Friday, May 04, 2007

Fission and Identity

Johnny-Dee quotes Swinburne's take on fission:
Reflection on this thought experiment shows that, however much we know about what has happened to my brain—we may know exactly what has happened to every atom in it—and to every other material part of me, we do not necessarily know what has happened to me. From that it follows that there must be more to me than the matter of which my body and brain are made, a further essential immaterial part whose continuing in existence makes the brain (and so the body) to which it is connected my brain (and body), and to this something I give the traditional name of ’soul’. I am my soul plus whatever brain (and body) it is connected to. Normally my soul goes where my brain goes, but in unusual circumstances (such as when my brain is split) it is uncertain where it goes.

I prefer Parfit's solution, which is to deny that there is any further fact here to know. The physical and psychological facts exhaust the facts. Once those are all specified, there is nothing left to know about the world. It would seem strange to posit two possibilities, alike in every objective and subjective respect, yet somehow differing in virtue of the "identity facts". For what would those consist in, and how could we ever grasp them? Our commonsense concept of identity tracks a familiar kind of continuity, but we have no reason to think there's anything further underlying it.


  1. It seems to be that when your mind is split (into two), it thinks on two tracks (think if certain mental illnesses where the two sides of the brain cannot communicate properly). Which also happens to be the pretty intuitive way in which it would work - (how else could it possibly work?).

    Demonstrating that his interpretation of his thought experiment is probably mistaken.


  2. By Swinburne's reasoning there's "something more" to amoebas than protoplasm and something more to the Ship of Theseus than planks. Maybe this is ok if we take "something more" in a minimalist sense. But the further claim that the something more is an immaterial part is doesn't fly--unless Swinburne is arguing for panpsychism (which he isn't).

  3. however much we know about what has happened to my brain—we may know exactly what has happened to every atom in it—and to every other material part of me, we do not necessarily know what has happened to me

    Doesn't this initial sentence assume Cartesian dualism in the first place? (i.e. that there is a difference between 'my brain' and 'me'). I'm no expert, but his conclusion just seems question-begging as a result.

  4. "however much we know about what has happened to my brain...we do not necessarily know what has happened to me. From that it follows that..."

    It follows that we don't have as clear-cut and well-defined a concept of a person as we do of a material thing. But didn't we know that all along?

    If one of the two results of the fission turns out to be a vegetable, then that might suggest there was an immaterial substantial soul - but to presume that outcome would beg the question.

    If, on the other hand, both the 'offspring' act like sentient people, then: (a) people don't have souls; or (b) you don't need a soul to be a person; or (c) souls can divide to create two souls.

  5. Not so fast Richard...

    There have been several suggestions about world that are alike in the physical and psychological facts, and yet differ in "identity facts." For instance, see Cameron Ross on the contingency of composition. He thinks that whether x and y fuse to form z is a matter of the laws of mereology (or some other contingent fact). So two worlds, both with only x and y touching, and in one world there are two things (just x and y) and in another world three things (x and y and x&y=z).

    I guess I am saying that the same physical facts need not imply the same objective facts.

    Could be the same in the other case: same psychological facts need not imply the same subjective facts.

  6. Curious. What sorts of reasons are typically offered to motivate those suggestions? I mean, I could sort of understand traditional mereological debates if they were analytic, concerned with explicating the most coherent and theoretically satisfying account of our ontological concepts. But I have a lot of difficulty understanding mereological composition as a contingent worldly fact! For example, I take it that the mereological laws and facts would, in that case, be strictly unknowable. But then why posit them at all? What do these extra facts explain?

  7. I am not sure that I buy into this contingent composition thing, but here is a motivation for it.

    (1) The universe has objective structure. That is, for any two simples x and y, there is a fact of the matter about whether their fusion, x&y exists.

    (2) There is no contradiction in holding that for all x and y, x&y exists (unrestricted mereology).

    (3) There is no contradiction in holding that for all x and y, x&y does not exist (mereological nihilism).

    (4) There might be no contradiction in holding that for some x and y, x&y exists, but not for all (commonsense mereology).

    Thus, given any arrangements of simples, there are (at least) two, maybe (at least) three ways the world could be. It could have no fusions, it could have as many fusions as possible and it could have a restricted range of fusions. That is, in bloody possible world speak, there is more than one possible world that corresponds to any history of spatiotemporal relations among the simples.

    Like I said, I am not sure that I buy this. In my paper-in-my-head about mereology, this is something that I tackle.

  8. Swinburne thinks that reductionists like Parfit are committed to a kind of verificationism that ignores the logical possibility that one or the other of two identical survivors could be (uniquely) identical to the original. I think Swinburne is a non-reductionist for the wrong reasons.

