Saturday, December 13, 2008

Cryonics and Continuity

Suppose we could be frozen at death, to be revived centuries later (perhaps as digital 'uploads') by a society of transhuman immortals. Sound good? Many of the folks at Overcoming Bias have little doubt. I lean in their direction, but rather more tentatively. Maybe some people would benefit from this transition, but I think there's a hidden incoherence in calls for more widespread adoption of cryonics.

The first thing to note is that, from an impersonal perspective, we're not doing the world any favours by populating the future with our primitive 20th(-21st) century minds. (What would they want us for? Anthropological interest? But this then raises the distressing prospect that we will 'awaken' only to live out our second lives in a scientific testing facility, or in some sort of transhuman zoo.) Cryonicists hope to be revived by benevolent transhumans, but assuming future altruists could get more bang for their buck by bringing into existence completely new (and better) minds, benevolence would presumably lead them to do that instead.

So cryonicists must assume that it is better to extend an existing life than to create a new one. Maybe we have special obligations to existing people, such that benefiting them personally takes priority over making the world a better place. So let's consider the question from a personal perspective: would cryonic revival make one better off?

To answer this we need to consider what makes for a good life (i.e. good for the person living it). It's not as though longevity is intrinsically desirable. Living longer is only good insofar as this enables us to live better. Put another way: death is only harmful insofar as our life, by ending then, is worse on the whole than it otherwise would have been.

A life has a kind of narrative structure, and a good one (like any good story) will fit together well. We have various life projects -- from developing a systematic philosophy to helping children develop into autonomous adults -- which give our lives much of their 'meaning' and worth. These projects are extended across time: decades, often, and sometimes an entire lifetime. But they are also typically situated in time, and may not survive one's sudden displacement into a radically different future society. So any would-be cryonicist must ask: "What is it that I really want from life? Could I still pursue it in the transhuman utopia?"

Some might. But many people are (quite reasonably!) wedded to the particularities of their life and situation, and would not welcome such drastic change. They would awaken to world where their central concerns and life projects no longer belong. Such radical discontinuity should concern any reductionist about personal identity. For insofar as this newly awakened person would be enculturated into a new society, acquiring new values and life projects, they are effectively becoming a new and different person. But notice that this completely undermines the only justification for reviving them. If the benefits of revival accrue to a different person, then revival is unjustified: a better new life could be created 'from scratch', so to speak.

So cryonics is (at best) only justified for people whose central concerns and life projects could continue to be fruitfully pursued upon revival in a transhuman society. That won't be everyone. For many, the cryovangelist's promise of a 'new life' may be taken literally: given its radical discontinuity with all that they care about, such a new life would do nothing to improve their (current) life.


  1. Aren't you underestimating the importance of memory in establishing the continuity of personal identity?

    If a revived cryonicist retained memories of their former life, wouldn't they be "the same person" in an important sense?

  2. Well, it's not as though there's really any further metaphysical fact to be "established". So what kind of relation we consider "important" is really just a question of ethics (broadly construed). I can't say I care a whole bunch whether some future person-stage has my memories. What I care about is the continuation of my values, projects, etc.

    (But your mileage may vary. I expect a committed hedonist, for example, would find it easier than most to project their identity into the revived life too.)

  3. Still, what's the harm in freezing yourself?

    If when you wake up, you don't like it, you can just kill yourself off, no?

  4. I don't know if rejecting personal identity makes a difference (if that's what you meant?).
    All else being equal, projecting at least some of your own mental contents into the future via uploading, cryonics or just bog-standard survival is still predictably better for the continuation of your projects, values etc then not.
    The only reason I'd see for not doing it (given your assumptions) is if you'd give better than even odds on your future self 'turning evil' or something. If they just turn out to be ineffectual in the future environment then no harm done.

  5. We are uncertain about what makes people's lives go best. In light of this uncertainty, signing up for cryonics seems to be the rational choice. Even if there is only a 10% probability that life in the posthuman future will be good for us, the value of a one-in-ten chance of living a good life for thousands of years clearly outweighs the disvalue of a 90% chance of a brief future existence that ends in suicide. In effect, we would be signing up for cryonics for the privilege of knowing whether we had good reasons for having signed up, and for the opportunity to act on the basis of that knowledge.

  6. I'm also skeptical about whether reductionism will make a difference here. Plausibly, reductionists should believe that many relations matter, including memory, dispositions, intention, plans, etc. I see no reason to think that many of these relations will not obtain between me and a future unfrozen self will fail to obtain.

