Sunday, June 18, 2006

Synthetic Survival

Kevin T. Keith asks why sci-fi characters would want to undergo a "cognitive upload", creating an immortal synthetic copy of themselves:
[Y]ou’re not buying anything extra for “you”, you’re just creating a new person who happens to share all your memories up to that instant (and presumably your personality traits as well), but then becomes its own entity and lives its own life. A good deal for that new person, of course, but why would “you” care?

He further claims:
[I]t does not grant immortality to the entity actually choosing to undergo the copying process... his consciousness, as an artefact of his brain, stays with his brain; the new body gets its own consciousness, resident in its synthetic brain, which picks up with the exact memories and thoughts the old one had had and then branches off on its own path, but the old one just remains what it was.

But what's the basis for this claim? I don't deny that the copying process would lead to two distinct seats of consciousness, with their respective organic and synthetic physical substrates, but why think that the consciousness of his past self "stays with his brain"? (It's question-begging to hold this on the basis that "his consciousness [is] an artefact of his brain". Each future consciousness is the product of its respective "brain". The question is which, if either, is to be identified with the past self. The above proposal presupposes that the person endures through his organic brain. We're not given any reasons to accept this view.) I agree that it's silly of the character to think he had “a fifty-fifty shot” of ending up in either body. But that's not because the self endures in its organic body. Rather, it's because the self doesn't endure at all.

According to Parfit's reductionism, which I'm sympathetic to, there is no enduring Cartesian Ego of the sort that Kevin's remarks presuppose. I'm having some conscious experiences now, and a future person-stage will have his own experiences and remember mine, but we are two distinct subjects of experience. We exhaust the relevant facts once we state the relations of physical and psychological continuity between the person-stages. There is no "deep further fact" about whether another person-stage is really "me" or not. (Parfit's sorites-type cases help support this conclusion.)

Crucially, on this view, my present consciousness does not "stay" anywhere apart from this very moment. It does not endure into the future. Each moment gives rise to "a new person[-stage] who happens to share all your memories up to that instant". Our everyday persistence consists in no more than the sort of "copying" that Kevin dismisses. Since I value my persistence all the same (even if it merely consists in perdurance, not endurance), I'm led to value "copying" likewise. What matters for our persistence, insofar as our persistence matters at all, are relations of psychological continuity. Future "copies" will be psychologically continuous with our present stage, and this provides us with all the reason we could coherently ask for in order to care about them as a future 'self'. (That's not necessarily to say that it's very much reason.)

Kevin goes on to discuss the tricky problem of double-survival, and ends up accepting a position that sounds much like Parfit's. At least, he suggests that the best way to describe a case of double-survival is to say that neither resulting person is identical to their shared past self (even though each would have been, had it not been for the other survivor). Though I'm not sure whether he goes all the way to accepting that this shows that identity is merely superficial, and isn't what matters for survival. (I'm also not sure whether he means this as a retraction of his earlier claim that the person's consciousness "stays with" their organic brain, and if not, how he would deal with Parfit's more symmetrical split-brain case.)

Anyway, I very much share Kevin's ultimate conclusion about the topic: "Good fun." Perhaps I should conclude by returning to his original question: given that survival doesn't "buy anything extra" for your momentary self, "why would “you[-now]” care?"

8 comments:

  1. Two further points:

    1) I responded to the "fifty-fifty shot" thing by suggesting that neither occurs. But there's another sense in which both do. A person in each body "wakes up" with memories of his past self and the transfer. One will say, "Great, it worked! I'm now immortal!" And the other will say, "D'oh, nothing happened, I'm still organic." The point is simply that neither of them is the enduring consciousness of the previous moment, for there is no such endurance. But both of them are perduring future stages of the shared past self, persisting just as well as any of us ever does. (Twice as well, in fact!)

    2) Another reason in favour of making a "copy" of yourself is that it would allow your future selves to diversify, and complete twice as many of your projects. No more heart-rending decisions about which of two dreams to pursue. Now your future selves can do both -- one each!