    I am a non-reductionist about our identity over time because I think identity over time itself is primitive. I am an ontological reductionist, however, in that I agree with Parfit that the physical and psychological facts exhaust what we are at a time and over time.

    The mistake is to move from this ontological reductionist claim to the claim that identity over time is itself analyzable in terms of the facts about physical and psychological continuity.

    As I mentioned in another thread, on this view the 'further fact' of identity over time is not mysterious in the way that a Cartesian soul or an (immaterial) Lockean consciousness would be, it is simply due to the fact that identity over time is primitive, which I think is consistent with the recognition that what we are is exhausted by the physical and psychological facts about us (both synchronically and diachronically).

  9. That still sounds mysterious to me. Once we fully specify all the qualitative (e.g. physical and psychological) facts, there doesn't appear to be any room left for further facts. Where are they to fit into our understanding of things? What in the world do they do? You say that "identity over time is primitive", but why posit further primitives when we don't need to? (Ockham's Razor.)

  10. The view that identity over time is primitive is motivated in part by the failure of all the proposed criteria of identity, including those for personal identity, but also including the standard criterion of 'spatio-temporal continuity' for ordinary physical objects.

    The standard view has it that criteria of identity over time are generated by the conditions of kind-membership, for any given kind. There are two general problems with this view: (1) to belong to a kind and to persist are two different things; and (2) the evidence for identity always underdetermines identity, such that for any given proposed criterion of identity over time for a given kind, two or more of that kind can always satisfy the criterion.

    What all of this mounting evidence points to, I propose, is that we are more reasonably justified in concluding that identity over time is primitive than otherwise.

    I also believe that reductionism about a kind (say, human animal) is compatible with the claim that identity over time is primitive. As Parfit might say, a human animal is nothing over and above a brain and a body and the interrelated set of physical and psychological states involved in the existence of that human animal. We don't need to go one step further and posit that identity over time for a human animal is analyzable into physical and/or psychological continuities in order to be ontological reductionists about what is involved in the existence of a human animal.

    I don't see any violation of Ockham's Razor here. I'm not positing the existence of any more entities in claiming that identity is primitive.

    To answer the question about personal identity, "What makes me today identical to myself of five years ago?" The answer is, nothing. Nothing makes a thing identical to itself, not at a time, and not over time.

  11. The evidence you point to could just as well be taken to indicate that our 'identity'-talk is simply misguided, not latching on to anything fundamentally real. That's my view: a full accounting of the facts can be provided without any mention of identity facts. We can introduce 'identity'-talk in any number of (more or less arbitrary) ways, but there's nothing fundamental that it's getting at. You haven't given me any reason to take identity seriously.

    P.S. What's the truthmaker for identity claims? I think you must say it is a primitive property, a haecceity or 'thisness', which "makes me today identical to myself of five years ago". (So you must violate Ockham's Razor.) If you really say there is "nothing" in reality to ground the alleged truth in question, then it isn't a truth at all, but empty words.

    Anyway, we can put the dispute like this: as a reductionist, I think the qualitative (non-identity) facts exhaust the facts. If you disagree, if you think that there is something more, something in addition, then you are positing additional primitives (whether they be facts, properties, or some other entity), and so I hold you to be violating Ockham's Razor.

  12. Do you think there is also no such thing as synchronic identity?

  13. I have no idea what you're asking when you say that. I can tell you this: I don't believe in haecceities. I don't see any other substantive questions here.

  14. Just to clarify, so I understand your view...

    For the synchronic case, to change Adams' example slightly, suppose Aristotle has a momentary perceptual state that includes hearing a robin sing and tasting an anchovy.

    Here it is implied that some individual that is hearing a robin sing is identical with one that is tasting an anchovy. Adams claims that "this sort of identity...could be primitive in a purely qualitative construction of reality."

    And you say...?

    Also, is your view any different from Lewis' view?

  15. Oh sure, that's just to say that someone has a momentary perceptual state that includes both hearing a robin sing and tasting an anchovy. Talk of "identity" adds nothing of substance to the description.

    I'm not aware of any differences from Lewis' view (though I couldn't say for sure).

  16. Right, I understand now. I actually agree with Lewis that identity per se is wholly unproblematic.

    I take it you do think that there are substantive metaphysical issues having to do with persistence though? It doesn't matter to me whether we state those issues in terms of the relation of identity or in terms of the relation of unity.


Visitors: check my comments policy first.
Non-Blogger users: If the comment form isn't working for you, email me your comment and I can post it on your behalf. (If your comment is too long, first try breaking it into two parts.)