Unmanifestable Dispositions

This morning I’ve been thinking about dispositions that cannot be manifested: that is, dispositions to phi under circumstances C, where either phi-ing or circumstances C are metaphysically impossible.

One thing I’m interested in is whether there are any such dispositions. Another is whether anything has such a disposition. Prima facie, there are some reasons to answer yes to both questions. I think, for instance, that I have a disposition to be puzzled when presented with a round square object.

In response to this suggestion, however, Daniel pointed out that a certain amount of coarse-graining about dispositions would enable us to accommodate that disposition without believing in dispositions which cannot be manifested. My disposition to be puzzled when presented with a round square object may be identical to my disposition to be puzzled when presented with an interesting and surprising object that I didn’t think existed, and this disposition can of course be manifested.

Lewis’s counterfactual account of dispositions in ‘Finkish Dispositions’, combined with his view that counterpossible conditionals are trivially true, delivers that everything has every disposition to phi in circumstances C for impossible C. But this does not by itself entail that there are any dispositions which cannot be manifested, since these trivial dispositions may for all we’ve said so far be identical to more familiar, manifestable, ones.

Nevertheless, for those of us inclined to be abundant with our dispositions, I think there is some reason to believe in unmanifestable dispositions (and instantiations thereof). And I don’t see any special reason why there shouldn’t be such things, given that dispositions don’t need to be manifested in order to be instantiated.

One Reply to “Unmanifestable Dispositions”

  1. You write about the metaphysically impossible. I think that a lot hinges on how impossible you want C to be. (I am leaving aside impossible phi-ing.) I shall start with a straightforward possible-worlds analysis where C is actual in some world or other, then try to stretch this analysis to apply when C is impossible in some fairly strong sense.

    The analysis

    Suppose that your disposition is to be surprised when confronted with a solid object that has five spatial dimensions. (I mean an object you can handle, say 20 cm across, not a string-theorist’s object.)

    If this were to appear to happen in the actual world, w*, you would of course not believe it. You would try to find out why the object appeared to have five spatial dimensions, even though it actually only had three.

    It could however happen in another possible world, w5, in which there were five accessible spatial dimensions.

    So how about saying that you, as you are in w*, have the disposition to be surprised when confronted with a solid object that has five spatial dimensions, because if you were suddenly moved to w5 you would indeed be surprised by the objects you found, at least until you got used to your new environment?

    This approach would stop everything from having every disposition to phi in circumstances C for impossible C, because some things would not phi if suddenly moved to the other worlds under consideration. A butterfly would not be surprised by the objects in w5, because it does not have the capacity to be surprised by anything, so we would not attribute to it, as it is in w*, the disposition to be surprised by five-dimensional solid objects.

    One question is whether we can talk about you in w5. Do we need to believe in trans-world identity to use this analysis, or could we make the analysis work even if we could only have your closest counterpart in w5? The disposition to be surprised is a special case, because the analysis depends on suddenly moving you from w* to w5 and seeing how you react. That picture of moving you from one world to another rather assumes trans-world identity, as well as encouraging the false picture that other worlds are joined to our world by corridors. (I don’t count physicists’ inter-universe wormholes as part of philosophers’ possible worlds theory. I am also cheerfully ignoring the practical consequences of suddenly placing organisms in environments with more spatial dimensions than they are used to. Membranes which can hold fluids in place in 3-space will not do so in higher-dimensional space.) Other dispositions, the analysis of which would not involve thinking about your suddenly being moved from one world to another, might make it easier to ask the question of whether we needed trans-world identity or whether counterparts would be good enough.

    Making C impossible in a strong sense

    There is no world wrs with round square objects in it. If a world might at first glance appear to be wrs, then it would not really be wrs because its geometry would be so different that the terms “round” and “square” would not mean what they do in w*. That would seem to block the above analysis. If you were suddenly moved to wrs, you would be puzzled by some of the things you found there, but you cannot be moved to wrs because it is not there to receive you.

    How about helping ourselves to incomplete sketches of worlds? The sketches could not be completed coherently, just as you cannot coherently combine the whole of an M C Escher drawing with the laws of physics or of geometry, but we could get enough into the sketch of wrs at once to see how you would react when confronted with round square objects. Then we could use the above analysis, ensuring that you would, and a butterfly would not, be puzzled when confronted with a round square object.

    This approach would raise the question of which bits of the impossible world we should sketch in, and which we should leave out. If we sketched in different bits, we might establish different dispositions. I fear that we might have to rely on case-by-case judgement here, just as we have to rely on case-by-case judgement when we need to choose the factors to consider when deciding which non-actual worlds are closest to w*, or when deciding which entities in some non-actual world are the counterparts to entities in w*. As with such judgements, it will often be obvious which bits of the impossible world to sketch in, but some philosopher will think of an example where it is not obvious.

    Finally, there are the extreme impossibilities, such as the worlds in which 2 + 2 = 5, not just because four apples form a critical mass which always magically creates a fifth apple but as a basic consequence of some crazily re-jigged arithmetic. Then we might not even begin to have a sense of how to start sketching such a world, or we might be able to sketch so little of it before our sketch became incoherent that we could not have any inkling of how we would get on in such a world. Perhaps that is where we should draw the line, and say that C is so strongly impossible that you do not have a disposition to phi in circumstances C.

Leave a Reply