    Imagine another example: I move to a radically different foreign culture. I need to acquire many new habits and dispositions and projects. Does a reductionist have to say that I am effectively a brand new person? I doubt it.

  7. How large of a social change is required before ordinary living people should kill themselves?

    Could the practice of suttee be defended on this basis - that a woman is nothing without the social context of her husband, especially if society sees her largely in terms of her husband's identity?

  8. EY - Actually no, I tend to think that blatantly false premises aren't a good basis on which to justify burning people alive. But your standards may differ.

    (Snark aside, it's hard to see how your question relates to my post. Note that on my account, the opinion of "society" is irrelevant. Each individual authors their own story. But even if someone can see no further point to living, it hardly follows that burning them alive is a good idea. I guess it's conceivable that someone might really desire this particular form of death for themselves, but that's true on any account. And of course none of it even comes close to justifying the coerced version - i.e. murder - found in the real world. So I don't appreciate your attempt to rhetorically associate my theory with that despicable practice.)

    "How large of a social change is required before ordinary living people should kill themselves?"

    You assume that refraining from life-extension is interchangeable with suicide. That strikes me as clearly mistaken: one might reasonably prefer to die of natural causes, after all. The acts have different "meanings"; a story that ends in suicide is different (and arguably less desirable) than other possible endings.

    The real question here is: how great must the external discontinuity be before an individual can no longer recognize themselves in the future that lies ahead? That's obviously going to differ from individual to individual, so I'm not sure what kind of answer you're looking for. (Given your later question, I rather doubt whether you were even asking this one in good faith.) But I suggested in the main post that it will depend in particular on whether one's "central concerns and life projects could continue to be fruitfully pursued". That is, one must ask oneself, "What is it that I really want from life? Could I still pursue it in the [radically different society]?"

    This is an open question. It clearly depends on the precise details of the individual and the possible future they're considering.

  9. Pablo - I don't think it's so easy to move from normative uncertainty (effectively: ignorance about what's rational) to conclusions about what's rational. But if we could, you'd also need to take into account the likelihood that what we really ought to do is bring about the impersonally best future, i.e. filled with better minds than ours. So [contra AnlamK and GLC] I think there is at least some "harm" to consider here -- though you might consider it worth the gamble even so.

    (Of course, any practical calculation would also need to include the financial costs and low chance of success that I bracketed for purposes of this post.)

    Javier - I assume that the important kind of continuity is a matter of 'narrative unity': can I tell a coherent life story that connects the one stage of life to the next? Or are they so different as to comprise two completely separate life stories?

    I agree that in ordinary cases of moving to a foreign culture, the requisite unity and continuity is there. (This is most obviously the case for someone who has long planned to make such a trip, and once there can execute various past plans, etc.) It may well hold for keen cryonicists too, who have incorporated the possibility of future revival into their self-conception. That's fine. My point is simply that they shouldn't be so quick to assume that everyone else must share this conception. For many, I suspect, the prospect of dying and reawakening centuries later in a post-human society will seem wholly alien -- and alienating.

    Again, those who are thrilled by the prospect may go right ahead. But others may be wholly satisfied by a life of eighty years (growing from childhood to raising kids of their own, advancing in their careers, retiring, and spending time with their grandkids). If they have achieved all that they wanted to in life, there's no great harm in death. Death is bad when it cuts short our life story, depriving us of the narrative resolutions we'd prefer. If post-cryonic revival would add nothing to the narrative arc I'm (now) living out, of what concern is it to me?

    [I might have impersonal reasons to care about other stories than mine, of course, but I've claimed that such considerations favour alternatives to cryonic revival. Better to see the future populated with brilliant new minds well fit to their cultural environment. No reason to want me there unless I want it myself. And no reason for me to want that unless I could continue to advance my life projects.]

    Of course, you might be skeptical of the antecedent of my conditional claim. Perhaps the discontinuity would not be so radical as to create a wholly new story. Perhaps I could indeed continue my present life-story, and do so in a desirable way. That's the big question (on my view).

    P.S. Note that reductionism makes a difference in at least allowing the question to be raised. If I were a magically enduring Cartesian Ego, the question wouldn't even arise. The future person would really be me, in some deep and important sense. So one might care about that even if there was nothing else in the future life that speaks to one's present interests and concerns. But if reductionism is true, we need some independent reason to care [personally] about some future stage, to project our identity into it, or to conceive of it as our "own".

