by e-mail would be greatly appreciated.

Here is the abstract:

In this paper, I distinguish two often-conflated theses—the thesis that all dispositions are intrinsic properties and the thesis that the causal bases of all dispositions are intrinsic properties—and argue that the falsity of the former does not entail the falsity of the latter. In particular, I argue that extrinsic dispositions are a counterexample to first thesis but not necessarily to the second thesis, because an extrinsic disposition does not need to include any extrinsic property in its causal basis. I conclude by drawing some general lessons about the nature of dispositions and their relation to their causal bases.


  1. Two quick thoughts, mostly to get clearer on the issue.
    Aren't your claims in secs. VI and VII made a bit too easy by the fact that you consider as an extrinsic property 'being such that the lock on Door is a K-lock' instead of, say, 'having a shape that corresponds to the arrangement of pins typical of locks of the type that happens to be on Door'? The second property, it seems to me, is less easily shown to be abundant, parasitic etc., and yet perfectly qualifies as a part of the relevant causal basis.
    I guess you'll reply that the latter quality is clearly analysable in two parts and reducible accordingly, so that shape is part of the causal basis and the shape-pin correspondence is not. But isn't this latter division arbitrary? Isn't the shape-pin *relation* what essentially grounds the disposition (as an extrinsic disposition)? I didn't follow your claims there.
    Independently of this, couldn't you avoid the claim that the possession of the relevant causal basis is not sufficient for the manifestation of the disposition by including the relevant extrinsic factors (e.g., the arrangement of pins in the lock of Door) in the stimulus? If so, wouldn't this make your claim that one should distinguish IDT and IBT more plausible?

  2. Ciao Matteo,

    Thanks for your comments. My first reaction to your first comment is that, insofar as it picks anything out, your description still picks out an intrinsic property of Key (its shape) although it characterizes it extrinsically. Say that you and I have the same mass, does that mean that i have two properties that of 'having a mass of 70kg' and 'having the same mass as Matteo'? As far as I can see, no.

    To your second comment, if one were to follow your suggestion, wouldn't all disposition and bases be intrinsic?

  3. Hi Gabriele,

    Well, the point of my first comment was exactly that extrinsic dispositions such as that to open Door are analogous to having the same mass as y, not to having mass x. It seems to me that those dispositions depend in a fundamental way on whether, on top of a certain intrinsic property, their bearer also is in a certain relation to things outside of it, i.e., possesses an extrinsic property. Besides, if I understand you correctly, in that section you don't question whether and how all extrinsic properties are reducible, but just the plausibility of a specific extrinsic property which - I claim - is too easy a target.

    Not sure I follow your response to my second point. My first reply would be "Not necessarily, it depends on what exactly one feels should go in the causal basis and in the stimulus in each case". More importantly, there I was following you in your claim that extrinsic dispositions may have entirely intrinsic causal bases, merely adding the suggestion that the other relevant factors may be included in the stimulus instead of left hanging. How does this make all dispositions intrinsic? Accepting this would be like saying that your main point in the paper depends on how one defines stimulus conditions!

  4. I guess I didn't explain myself clearly. You were saying that the property 'having a shape that corresponds to the arrangement of pins typical of locks of the type that happens to be on Door' is less easily shown to be abundant, parasitic etc. I was claiming that it is no less abundant or parasitic. First, I think that anyone who favors a sparse conception of properties would agree that Key doesn't have that property on top of having the property of having such-and-such shape. Second, even if one admits that there is such a property as the one you envisage, I think it would be hard to claim that it is a fundamental property. It would seem that Key has it only insofar as it has a certain shape and the lock on Door is a K-lock, so I don't see how the arguments I use against 'being such that the lock on Door is a K-lock' do not apply to your property as well.

    As for your second point, a disposition is extrinsic iff something can gain it or lose it without undergoing any intrinsic change. But if you include the external conditions in the stimulus then the disposition is no longer acquired and lost when those external conditions do not obtain and so the resulting disposition is not extrinsic. So yes, any extrinsic dispositions can be turned into an intrinsic one by appropriately redefining the stimulus, but this is no reason to worry, as it doesn't mean that there are no true extrinsic disposition ascriptions. Isn't it?

  5. 1) I totally agree, but what is relevant is the relation between the two 'parts' (the shape and the pins, in this case), and the disposition appears to be extrinsic only if this relation holds. If this is so, one may think, the causal basis of the disposition must include at least (the property corresponding to) one relational state of affairs that should not be analysed. I have the impression that being a radical sparse theorist with respect to the latter would clash with realism about extrinsic dispositions: typical arguments against these are exactly that they are parasitic on the 'true' dispositions in terms of which they can be analysed.
    2) Isn't it the case that introducing the relevant extrinsic factors in the stimulus would make the disposition intrinsic only if the latter is defined entirely on the basis of the stimulus itself? My thought was instead that if key has intrinsic properties (B) grounding its disposition 'to open door' (D) when it is inserted in door and door has lock of type K (S), you get what you need while avoiding the (vague) claim that dispositions are ethereal: B is intrinsic, D is extrinsic, and the stimulus S suffices for the manifestation of the disposition. Of course I am not saying that the disposition should be 'to open doors with locks of type K'!
    Intuitively, I would reject extrinsic dispositions, here I am trying to follow your 'flexibility'. Did this lead me totally astray?

  6. 1) If that was the case, wouldn't the disposition to unlock K-locks also be extrinsic? (Btw, as I say in the paper, I am not an extreme realist about dispositions (I don't believe that there are dispositional properties (i.e. properties picked out by definite descriptions of the form 'the dispositon to M when S')), I am only a moderate realist about dispositions (I believe that some disposition ascriptions are true)).

    2) I'm not sure if I'm following you, but to me it still looks like the disposition becomes intrinsic once you include the extrinsic factors in the stimulus. Isn't it?