    ReplyDelete
  2. I don't think Kevin's argument needs to appeal to a Cartesian Ego, as you suggest. All it needs is to point out that the upload is actually a copy of the organic, which, ex suppositione, it is. Even if the upload is conscious, it's a copied consciousness. It's considerably more elaborate and detailed than having someone skillfully impersonate you, but it's not really much different in the end.

    If our everyday persistence consists in copying, what is the copying mechanism?

    ReplyDelete
  3. I think continuality is still a pretty good reason to care... as a half full person I would say it is the best possible reason to care!

    Brandon,
    I think Richards point is that ALL consciousness is "copied consciousness" (in a sense). So pointing out that the upload is a copy doesn’t help in the slightest.
    Also impersonation would probably not provide continuality - so it wouldn’t be "copied consciousness".

    ReplyDelete
  4. Yes, that's right. It isn't literally copying, or at least it seems strange to think of everyday persistence in those terms. But my point is that standard persistence doesn't involve anything more than persistence through replication does (or at least, not anything relevant -- I guess it has more physical continuity). In particular, it doesn't involve the sort of enduring consciousness we intuitively believe in as commonsense Cartesians. There are just the relations of psychological continuity, memories, etc. All those things that the "copy" has just as well as one's "natural" future self. (Impersonation is rather superficial in comparison. But if made more completely encompassing -- say through memory transfers and the like -- then sure, I'd agree that this too would be "not really much different in the end.")

    I may have misunderstood Kevin's argument, but the suggestion that the agent's consciousness "stays with his brain" sure sounds like a commitment to an enduring Cartesian Ego, albeit one with a neural basis.

    ReplyDelete
  5. 2) Another reason in favour of making a "copy" of yourself is that it would allow your future selves to diversify, and complete twice as many of your projects. No more heart-rending decisions about which of two dreams to pursue. Now your future selves can do both -- one each!

    While "I" might not actually endure, I still feel like I do, and so the thought of another "me" pursuing projects that I would like to pursue but can't because of, say physical or financial restrictions, cheers me not at all, for "I" wouldn't experience the projects or any occurent benefits attached to those projects.

    ReplyDelete
  6. I may have misunderstood Kevin's argument, but the suggestion that the agent's consciousness "stays with his brain" sure sounds like a commitment to an enduring Cartesian Ego, albeit one with a neural basis.

    I think I was reading more in the sense that the original remains the original even when copied; the copy doesn't magically become the original in another form just because it was copied from the original.

    I do think it's clear that personal persistence even under a non-endurance view has to involve more than persistence through replication does, because nothing persists as an individual through replication -- only the species persists, because only the specific characteristics are carried over. For personal persistence to involve nothing more than persistence through replication does, we would have to argue that persons are actually species, not individuals; which is a possible position, but requires, I think, that we be able to identify clearly defined individuals in the species -- which I don't think perdurantists can do. Time-slices just aren't well-defined individuals, because you can make them as thick or as thin as you please without affecting the perdurantist position; and if this is right, I don't think there can be much analogy with copying. In a coherent perdurantism, time-slices aren't like copies; they are capable of extending into each other and overlapping, and that makes a serious difference. (In other words: continuity is actually not much of an issue in copying, as such, except insofar as the copying mechanisms are involved; but it is the whole issue with persistence.)

    ReplyDelete
  7. Children are little organic "copies" that people persist in creating. An uploaded consciousness copy is a type of child, yet another poor substitute for immortality.

    The uploaded consciousness-copy of some persons might be worth more to society than millions of human beings in the flesh. Some human minds are exceptionally fertile in a beneficial way and worth preserving.

    ReplyDelete
  8. Psychological continuity makes for a pretty significant disanalogy there. (Kids don't inherit your memories, for example.) So you haven't given any reason to think that uploads/copies are a "poor substitute for immortality". (See above for my arguments that this is as much immortality as anyone could coherently ask for.)

    ReplyDelete

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.)

Note: only a member of this blog may post a comment.