  10. I expect a committed hedonist, for example, would find it easier than most to project their identity into the revived life too

    I'm a committed (universal) hedonist, and I think this is right. I'd keep trying to accelerate the progress of humanity into the Experience Machine, just like I do these days.

  11. Richard,

    I expect, based on your references to narrative unity of self and notions related to character, that the OB folks are on much different footing philosophically than you are.

    I don't know if you're a utilitarian, but the OB folks (and, I think, most transhumanists) are. I find that makes them very hard to reason with wrt ethics, for instance leading to some simplifications of what would constitute a good life for the revived.

  12. Neil - just out of curiosity, do you think it's better to give some hedonic bonus to an already existing person, or to instead bring into existence an additional person who would add even more to the net happiness?

    Robin - you don't really say very much about my view, other than that you find it "a bit odd". But I'll add a quick comment over there.

  13. I think my view commits me to ambivalence between those two options. Which I admit is counterintuitive -- the intuitive position here, I think, is that it's better to give existing people hedonic bonuses than to create new ones who get the bonuses.

    I don't think I agree with you on the general question of whether the unfrozen people are additional persons as opposed to the same old persons. This consideration doesn't do a lot for me:

    They would awaken to world where their central concerns and life projects no longer belong.

    If for some reason they got unfrozen not caring at all about their old projects and having weird new futuristic projects, okay, that might make them new people.

    But I take it that's not the scenario we're talking about. If you unfreeze William Wallace in a world where the Sun went Red Giant, burning up Scotland and all of Earth so that everybody lives on starships, I'm assuming that he goes through a complicated and emotionally difficult process of readjusting himself to his new world.

    I understand that process as sort of marking his psychological continuity. His emotional difficulties upon waking up in the far future mark him as psychologically connected to his old self. He has those difficulties because he is William Wallace who loves Scotland more than life itself. So, identity holds.

  14. Perhaps in some sense. (There are no "deep" facts about identity. And we can construct any number of identity-like relations, each perhaps serving a different philosophical purpose. Your property of "being William Wallace" is serving a role in psychological explanation. That seems a different kettle of fish from what I'm talking about.)

    My claim is more limited: this is not the kind of 'identity' or continuity that matters for ethics. In particular, it is not the sort of continuity that can justify less total welfare. We may put it this way: the "locus" of welfare value is a life-story. If Wallace effectively lives out two of these, then for moral purposes they count as two and not just one. There is no reason, even for an average utilitarian, to prefer the "same old" Wallace to wake up in a starship rather than some entirely new person. Because for ethical purposes, the "old" Wallace -- put into this radically new context -- counts as a new life.

    That's the claim. Why believe it? Because the kind of 'narrative unity' I'm talking about is the best candidate for a kind of continuity that matters. We can all see the tragedy in a life-story cut short (life projects thwarted, past efforts for naught). But what possible reason is there for preferring the newest member of the starship to have Wallace's memories and love of Scotland? Normally, continued existence changes the moral qualities of our past -- mistakes are redeemed, efforts are rewarded, etc. -- but none of that seems possible given the kind of radical external discontinuities we're considering here. So whatever continuity may hold, it does not seem to be of a sort that matters.

  15. I guess what I'd want to say here is that my intuitions push towards regarding the kind of personal identity important for (some) ethical purposes as the same as the kind of personal identity important for psychological explanation.

    Suppose I made a promise to Wallace when he was alive, and I never got to keep it because he died. We've got Wallace's frozen body and we're trying to revive him. If I'm interested in making good on the promise, it will matter to me whether Wallace revivified will have psychological continuity with the Wallace I knew, as manifested in his pro-Scottish projects among other things. This is the case even if time has made it impossible for him to advance those projects. Bring back a person who doesn't care for any of the people he knew or have any love of Scotland, and I may not see him as the legitimate promisee. But if he does care about these things, I'll be able to see him as the person I made the promise to.

    The moral notions you're interested in seem to be rather different ones -- you're working with a normative theory where welfare is a matter of living out good narratives. I guess I'd have to get a better understanding of the structure of the view (for example, how you intend to aggregate the values of multiple lives with good narrative unity) before knowing how to handle questions like this. Otherwise I'm not sure what hangs on regarding Wallace as living out two narratives rather than one, or if it's even useful to talk about identity here.


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.)